Management Information Systems for Microfinance : Catalyzing Social Innovation for Competitive Advantage [1 ed.] 9781443885102, 9781443853514

Microfinance is a double bottom line sector which is growing fast, making money and doing well in a variety of socially

319 106 2MB

English Pages 372 Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Management Information Systems for Microfinance : Catalyzing Social Innovation for Competitive Advantage [1 ed.]
 9781443885102, 9781443853514

Citation preview

Management Information Systems for Microfinance

Management Information Systems for Microfinance: Catalyzing Social Innovation for Competitive Advantage

Edited by

Arvind Ashta, Bryan Barnett, Karl Dayson and Godfrey Supka

Management Information Systems for Microfinance: Catalyzing Social Innovation for Competitive Advantage, Edited by Arvind Ashta, Bryan Barnett, Karl Dayson and Godfrey Supka This book first published 2015 Cambridge Scholars Publishing 12 Back Chapman Street, Newcastle upon Tyne, NE6 2XX, UK

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

Copyright © 2015 by Arvind Ashta, Bryan Barnett, Karl Dayson, Godfrey Supka and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-5351-8, ISBN (13): 978-1-4438-5351-4

TABLE OF CONTENTS

List of Reviewers...................................................................................... viii Foreword ..................................................................................................... x Raghavan Kunigahalli Preface ...................................................................................................... xiii Arvind Ashta Preliminary Background Introduction ................................................................................................. 1 Karl Dayson Chapter One ................................................................................................. 5 The Evolution of Microfinance Arvind Ashta and Saleh Khan Chapter Two .............................................................................................. 21 Introduction to MIS for Financial Services Sunder Annamraju Part One: Information Requirements for Microfinance Introduction ............................................................................................... 39 Glòria Estapé-Dubreuil Chapter Three ............................................................................................ 44 MIS as a Potential Catalyst for Social Performance Management Frances Sinha, Rupal Patel and Nitin Madan Chapter Four .............................................................................................. 64 MIS and Reporting in Microfinance in the Framework of the Triple Bottom Line Glòria Estapé-Dubreuil, Consol Torreguitart-Mirada and M. Rosa Rovira-Val

vi

Table of Contents

Chapter Five .............................................................................................. 87 An Exploratory Assessment of Customer Intelligence Information Systems in Microfinance Transactions: Evidence from India Djamchid Assadi, Sharam Alijani and Satchidananda Sogala Part Two: Software for Microfinance Introduction ............................................................................................. 101 Marc Ingham Chapter Six .............................................................................................. 108 Drivers to Action: Organisational Buyer Behaviour in the Microfinance Management Information System Market Godfrey Supka Chapter Seven.......................................................................................... 131 The Challenges of Being an MIS Service Provider in Microfinance: Cases from India Gaurav Sinha Chapter Eight ........................................................................................... 155 The Evolving Industry for Microfinance Software: Evaluation and Guide for MFIs and MIS Vendors Arvind Ashta, Vitalie Bumacov, Mikhail Cherkas and Dinos Constantinou Chapter Nine............................................................................................ 178 The Open Source Attitude in Microfinance: the Case of Airdie Vitalie Bumacov, Frederic Lanet and Arvind Ashta Chapter Ten ............................................................................................. 200 SaaS: Strategic Innovation in MIS for Microfinance Markets Arvind Ashta Chapter Eleven ........................................................................................ 225 Risks and Mitigation in Cloud Computing for Microfinance Bryan Barnett

Management Information Systems for Microfinance

vii

Part Three: MIS Implementation in MFIs Introduction ............................................................................................. 241 Gaurav Sinha Chapter Twelve ....................................................................................... 245 Peculiarities of the Microfinance Sector: Success Factors for MIS Implementations Krishna Nyapati and Sandeep Mysore Seshadrinath Chapter Thirteen ...................................................................................... 261 MFI Growth Phase: Difficulties in the Management Information System Mawuli K. Couchoro Chapter Fourteen ..................................................................................... 276 Status of Information Systems in Microfinance Institutions and Overindebtedness of Clients in Democratic Republic of Congo Jacques Bongolomba Isoketsu, Gaurav Sinha, Sunder Annamraju, Rakesh Sud and Aishwarya Srinivasan Part Four: Information Systems Catering to the Microfinance Industry Introduction ............................................................................................. 297 Krishna Nyapati Chapter Fifteen ........................................................................................ 300 Designing a Distributed Microfinance Credit Bureau System Sudeep K Krishnan and Debdatta Pal Chapter Sixteen ....................................................................................... 320 Institutional Work in Building a Credit Bureau for Microfinance: The MFIN Case Study Alok Prasad and Vibhu Arya Biographical Profiles of Authors ............................................................. 335

LIST OF REVIEWERS

Academic Advisory Board Hayyan Alia, University of Franche-Comté, France Isabelle Allemand, Burgundy School of Business, France Djamchid Assadi, Burgundy School of Business, France Laurence Attuel-Mendes, Burgundy School of Business, France Glyn Atwal, Burgundy School of Business, France Vitalie Bumacov, Oxford Brookes University, UK Anuja Cabraal, RMIT University, Melbourne, Australia Mawuli Couchoro, Université De Lomé, Togo Karl Dayson, University of Salford, UK Jason Dedrick, Syracuse University, USA Isabelle Delalex, Columbia School of International & Public Affairs, USA Glòria Estapé-Dubreuil, Universitat Autònoma de Barcelona, Spain Mark Hannam, University of London, UK Marc Ingham, Burgundy School of Business, France Leland Jordan, Christopher Newport University, Virginia, USA Frank Lentz, Burgundy School of Business, France Anabela Mesquita, Polytechnic School of Porto (IPP), Portugal Philipp Otto, European University Viadrina, Germany Rashidah Rahman, Universiti Teknologi Mara, Malaysia Sophie Reboud, Burgundy School of Business, France Rajiv Sabharwal, University of Arkansas, USA Claudio Vitari, Grenoble Ecole de Management, France Damien Wilson, Burgundy School of Business, France

Technical Advisory Board Signis Aliks, FAO-GIZ Microbanker Project, Thailand Sunder Annamraju, Serious Games Consultant, UK Steve Bell, Lean IT Strategies, USA Kevin Day, RISKebiz, Canada Michael Feller, Investment Strategist & Financial Analyst, Australia Yoann Guirimand, Microcred, France Jatinder Handoo, Fino, India

Management Information Systems for Microfinance

ix

Alberto Jimenez, Citigroup, USA T.T. Krishnan, Agricultural Finance Corporation, India Nandu Kulkarni, Banking Technology Consultant, India Raghavan Kunigahalli, Navy Federal Credit Union, USA María Jesús Mariño Gómez, Risk Management Consultant, Spain Minh Huy Lai, CGAP, USA Abu Saleh Mohammad Musa, Credit & Development Forum, Bangladesh Indira Pattni, Entrepreneurship Consultant, Kenya Prateek Shrivastava, Accendo Associates Limited, UK Vikas Kumar Singh, UOB Bank, Singapore Frances Sinha, EDA Rural Systems, India Steve Thomson, Thomson + Associates, Canada Irawan Tjandra, Catur Adhimukti, Indonesia

FOREWORD RAGHAVAN KUNIGAHALLI

We live in an age where technology is rapidly reinventing the financial services industry. From online banking through to digital wallets, how we consume finance has been altered profoundly in recent years. At one level this isn't necessarily a new phenomenon; banking has always embraced technology—from double entry bookkeeping in the fourteenth century to Automated Teller Machines in the 1960s. What makes today’s changes unique, and pertinent for this book, is the global nature of this transformation, and how technology is helping us improve financial services to the poorest communities in the world, namely through the use of management information systems (MIS) in microfinance. Microfinance plays an important role in global economic development by providing opportunities for job creation and increasing financial stability to low-income families. With more than a billion people living under US $1.25 a day, microfinance has a significant potential to play a major role not only in economic uplift of the poor but in creating business opportunities for financial institutions. Year by year growth of 30% during the last decade, highlighted in the Preliminary Background of this book, clearly demonstrates entrepreneurships and global economic uplift opportunities offered by technology advancement in microfinance. With this phenomenal growth it is essential to modernize the information systems and technology infrastructure supporting Microfinance Institutions (MFIs) and their customers. This should ensure sustainable growth and substantially improve outreach, as well as profitability, of microfinance institutions around the world. Advancements in open-source software solutions, cloud-based service offerings and social networks present low-cost opportunities for improved management of loan portfolio, liquidity and credit risks. Several chapters in this book provide useful insights to both practitioners and researchers on opportunities presented by these technological innovations. Microfinance lenders, entrepreneurs and investors will greatly benefit from technology insights presented in these chapters.

Management Information Systems for Microfinance

xi

Identifying market opportunities and reaching out to potential borrowers with low credit risks will significantly enhance capabilities of microfinance institutions to increase market penetration and improve the health of their loan portfolio. In this collection, a number of tools are examined which should help identify market opportunities and target potential microcredit borrowers with appropriate and affordable products. The banking sector has benefited from automated credit rating systems and this could be extremely helpful in improving the financial health of microfinance institutions seeking to manage default rates. The final two chapters discuss this in detail and look at potential approaches to establish a distributed Microfinance Credit Bureau system. This begs the question whether the establishment of credit reporting agencies solely focused on microfinance in various regions of the world would enhance sectoral performance. While it is necessary to look at the health and profitability of microfinance institutions to ensure sustainable growth, it is essential to measure the performance in terms of achieving broader objectives such as economic uplift of the poor. Content presented in the chapters focused on Financial Reporting and Performance Management will help readers to gain a deeper understanding of Social Performance Management and discover Information Technology opportunities to improve the social reporting aspects of microfinance institutions. I believe that this text will serve as a great reference material for all individuals interested in microfinance and the role information systems and technology will play in sustaining the growth observed during the last two decades. This edited collection should help inform the thinking of microfinance practitioners and technology professionals as they plan and implement the use of information systems and technology to improve the performance of their businesses. But most importantly it will help improve the lives of their customers around the world. This book is a culmination of about two years hard work by several experts in microfinance and information technology. Editor-in-Chief Dr. Ashta managed the compilation, peer review and publication process extremely well by partnering with leading microfinance practitioners and researchers around the world to produce a timely and high quality publication. Presentation quality improvement efforts from Dr. Bryan Barnett, Prof. Karl Dayson and Mr. Godfrey Supka helped in meeting the higher quality assurance required for the publication of this book. This book is a valuable asset for microfinance institutions, investors, entrepreneurs and researchers with abundant reference information

Foreword

xii

pertaining to microfinance.

information

systems

and

technology

supporting

PREFACE ARVIND ASHTA

This preface is about my personal journey to editing this book and to organizing it, and will therefore refer a lot to my own work. A fairly exhaustive and far more balanced literature review on research in microfinance was provided by Milana and Ashta (2012) and a more balanced bibliography for Management Information Systems (MIS) issues in microfinance is available at the end of each chapter of this book. This book was initiated thanks to the push from Carol Koulikourdi of our publishers. If she hadn't pushed, and if Stéphan Bourcieu, the Dean of our school and Sophie Reboud, then Research Director of our team at the Burgundy School of Business, had not supported her, it is probable I would have continued to stay in my routine of teaching and writing—a routine which is of immense pleasure to me. Instead, once again, I opted for the toil of editing a book and, once again, learnt more than I contributed. Once again, I was humbled by the number of persons who lent a hand to write and review chapters. Why MIS for Microfinance? Why not something more sexy such as gender issues, for example? To understand this, it is essential to review the research I have been involved in with my co-authors, many of whom are researchers associated with the Banque Populaire Chair in Microfinance of the Burgundy School of Business. From my first paper on microcredit, which was my own introduction to this field (Ashta, 2009), I have been fascinated by this sector, not only because it is a social innovation (Ashta et al., 2013) and has potential for encouraging entrepreneurship (Ashta, 2013; Ashta, Couchoro and Musa, 2013) but also because it is a test for managing double bottom lines, enabling rich lessons to take to our classroom of budding managers and CEOs. Managing a double bottom line, even a triple bottom line, has become important to even mainstream business. I initially looked at what makes microcredit succeed in some countries rather than in others (Allaire et al., 2009; Ashta and Fall, 2012). Don't look at the dates of the publications, because for some inexplicable reason, some papers are published faster than others, depending perhaps on the

xiv

Preface

journal one submits to, the reviewers, and many other factors. In any case, our search for broader influential factors focused on religion (Ashta and De Selva, 2011; Attuel-Mendes, Ashta and Pic, 2012), the state of the economy (Constantinou and Ashta, 2011), laws and ethics. Notably, we were made to look at the social failure of microfinance, brought to the fore by Compartamos's IPO and the controversy it created: perfectly executed private equity model success on the one hand and debatable ethics of charging high interest rates on the other hand (Ashta and Bush, 2009; Ashta and Hudon, 2012; Hudon and Ashta, 2013). This in turn made us look at the need for high interest rates and whether ceilings were necessary. We found that removing ceilings does not lead to more microfinance (Attuel-Mendes and Ashta, 2008; Ashta, Attuel-Mendes and Ratsimalahelo, 2013); therefore, why not keep them? Moreover, we found that the ethical problems persist whenever high profit are made, even if interest rates/yields are lower than the global average of 27% per annum, as was the case of SKS in India (Ashta, Khan and Otto, 2011). Perhaps interest rate transparency would help (Attuel-Mendes and Ashta, 2013). We came to the conclusion that the most important factors to make microfinance successful perhaps lie within the microfinance institutions and not just those in the environment. These factors could be cost management (Khan and Ashta, 2012) and the management of risks (Khan and Ashta, 2013). How can costs be reduced? We have been gradually led to the conclusion that the most important factor is the use of technology. We had been looking at internet technology in what used to be called online lending, social lending or P2P lending, now treated as a part of crowd funding (Ashta and Assadi, 2010a, b; Johnson, Ashta and Assadi, 2010; Assadi and Ashta, 2009, 2012). Based on our research, I decided to edit a first book on advanced technologies for microfinance (Ashta, 2011). The research indicated that besides what is being done on the Internet (Assadi and Hudson, 2010; Ashta and Assadi, 2011; Brett and Stefanakis, 2011), there is a great role for MIS (Augsburg, Schmidt and Krishnaswamy, 2011; Das, 2011; Khan, 2011; Musa and Khan, 2011; Nyapati, 2011; Quadri, Singh and Iyengar, 2011) and mobile banking (Jawadi, Jawadi and Ziane, 2011; Kulkarni, 2011; Morawczynski, 2011; Shrivastava, 2011), as well as for new techniques such as credit scoring (Simbaqueba, Salamanca and Bumacov, 2011) and fuzzy models (Lozano and Fuentes, 2011), and also for new applications (Estapé-Dubreuil and Torreguitart-Mirada, 2011; Hettihewa and Wright, 2011; Sim and Dayson, 2011). Based on that book, and my conversations with leading MFI managers, I became convinced that the single most important factor for scalability

Managementt Information Systems for Miccrofinance

xv

and cost maanagement waas a good MIS S. Delving deeeper, I learned d that the MIS is not oonly an aid too control and performance measurement, but also yields inform mation for strrategic decisio on making. A At times, the MIS M may even be a strrategic strength, enabling a firm to outcoompete (Baruaa, Kriebel and Mukhoppadhyay, 19911; Brown, Gattian and Hickss Jr, 1995). While W MIS technology bby itself is noot the panaceaa (Clemons, R Reddi and Ro ow, 1993) except in ceertain sectors (Clemons an nd Row, 19911), when coup pled with strong and effective opeerational process managem ment, MFIs can gain significant benefits thaat flow to its i bottom-linne, strengtheening its competitive positioning and improving its abilitty to meet customer demands. M Moreover, I disscovered that social innovaations and cosst sharing could becom me possible thrrough cloud computing c andd Software as a Service (Ashta and P Patel, 2013). This ledd me to pondeer if indeed MIS M was as ddry a subject as I had thought, andd whether som mething eclecttic was not haappening in th his sector. Hence the cchoice of theeme for this book. b The quuest for manaaging the double bottoom line of susstainability an nd social ethiccs required foccusing on cost reductiion, through social s innovattion enhancedd by high tecchnology. Indeed, whaat could be sexxier than a forray into such a field at the border of many interaccting disciplinnes? Figure 1 resumes the ddiscussion.

Figure 1

xvi

Preface

It would be appropriate here to define social innovation. Schumpeter (1951, 1989) considered innovation as "a historic and irreversible change in the way of doing things" (p. 138): the introduction of a new product, new method of production, the opening of a new market, the conquest of a new source of raw material; the creation of a new organization. The innovation could be technological, but it may equally be social. "Whereas technological innovation is concerned with application of new technology, social innovation deals with application of new social patterns of human interaction." (Holt, 1971, p. 235)

Holt (1971) was dealing with social innovations within organizations, notably new management practices. Other authors even consider social innovation to be intra-organizational (Westley and Antadze, 2010). It can include new arrangements in society which help to meet unmet social needs, and these actions could take place on local, regional, national or even global levels (Moulaert et al., 2005). Social innovation requires the creation of new social relations to overcome economic and social exclusion and create collective empowerment by boosting the confidence of the excluded and providing them hope of success through better governance structures (Moulaert et al., 2013), including through interaction with eminent but locally embedded actors or "known strangers" (Marti, Courpasson and Dubard Barbosa, 2013). The social innovator may require external partnerships to break the vicious circle (Nurkse, 1952) since they cannot achieve anything alone, but they require the aid of a multiplicity of local actions that permit bringing together not only financial, technical and human capital, but also bridging, bonding and linking social capital (Gerometta, Häussermann and Longo, 2005; Lybbert, 2008) which, on his own, the social innovator lacks. Once a social innovation has been introduced, others may replicate it. A lot of research is being done on seeking competitive advantage through radical technological innovation either by incumbents or by disruptive outsiders (Christensen, Bohmer and Kenagy, 2000), through creating new markets (Christensen, Johnson and Rigby, 2002) rather than just entering the low-end of existing markets. Now, a new sub-category of disruptive innovation, termed catalytic innovation, has been applied to social change (Christensen et al., 2006). The social change to meet an unmet social need is served through a low cost or simpler but satisfactory model, by tapping resources in manners that are initially unattractive to incumbent competitors or their existing business model. Once a workable solution is found, it is scaled through replication.

Management Information Systems for Microfinance

xvii

There are many case studies of successful social innovation bringing out some aspects of the theoretical developments indicated above, including not only microcredit (Yunus, 2003; Maak and Stoetter, 2012; Khavul, Chavez and Bruton, 2013), but also venture philanthropy (Wagner, 2002), participatory budgeting in local finances (Novy and Leubolt, 2005), and urban governance (González and Healey, 2005). In this book we are looking at MIS and a number of chapters address new information needs, new technology adoption and innovations in implementation. The ensemble of these chapters allow us to study whether MIS can help to catalyze the social innovation, which means initiating as well as scaling it, by creating a competitive advantage for those microfinance institutions who invest in such systems. This is not totally novel since it is a continuity of the literature which emerged from notions of competitive advantage linked to strategic planning (Gluck and Kaufman, 1980) and corporate strategies linked to sharing rather than diversification (Porter, 1987), but also linked to other aspects, notably information (Porter and Millar, 1985). Notably, it was found that new IT technology alters industry structure, supports cost differentiation strategies and spawns totally new businesses. In this book, we can see how MIS, notably with new technologies, but even in implementation of existing ones, may create competitive advantage in a socially innovative sector. All this reflects the final sub-title of the book (Catalyzing Social Innovation for Competitive Advantage) which emerged from brainstorming with the authors, the reviewers and the publishers once the book was complete. As a result of the call for papers with the main title (Management Information Systems for Microfinance), I received 31 abstracts and draft papers. Of these, sixteen have made it to the final lap. Some dropped off at the abstract stage; this is not unusual. Others submitted a paper which was reviewed by a team of academic and practitioner reviewers—all volunteers—whose specific mission was to help authors develop their papers further. Most of the authors continued with these revisions, but some were unable to find the time or inclination to revise it and dropped out. The revised paper was then sent back to the reviewers, sometimes to new reviewers as the old ones had run out of their volunteer time. These second stage reviewers then either approved the paper or suggested further minor modifications. The end result of the process is before you. My thanks to all of these as well as to those who submitted work that nonetheless was not included in the final book. The latter encouraged us, by their presence—however brief. The former pressurized me to find the time to complete this work, and not wander off into other pursuits in microfinance. And of course my thanks to all the reviewers.

xviii

Prefaace

Brainstorming with Anabela Mesquita, M thhe co-editor of the Internationaal Journal off Technology y and Humann Interaction n (IJTHI) provided mee with a clear perspective on how to dividde the chapterrs for this book into diifferent parts. An introduction to each off the parts, by y eminent specialists, w will provide more m detail off the chapters enclosed therrein, but I provide a oone-line introdduction to eacch, in recognnition of the enormous e efforts madee by these authhors. The twoo introductoryy chapters ob bviously cam me first. In Chapter 1, Saleh Khan and I presentt the microfinaance sector annd its evolutio on so that a layman ggets a brief inntroduction to o the sector. In Chapter 2, 2 Sunder Annamraju, who spent considerablee time leadinng and guid ding MIS implementattions and tecchnology-led business trannsformation programs p within the ffinancial servvices industry y, provides hiis perspectivee on how MIS evolvedd in the finanncial services sector in the developed wo orld. This provides thee reader with a general ideaa of a benchm mark on what he could expect in miicrofinance. Karl K Dayson provides an inttroduction to these t two chapters. For the rem maining chappters, I used d the followiing Figure 2 on the stakeholderss of microfinaance to bring g clarity to m my thoughts on n how to order the parrts.

Figure 2: Adaapted from Milaan and Ashta (2 2012).

Management Information Systems for Microfinance

xix

Clearly, microfinance has many stakeholders (see shaded boxes) with different information requirements. This was the demand side for information and obviously required to be spelt out first, and we take this up in part I, introduced by Glòria Estapé-Dubreuil. These stakeholders are the institutional partners that need to be satisfied to allow microfinance to innovate successfully. We start, in Chapter 3, with the information requirements on a double bottom line presented by Frances Sinha, Rupal Patel and Nitin Madan, a team from EDA Rural Systems who have considerable experience in this line. Then, in Chapter 4, Glòria EstapéDubreuil, Consol Torreguitart-Mirada and Maria Rosa Rovira-Val, all from the Universidad Autonoma de Barcelona, show us how including a third bottom line, that of sustainable development, would further increase the reporting requirements. Finally, in Chapter 5, Djamchid Assadi, Sharam Alijani and Satchidananda Sogala, a partnership of academics with a practitioner consultant, focus on just one stakeholder, the customer, and propose a customer-intelligence-based model to improve the distribution of financial services to the rural population. Along with these innovations on the information demand side, there have been developments in the supply of information systems, where software suppliers are bringing their own innovations into the process. This supply side of information tools therefore became the second part, introduced by Marc Ingham. We start, in Chapter 6, with Godfrey Supka, a long-time executive with Fern, supplying software to the microfinance sector. He studies MFIs and indicates how MFIs select the software which would produce the information required by the different shareholders. As opposed to the look at MFIs, in Chapter 7, Gaurav Sinha looks at the software vendors themselves and the challenges they face while dealing with MFIs. Then, in Chapter 8, I team up with Vitalie Bumacov, who has now started his own MFI in Moldova, and Mikhail Cherkas and Dinos Constantinou, who are Microfinance Strategy consultants in Switzerland, and we go further to study the features of a large range of software, necessitating incremental innovation and differentiation to retain competitive advantage. In view of the cost of such software for small MFIs that comprise the majority of MFIs, in Chapter 9, Vitalie Bumacov and I team up with Frederic Lanet of Airdie, which was the first MFI to experiment with the first open-source software, Octopus, to understand whether the open-source attitude of microfinance can be combined with the open-source attitude of free software. In Chapter 10, I present a different kind of technological innovation, that of SaaS, and discuss how it would change social relations between the suppliers and the MFIs as well as relations between the MFIs, disrupting the competitive advantage

xx

Preface

equation. Finally, in Chapter 11, Bryan Barnett, a long time consultant in IT in the microfinance sphere, cautions that cloud computing technologies, such as SaaS, carry their own risks. Irrespective of the kind of software adopted, bought, rented, free or home-grown, the proof of the pudding is in the eating. Part III, introduced by Gaurav Sinha, looks at the implementation of software in MFIs. In Chapter 12, Krishna Nyapati and Sandeep M.S, with considerable experience in IT, start the ball rolling by indicating the success factors for implementing MIS and examine a case study of MIS implementation in an Indian MFI, BSFL. In Chapter 13, Mawuli Couchoro studies the difficulties of MFI software in coping with the rapid growth of microfinance in Togo. In perhaps the most challenging work, in Chapter 14, Jacques Bongolomba Isoketsu is joined by a team of researchers including Gaurav Sinha, Sunder Annamraju, Rakesh Sud and Aishwarya Srinivasan to comment on the need for information systems to help overcome client over-indebtedness in the war-torn Democratic Republic of Congo, perhaps the poorest country on the planet. The over-indebtedness problem is perhaps the most important problem facing microfinance today, leading to reputation risk as well as credit risk for the sector (CSFI 2011, 2012). Recognizing these risks, a number of researchers and policy makers have suggested that credit bureaus be introduced to overcome information asymmetry faced by microfinance institutions in the wake of competing MFIs lending to the same borrowers. Part IV, introduced by Krishna Nyapati, looks at these macro-industry needs and the rapid response of the industry to provide these macro-industry information systems. In Chapter 15, Sudeep Krishnan and Debdatta Pal review the literature and interview experts to propose a distributed system designed for Microfinance Credit Bureau Management Information Systems. Finally, in a sharp response to theory, Alok Prasad, the head of the Microfinance Institutions Network and consultant Vibhu Arya show how the institutional work of creating and developing credit bureaus in India at a rapid pace was supported by the microfinance industry, in an effort to reduce the risks and convey to the stakeholders that the message had been heard. This rapid capability of an industry to come together to address industry issues perhaps reminds us that competitive advantage is important but continuing a social innovation which serves so many is even more important. Thus information systems at the firm level as well as those at the industry level are both required. I hope that you will enjoy reading the book and find it useful. Dijon, France, 2013

Management Information Systems for Microfinance

xxi

Bibliography Allaire, Virginie, Arvind Ashta, Laurence Attuel-Mendes, and Karuna Krishnaswamy. 2009. "The success of Moroccan Microfinance institutions: more than just culture?" Euro-Mediterranean Economics And Finance Review no. 4 (3, Special Issue for IFC 5):53-74. Ashta, Arvind. 2009. "Microcredit Capital Flows and Interest Rates: An Alternative Explanation." Journal of Economic Issues (M.E. Sharpe Inc.) no. 43 (3):661-683. —. 2011. Advanced technologies for microfinance: Solutions and challenges. Hershey, PA: IGI Global. Ashta, Arvind. 2013. "Cooperating For Entrepreneurship: The French Regime of Cooperative of Activities and Employment." In Cooperative and Microfinance Revolution, edited by Onafowokan O. Oluyombo, 53-61. Lagos, Nigeria: Soma Prints Limited. Ashta, Arvind, and Djamchid Assadi. 2010a. "An analysis of European online micro-lending websites." Innovative Marketing no. 6 (2):7-17. Ashta, Arvind, and Djamchid Assadi. 2010b. "Should online microlending be for profit or for philanthropy? DhanaX and Rang De." Journal of Innovation Economics no. 2 (6):123-146. Ashta, Arvind, and Djamchid Assadi. 2011. "The use of Web 2.0 technologies in online lending and impact on different components of interest rates." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Ashta, Arvind, Laurence Attuel-Mendes, and Zaka Ratsimalahelo. 2013. "Another "French paradox": Explaining why interest rates to microenterprises did not increase with the change in French usury legislation." European Journal of Law and Economics no. 35 (1). Ashta, Arvind, and Matthew Bush. 2009. "Ethical Issues of NGO Principals in Sustainability, Outreach and Impact of Microfinance: Lessons in Governance from the Banco Compartamos" I P O. Management Online Review November: 1-18, www.morexpertise.com Ashta, Arvind, Mawuli Couchoro, and Abu Saleh Mohammad Musa. 2013. "Microfinance and Entrepreneurship." In Encyclopedia of Creativity, Invention, Innovation, and Entrepreneurship, edited by Elias G. Carayannis. New York, NY: Springer. Ashta, Arvind, Karl Dayson, Rajat Gera, Samanthala Hettihewa, N.V. Krishna, and Christopher Wright. 2013. "Microcredit as a Social Innovation." In The International Handbook on Social Innovation, edited by Frank Moulaert, Diana MacCallum, Abid Mehmood and Abdelillah Hamdouch, 80-93. Cheltenham, UK.: Edward Elgar.

xxii

Preface

Ashta, Arvind, and Rosita De Selva. 2011. "Religious Practice and Microcredit: Literature Review and Research Directions." Postmodern Openings no. 2 (8):33-44. Ashta, Arvind, and Ndyey Salimata Fall. 2012. "Institutional analysis to understand the growth of microfinance institutions in West African Economic and Monetary Union." Corporate Governance no. 12 (4):441 - 459. Ashta, Arvind, and Marek Hudon. 2012. "The Compartamos microfinance IPO: Mission conflicts in hybrid institutions with diverse shareholding." Strategic Change: Briefings in Entrepreneurial Finance no. 21 (7-8):331-341. Ashta, Arvind, Saleh Khan, and Philipp E. Otto. 2011. "Does Microfinance Cause or Reduce Suicides? Policy Recommendations for Reducing Borrower Stress." Available at SSRN: http://ssrn.com/abstract=1715442. Ashta, Arvind, and Jiten Patel. 2013. "Software as a Service: An opportunity for disruptive innovation in the Microfinance software market?" Journal of Innovation Economics no. 11 (1):55-82. Assadi, Djamchid, and Arvind Ashta. 2009. "How do People Trust on Peer-to-Peer Lending Websites? Analysis of the Impacts of the Web 2.0 Technologies and Intermediation Roles." In Advances in Technology and Innovation in Marketing, edited by Rajat Gera, 49-69. Delhi: MacMillan Publishers India Ltd. Assadi, Djamchid, and Arvind Ashta. 2012. "Brand Value Building In Online Social Lending Startups." Journal of Innovation Economics no. 1 (9):139-161. Assadi, Djamchid, and Meredith Hudson. 2010. "Marketing Analysis of Emerging Peer-to-Peer Microlending Websites." In Advanced Technologies for Microfinance, edited by Arvind Ashta, 185-205. Hershey, PA: IGI Global. Attuel-Mendes, Laurence, and Arvind Ashta. 2008. "French Legislation and the Development of Credit Availability for Microenterprise." Global Journal of Business Research (GJBR) no. 2 (2):123-137. Attuel-Mendes, Laurence, and Arvind Ashta. 2013. "The Truth, But Not Always The Whole Truth In Lending Laws." Cost Management no. 27 (2):6-19. Attuel-Mendes, Laurence, Arvind Ashta, and Emmanuel Pic. 2012. "La finance responsable face aux grandes religions." In Management et religions: Décryptage d'un lien indéfectible, edited by Isabelle Barth, 171-186. éditions EMS.

Management Information Systems for Microfinance

xxiii

Augsburg, Britta, Jan Philipp Schmidt, and Karuna Krishnaswamy. 2011. "Free & Open Source Software for Microfinance: Increasing Efficiency and Extending Benefits to the Poor." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Barua, Anitesh, Charles H. Kriebel, and Tridas Mukhopadhyay. 1991. "An Economic Analysis of Strategic Information Technology Investments." MIS Quarterly no. 15 (3):313-331. Brett, Daniel, and Nikias Stefanakis. 2011. "EDA CapitalConnect: An Online Platform for Social Enterprise Financing." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Brown, Robert M., Amy W. Gatian, and James O. Hicks Jr. 1995. "Strategic Information Systems and Financial Performance." Journal of Management Information Systems no. 11 (4):215-248. Christensen, Clayton M., Heiner Baumann, Rudy Ruggles, and Thomas M. Sadtler. 2006. "Disruptive Innovation for Social Change." Harvard Business Review no. 84 (12):94-101. Christensen, Clayton M., Richard Bohmer, and John Kenagy. 2000. "Will Disruptive Innovations Cure Health Care?" Harvard Business Review no. 78 (5):102-112. Christensen, Clayton M., Mark W. Johnson, and Darrell K. Rigby. 2002. "Foundations for Growth." MIT Sloan Management Review no. 43 (3):22-31. Clemons, Eric K., Sashidhar P. Reddi, and Michael C. Row. 1993. "The Impact of Information Technology on the Organization of Economic Activity: The "Move to the Middle" Hypothesis." Journal of Management Information Systems no. 10 (2):9-35. Clemons, Eric K., and Michael C. Row. 1991. "Sustaining IT Advantage: The Role of Structural Differences." MIS Quarterly no. 15 (3):275292. Constantinou, Dinos, and Arvind Ashta. 2011. "Financial crisis: lessons from microfinance." Strategic Change: Briefings in Entrepreneurial Finance no. 20 (5/6):187-203. doi: 10.1002/jsc.895. CSFI. 2011. “Microfinance Banana Skins 2011: the CSFI survey of microfinance risk.” U.K.: Centre for the Study of Financial Innovation (CSFI). —. 2012. Microfinance Banana Skins 2012: the CSFI survey of microfinance risk - Staying Relevant. U.K.: Centre for the Study of Financial Innovation (CSFI).

xxiv

Preface

Das, Puspadhar. 2011. "A Case Study of Mifos implementation at Asomi." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Author. Ashta, Arvind. Estapé-Dubreuil, Glòria, and Consol Torreguitart-Mirada. 2011. "Information Technology and Microfinance in Developed Countries: The Spanish Case, with a Focus on Catalonia." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Gerometta, Julia, Hartmut Häussermann, and Giulia Longo. 2005. "Social innovation and civil society in urban governance: Strategies for an inclusive city." Urban Studies (Routledge) no. 42 (11):2007-2021. doi: 10.1080/00420980500279851. Gluck, Frederick W., and Stephen P. Kaufman. 1980. "Strategic management for competitive advantage." Harvard Business Review no. 58 (4):154-161. González, Sara, and Patsy Healey. 2005. "A Sociological Institutionalist Approach to the Study of Innovation in Governance Capacity." Urban Studies (Routledge) no. 42 (11):2055-2069. doi: 10.1080/0420980500279778. Hettihewa, Samanthala, and Christopher S. Wright. 2011. "Microfinance from LDCS to DCS: Are Socio-economic Differences Important?" In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Holt, Knut. 1971. "Social Innovations In Organizations." International Studies of Management & Organization no. 1 (3):235-252. Hudon, Marek, and Arvind Ashta. 2013. "Fairness and microcredit interest rates: From Rawlsian principles of justice to the distribution of the bargaining range." Business Ethics: A European Review no. 22 (3):277-291. Jawadi, Fredj, Nabila Jawadi, and Ydriss Ziane. 2011. "Could ICT Improve Microfinance Efficiency and Reduce Its Imperfections? Further Evidence from Developing and Emergent Financial Market." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Johnson, Susan, Arvind Ashta, and Djamchid Assadi. 2010. "Online or Offline? The Rise of "Peer-to-Peer" Lending in Microfinance." Journal of Electronic Commerce in Organizations no. 8 (3):26-37. Khan, Saleh. 2011. "Automating MFIs – how far should we go?" In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta.

Management Information Systems for Microfinance

xxv

Khan, Saleh, and Arvind Ashta. 2012. "Cost control in Microfinance: Lessons from the ASA case." Cost Management no. 26 (1):5-22. Khan, Saleh, and Arvind Ashta. 2013. "Managing Multi-Faceted Risks in Microfinance Operations." Strategic Change no. 22 (1/2):1-16. doi: 10.1002/jsc.1918. Khavul, Susanna, Helmuth Chavez, and Garry D. Bruton. 2013. "When institutional change outruns the change agent: The contested terrain of entrepreneurial microfinance for those in poverty." Journal of Business Venturing no. 28 (1):30-50. doi: http://dx.doi.org/10.1016/j.jbusvent.2012.02.005. Kulkarni, Nandu. 2011. "A Technology Requirements and Governance Framework for Rural Microfinance - an Indian perspective." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Lozano, Carmen, and Federico Fuentes. 2011. "A Systemic - Fuzzy Model to Evaluate the Social Impact of Microcredits." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Lybbert, Travis J. 2008. "Exploring the Role of Spiritual Capital in Poverty Traps and Microfinance." Faith and Economics (51):57-79. Maak, Thomas, and Nicolas Stoetter. 2012. "Social Entrepreneurs as Responsible Leaders: 'Fundación Paraguaya' and the Case of Martin Burt." Journal of Business Ethics no. 111:413-430. doi: 10.1007/s10551-012-1417-0. Marti, Ignasi, David Courpasson, and Saulo Dubard Barbosa. 2013. "“Living in the fishbowl.” Generating an entrepreneurial culture in a local community in Argentina." Journal of Business Venturing no. 28 (1):10-29. doi: http://dx.doi.org/10.1016/j.jbusvent.2011.09.001. Milana, Carlo, and Arvind Ashta. 2012. "Developing Microfinance: A Survey of the Literature." Strategic Change: Briefings in Entrepreneurial Finance no. 21 (7-8):299–330. Morawczynski, Olga. 2011. "Saving through the mobile: A study of MPESA in Kenya." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Moulaert, Frank, Diana MacCallum, Abid Mehmood, and Abdelillah Hamdouch. 2013. "General Introduction: The Return of Social Innovation as a Scientific Concept and a Social Practice." In The International Handbook on Social Innovations, edited by Frank Moulaert, Diana MacCallum, Abid Mehmood and Abdelillah Hamdouch, 1-6. Cheltenham, UK: Edward Elgar Publishing.

xxvi

Preface

Moulaert, Frank, Flavia Martinelli, Erik Swyngedouw, and Sara González. 2005. "Towards alternative model(s) of local innovation." Urban Studies (Routledge) no. 42 (11):1969-1990. doi: 10.1080/00420980500279893. Musa, Abu Saleh Mohammad, and Mostafa Saidur Rahim Khan. 2011. "Implementing Point of Sale Technology in Microfinance: An Evaluation of Come To Save (CTS) Cooperatives, Bangladesh." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Novy, Andreas, and Bernhard Leubolt. 2005. "Participatory budgeting in Porto Alegre: Social innovation and the dialectical relationship of state and civil society." Urban Studies (Routledge) no. 42 (11):2023-2036. doi: 10.1080/00420980500279828. Nurkse, Ragnar. 1952. "Some International Aspects Of The Problem Of Economic Development." American Economic Review no. 42 (2):571. Nyapati, Krishna. 2011. "Stakeholder Analysis of IT Applications for Microfinance." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Porter, Michael E. 1987. "From competitive advantage to corporate strategy." Harvard Business Review no. 65 (3):43-43. Porter, Michael E., and Victor E. Millar. 1985. "How information gives you competitive advantage." Harvard Business Review no. 63 (4):149160. Quadri, S.Mohd. Najmullah, Vikas Kumar Singh, and Kishen Parthasarathy Iyengar. 2011. "IT and MIS in Microfinance Institution: Effectiveness and Sustainability issues." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Schumpeter, Joseph A. 1951, 1989. "Essays: On Entrepreneurs, Innovations, Business Cycles, and the Evolution of Capitalism." In, edited by Richard V. Clemence. New Brunswick, NJ: Transaction Publishers. Shrivastava, Prateek. 2011. "Predicting the attitude toward mobile financial services in developing countries." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Sim, Jack, and Karl Dayson. 2011. "Portal to Portaloo: Can Microfinance and IT Help Solve the World’s Sanitation Crisis?" In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Simbaqueba, Lilian, Gehiner Salamanca, and Vitalie Bumacov. 2011. "The Role of Credit Scoring in Micro Lending." In Advanced

Management Information Systems for Microfinance

xxvii

Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Hershey, PA: IGI Global. Wagner, Lilya. 2002. "The 'new' donor: Creation or evolution?" International Journal of Nonprofit & Voluntary Sector Marketing no. 7 (4):343. Westley, Frances, and Nino Antadze. 2010. "Making a Difference: Strategies for Scaling Social Innovation for Greater Impact." Innovation Journal no. 15 (2):1-19. Yunus, Muhammad. 2003. Banker to the Poor: Micro-Lending and the Battle Against World Poverty. New York: Public Affairs.

PRELIMINARY BACKGROUND INTRODUCTION KARL DAYSON

'People asked me: "How did you figure out all those rules that you say you developed?" I tried to explain, but they didn't understand what I was saying. So now I put it a different way. I tell them that whenever I need a rule in a particular situation, I just look at how the conventional banks do it. After all, they have been in business for a long time. When I find out how they do it, all I have to do is do the opposite! And it works beautifully.' (Mohammed Yunus, Egon Zehnder International Speech, January 2009).

As Yunus highlights, microfinance is inherently contrarian. To understand it requires knowledge of the mainstream banking system and how it functions, but to act involves ignoring its assumptions and ideology. These first two chapters explain each side of this dichotomy. Annamraju details the banks’ long-standing investment, punctuated by periods of both technological innovation and sclerotic technology, caused by their thirst for merger and acquisitions-driven expansion. By contrast, Ashta and Khan describe the spectacular and predominantly organic growth of the microfinance sector. Both, though, indicate the dangers and risks ahead for microfinance as it begins to morph into a larger, more consolidated industry. But first let’s take a step back and address the question central to the Preface, introductory chapters and the wider book. Should technology matter to microfinance institutions? As a social scientist, one is always wary of technological determinism and narratives that justify action on the basis of a specific manufactured innovation. Sociology emerged from the furnace of the Industrial Revolution and has struggled with the question about whether technology drives societal change or whether societal change enabled the space for the adoption of new technologies. The answer is probably far more complex. But that debate explains why this book matters and why the introductory chapters are necessary.

2

Preliminary Background

Microfinance was invented (or more accurately, re-invented) in the period following decolonisation and emancipation of the developing world. Mohammed Yunus found that economic freedom only applied to some, specifically men and the new ruling elites. As in Orwell's parable 'Animal Farm,' these elites replicated the practices and institutions so reviled under the ancien régime. This meant local and national banks served their interests and ignored the needs of the poor, particularly women, crashing into the gender uprising of the 1970s when women from across the world began to assert their rights and demand equality. Starting in South Asia, microfinance spread throughout the developing world, eventually reaching Europe following the collapse of the Eastern Bloc in 1989. Meanwhile, the Cold War, driven by the arms and space race, had become a technology sprint as both sides sought to deliver more powerful weapons, satellites and rockets. Ideological and financial inducements pushed the Americans to realise the benefits of the microchip and the Internet long before they became available to consumers. Enabling this was an open scientific community where challenge and criticism were embedded in the culture, and along with enforced patent laws created an environment where design and development processes could respond quicker to IT inventions. When discussing the relationship between IT and microfinance, it is prudent to be mindful of these societal factors. Technology is not a random thing; it is a function of ideology, finance, anticipated return and some degree of transparency. It could be argued that the evolution of microfinance has the same characteristics. Viewed through a societal lens, we are able to see why technology matters to microfinance. The data used by Ashta and Khan is from MIX Market, a non-profit foundation based in Washington DC, reporting on about 2000 microfinance providers and with an avowed objective of strengthening the sector by promoting transparency. The design of the website makes MIX an essential and flexible tool to help managers and investors to compare performance and identify good practice. With every territory having its own regulatory framework it would be almost impossible to bring these information sources together in a comparative form without the innovation of MIX Market. By empirical capture of evidence around good practice, MFIs are able to improve profitability and provide a better return for investors. This in turn establishes a virtuous circle where good returns encourage further investment, which in turn enables management to invest in management systems that can better monitor and improve financial performance. As Annamraju indicates, not all IT investment systems produce such an

Introduction

3

outcome, but given the historically low level of expenditure on these services within MFIs, it should be possible to make strategic choices that can deliver immediate benefit (e.g. telephony based applications or field officer to headquarters communications). These 'low-hanging fruit' allow time for MFIs to build confidence and skills in operating MIS, thereby creating an environment where more expansive systems can be considered. For investors, the capacity of an MIS to record and present the use of their money is easier and more reliable, probably giving the MFI a competitive edge over those reliant on paper-based systems. It is especially useful if the investor is interested in a specific sub-group of clients and demands regular reporting. The cost of collecting and compiling this data is often overlooked by MFIs, invariably because it was made by a social or government investor that anticipated a lower return. As MFIs draw more heavily on commercial lenders, finding mechanisms to reduce reporting costs will become more important and draw more institutions towards increasingly sophisticated systems. Despite this talk of profit and investment return, most MFIs have a social mission and ideologically this creates an imperative for better and more efficient systems. It is this that separates MFIs from the mainstream banks. As Ashta convincingly argues in the Preface, if an MFI is serious about delivering the highest quality service at the lowest possible price, finding ways to reduce management and operational costs should be central to its activity. Annamraju points out how the banks’ inorganic growth has led to duplication and obsolescence of systems. Ultimately, the cost of this inefficiency is either borne by the shareholders or the customer. With the pressure to maximise shareholder value, it is the consumer that pays through lower savings rates, higher interest rates on loans, or just poorer service. If we adopt Yunus's position, MFIs should reject this approach and always look to minimise the cost of doing business and pass on the benefit to the customer. This is the ideological justification for investing in quality technological solutions. If we focus on Yunus' maxim, Ashta and Khan's introductive chapter explains how MFIs have reached ever-increasing numbers of people, but the risks are all too apparent. While profitability helps fund growth, there is a danger some MFIs will prioritise this over their social mission. Are MFIs serving the right people, do they have an appropriate credit mix and does the pricing reflect the risks and costs involved in delivering the service? Ashta and Khan's data is partially reassuring, and their argument around streamlining costs is persuasive, but questions still remain as the sector matures.

4

Preliminary Background

It is maturity and the relentless concentration on returns that is striking from Annamraju's review of the use of MIS within the conventional banking community. Banks with global pretensions engaging in complex takeovers require the integration of a plethora of IT systems. Following Yunus, it is probably best if MFIs don't expand in this way. Equally, the reliance on in-house systems seems to result in legacy problems which should encourage MFIs to consider Cloud and other third-party networked approaches, without losing ownership of the data. Finally, Annamraju's table comparing banks and MFIs suggests that operating the latter is a more complex business with a wider range of stakeholders and greater difficulty in reaching clients. For this reason alone MIS could deliver greater benefit to microfinance institutions, and for me this is why this book matters.

CHAPTER ONE THE EVOLUTION OF MICROFINANCE ARVIND ASHTA AND SALEH KHAN

Over recent decades, microfinance has been the poster child for economic development around the world. Riding on the image of providing micro loans, and helping the entrepreneurial poor to progress out of poverty, the sector has been booming. This extraordinary growth has been fueled by both public and private capital, investment funds, commercial bankers providing loans, and more recently, equity from venture capital funds. There have even been Initial Public Offerings (IPO) in the sector. Number of institutions Total number of clients reached

4000 3500

200 3000 2500

150

2000 100

1500 1000

50

Number of institutions

Number of Clients Reached, Millions

250

500 0

0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2009 2010 2011

Figure 1: Growth of the microfinance sector. Source: MIX Market, 2013.

This amazingly high growth rate has averaged around 30% year on year over the last decade (depicted by the overall trend in Figure 1). Within this period some institutions have doubled their loan portfolio

6

Chapter One

every year. However, this phenomenal growth can only be sustainable if the microfinance institution (MFI) is extremely well managed. If appropriate internal control mechanisms and information systems are not put in place, this growth can go awry—leading to over-indebtedness among clients. The debt-trap cycle unmistakably sows the seeds of immitigable credit risk, loan portfolio loss and possibly bankruptcy. We argue that the single most important enabler for sustainable growth is an effective and efficient Management Information System (MIS), hence the purpose of this book. The aim of this chapter is to highlight specific features of microfinance operations that would permit the reader to gain an appreciation of the chapters that follow.i For this chapter and unless otherwise stated, we draw upon publicly available data from the Microfinance Information Exchange (MIX). MIX obtains fairly detailed data from microfinance institutions (MFIs) that voluntarily submit this information. Today, much of the available information is based on audited figures reported by the MFIs. In 2011, 1361 MFIs reported to MIX, representing total assets worth $118 billion, of which $88 billion was their gross loan portfolio. The assets are financed through a blend of deposits ($67 billion), debt ($20 billion) and equity ($18 billion). The MFIs reach out to a total of 94 million borrowers and 77 million depositors, and they employ 545,000 people, including 258,000 loan officers. The maturity of MFIs, their outreach and growth varies by region. MIX classifies the world of microfinance in six regions: Latin America and Caribbean (LAC), Sub-Saharan Africa (SSA), Middle East and North Africa (MENA), South Asia (SA), East Asia and Pacific (EAP), and Eastern Europe and Central Asia (EECA). The number of MFIs per region varies from 58 in MENA to 379 in LAC. The combined total assets and gross loan portfolio are highest in EAP ($46 billion and $31 billion respectively) and lowest in MENA ($2 billion and $1 billion). Deposits are also highest in EAP ($30 billion). However, the greatest number of borrowers and depositors are in South Asia (50 million and 24 million). Approximately 214,000 people are employed in this region.

Overcoming information asymmetry Lending to the poor with little or no collateral is one of the key features of microcredit. In lieu of tangible collateral, microfinance institutions (MFI) often focus on evaluating a client’s “social collateral” as a mechanism to managing credit default risk. Social collateral involves using guarantees from one's group or social network. From a risk

The Evolution of Microfinance

7

management perspective, it therefore becomes important to properly identify borrowers and establish enough processes to ensure timely repayment of the loans. This is where technology and, especially, a good information system comes into play (Ashta, 2011). As with the formal financial industry, the microfinance sector can benefit from an effective information management strategy, underpinned by a strong Management Information System (MIS) platform. However, they become particularly essential to an industry which continually struggles with overcoming information asymmetry in the credit risk management process. Traditionally, formal financial institutions have not been able to provide small loans to the poor because of three major problems: information asymmetry (adverse selection, moral hazard, difficulties of recovering delinquent loans), high transaction costs associated with low loan sizes, and the lack of other complementary capital (human and social) with the poor (Armendàriz and Morduch, 2010). Microcredit overcame most of these problems by using a group lending model. The nature and size of the groups varied, but early experiments in different parts of the world showed that groups of borrowers, based on social peer groups, had insight and information on a potential client’s creditworthiness— something that the lender did not have. Therefore, a lender could leverage the information available with the group to distinguish between safe and risky borrowers, to monitor them and to enforce repayment. The group dynamics can also be used to ensure timely repayments and reduced transaction costs. Finally, through peer support networks, members of the group can help each other succeed in their enterprises. Once a group member’s credit worthiness has been determined, i.e. she has successfully paid back a loan, the information asymmetry is reduced and the lender can then provide an individual loan to the borrower— essentially doing away with the need for a group dynamic. An efficient MIS goes a long way towards capturing this information and providing institution memory of a single borrower’s behavior. This data can also support local Asset and Liability Committees (ALCOs) make more informed decisions in the loan approval process. Over time the successful capturing of payback information by various categories of borrowers creates a knowledge base within the MFI—and, if properly leveraged, can be used to create its own credit scoring algorithm based on past experiences and performance. The move to this step, of course, pre-assumes that there is a strong MIS in place which can store and analyze historic information.

Chapter One

8

There are numerous ways to capture the transaction history, and successive chapters in this book touch upon some of them.

Increasing loan sizes Since microfinance loans are made to the poor, usually with no physical collateral, the size of the loan is typically small, or at least smaller than what is granted by commercial banks or even SME banks. But exactly how small are these loans? The popular image of microfinance used to be that of a twenty-dollar loan which changed a person’s life. However, inflation has eaten into the value of the dollar and twenty dollars no longer permits people to finance enterprises even in the lowest end of the socio-economic pyramid. Even at the turn of the century, the average loan size was ten times larger. We therefore need to get an updated and realistic idea of what exactly is a microcredit. MFIs ranked by average loan outstanding First quartile Median Third quartile 90th percentile % of MFIs with average loan balance > $5,000

1999 $88 $239 $644 $1 610

2002 $83 $192 $671 $1 459

2005 $106 $297 $902 $1 987

2008 $182 $511 $1 443 $2 988

2011 $227 $607 $1 605 $3 737

0.8%

2.2%

2.2%

5.3%

6.6%

Table 1: Average size of microcredit loans (USD). Source: MIX Market, 2013. Based on MIX data, Table 1 above indicates that the median MFI in 1999 had an average loan size of $239 and that this grew to $607 in 2011. The first quartile loan size has grown from $88 to $227 during this period. The third quartile loan size has grown to $3,737. Moreover, the number of MFIs providing loans greater than $5,000 has grown from less than 1% to almost 7%. One is tempted to hypothesize that MFIs are more and more ‘up scaling’ their lending to reach out to micro small enterprises (MSME). However, one must keep in mind that there is regional diversity in lending with loan sizes varying across regions. Thus, the current median microloan in South Asia is only $154, while in EECA it is $1,862 and in Latin America it is $1,027.

The Evolution of Microfinance

9

Thus, the image of microloans being less than a hundred dollars came from early books based on lending experience in Bangladesh and India. We need to update our understanding of global microloans to be in the ballpark of $600. To explain this trend of increasing loan size, it is customary for proponents in the microfinance sector to explain that the average loan size increases with inflation and due to increased financial need that comes with the organic growth of the borrower’s business. Both inflation and business size growth, for the economy as a whole, are captured in the growth rate of Gross National Income (GNI) per capita. Therefore, it is customary to compare the microcredit loan sizes to GNI per capita. The Microfinance Information Exchange (MIX) classifies lenders as being MFIs if their average outstanding loan balance is below 250% of per capita GNI. MFIs Ranked by loan size as % of GNI per capital First quartile Median Third quartile 90th Percentile % of MFIs with average loan balance > 2.5 times GNI per capita

1999 13% 29% 67% 159%

2002 15% 34% 81% 175%

2005 14% 31% 72% 158%

2008 12% 28% 65% 146%

2011 11% 28% 68% 148%

5.3%

7.4%

5.7%

4.4%

5.0%

Table 2 : Average loan size as a % of GNI per capita. Source: MIX Market, 2013. We can see from Table 2 above that the average loan size for a first quartile MFI was 13% of the per capita GNI in which it operates. For the median MFI, it averages around 29% of the per capita GNI, while for the third quartile it hovers around 70%. Regional variations also come into play with average loans from the median MFI in South Asia being only 12% of GNI per capita, while in sub-Saharan Africa (SSA) they are at 60%. However, by and large, the ratio or average loan size to per capita GNI has largely remained in the same range even through the absolute loan size has increased—adding credibility to the argument that the increase in loan

Chapter One

10

size is primarily due to the growth of the economy and accounting for inflation. But, even using MIX’s exaggerated definition of microloans being anything less than 250% of the per capita GNI (which is far removed from the 11-68% range we see), there are about 5% of microfinance institutions reporting to the MIX who are clearly not providing “microloans.”

Increasing client outreach These $600 loans are made to thousands of clients, if not hundreds of thousands. We can see that large MFIs such as the Vietnamese Bank for Social Policies (VBSP), Grameen Bank, ASA, BRAC and SKS all have millions of borrowers. However, these extreme outliers do not really explain a sector very well. The panorama from the MIX in Table 3 below shows that the first quartile of MFIs have remained with less than 2,000 borrowers on average, perhaps reflecting newer MFIs who start reporting to the MIX database. Nevertheless, the median MFI has grown from 6,500 borrowers to 8,000 borrowers. The third quartile figure has nearly doubled from 19,000 borrowers to 35,000 borrowers. Finally, the 90th percentile has more than doubled from 49,000 borrowers to 117,000 borrowers. Therefore, we can see that growth of the MFIs has come from increasing outreach to more and more new clients promoting financial inclusion and ensuring that more of the entrepreneurial poor population around the world has access to financial services. MFIs ranked by average number of borrowers First quartile Median MFI Third quartile 90th percentile

1999

2002

2005

2008

2011

1 968 6 459 18 974 48 819

1 284 5 164 15 652 41 654

1 453 5 455 16 468 54 117

1 477 6 987 26 436 82 302

1 857 8 031 35 211 116 578

Table 3: Average number of borrowers per MFI. Source: MIX Market, 2013. The MIX benchmarks MFIs into three distinct categories: x Small: with less than 10,000 borrowers, x Medium: between 10,000 and 30,000 borrowers, x Large: with more than 30,000 borrowers.

The Evolution of Microfinance

11

All three categories of MFIs have grown considerably, but this reflects the fact that the number of MFIs reporting to MIX has grown tenfold during the last 12 years. However, looking from the data in Table 4, one can see that the distribution between the three categories has changed. The percentage of large MFIs has gone up while the number of medium institutions has gone down—but the percentage share of small MFIs has remained almost the same at 56 percent of the sector. 1999 Number of small MFIs ( 30,000 borrowers) % of large MFIs

2002

2005

2008

2011

77

334

717

764

684

57.5%

65.9%

63.0%

56.5%

55.2%

35

100

234

281

232

26.1%

19.7%

20.6%

20.8%

18.7%

22

72

187

307

324

16.4%

14.2%

16.4%

22.7%

26.1%

Table 4: Categories of MFI and their distribution. Source: MIX Market, 2013. This perhaps indicates that the number of new entrants in the market has helped grow the sector, whilst a number of ‘medium’ sized institutions have grown to be ‘large’ institutions. A gradual progression of growth both in terms of client outreach and loan size is indicative of any maturing financial industry. Again, the regional disparity is interesting. The worldwide median MFI may have 8,000 borrowers, but the median MFI in Eastern Europe (EECA) has only 2,000, whilst the median MFI in South Asia has 33,000.

Fostering a sustainable financial sector The takeaway, therefore, is that a median MFI in 2011 services 8,000 loans of $600 in size, which are repaid in either a weekly, fortnightly or monthly basis. If on a weekly basis the repayment amounts on this $600 loan are likely to be very small, consequently operating costs around counting, checking and recording thousands of weekly small transactions

Chapter One

12

are very high. So how do MFIs become sustainable and finance their growth?

Interest rates are high One of the answers is that MFIs typically tend to charge higher interest rates than commercial banks. In microfinance, the trend is to use the yield on gross loan portfolio (GLP) as a proxy for actual interest rate being charged to the borrowers. This is because different MFIs use different methods for charging interest rates (flat or declining interest rate calculations), include various service fees, often add compulsory savings to the loan requirement and sometimes require an initial down payment. There is a movement to increase transparency in lending and divulge the true cost of borrowing to the client; however, while an industry benchmark is being reached, the yield on GLP serves as a suitable proxy. Region

2003

2005

2007

2009

2011

Sub-Saharan Africa

35%

40%

37%

40%

37%

East Asia and the Pacific

42%

39%

34%

33%

29%

Eastern Europe and Central Asia

42%

32%

30%

32%

30%

Latin America and The Caribbean

37%

36%

35%

36%

37%

Middle East and North Africa

35%

33%

29%

30%

30%

South Asia

24%

21%

25%

24%

24%

Global average

37%

34%

32%

33%

32%

Table 5: Average yield on GLP (nominal), per region. Source: MIX Market, 2013. Based on information from the MIX, Table 5 indicates that the average yield for a MFI in 2003 was 37% but this had gradually reduced to 32% in 2011. Regional variances are important to note here, with South Asian MFIs having the lowest yield on their portfolio which has remained steady at around 24%, whilst sub-Saharan African and Latin American MFIs consistently had a higher yield on their portfolio. The decrease in portfolio yield has come mostly in East Asia and the Pacific, Easter Europe and Central Asia, and the Middle East and North Africa.

The Evolution of Microfinance

13

True cost of borrowing is often higher Since their interest rates are high, some MFIs might not be entirely transparent about the real cost of borrowing to their clients and would not report effective annual rates unless required by law to do so (AttuelMendes and Ashta, 2013). Even then, the law may permit a lot of flexibility in how the MFI states its interest rates and how it structures the repayments from their clients. Rosenberg (2002) provides examples of eight different elementary pricing models and shows how an interest rate of 3% per month can have different impacts on the total cost of borrowing with minor changes in the repayment schedules, fees structure and interest rate calculation. He cites the following scenarios: 1. A base case with monthly repayments and declining base 2. Taking interest up front 3. Charging an initial fee 4. Taking weekly repayments 5. Taking a flat interest rate 6. Taking the flat interest rate up front 7. Taking compulsory savings before giving the loan 8. Flat up front interest rates and compulsory savings. Using these scenarios and difference mixes of the requirements, he calculates that the Annual Percentage Rate (APR) almost triples from 36% to 92%—depending on the mix of scenarios applied.

Costs are streamlined Like any enterprise, MFIs have a basic set of operational costs; we tend to categorize them in three major headings: financing costs, operating costs and loan loss provisions. Each of these has its own peculiarities. Financing costs for a MFI also vary based on its source of funding mix. It can opt to take low cost or concessional loans from development finance institutions. It could operate on a mix of equity financing and grant financing, rely entirely on donations, or it could borrow from the financial markets for on-lending. An efficient MFI balances all of these sources in order to have a sustainable weighted average cost of capital (WACC) whilst ensuring it remains competitive within its marketplace.

Chapter One

14

A MFI has standard operating costs such as rent, electricity, personnel costs and so on. However, the fixed infrastructure costs tend to be lower than commercial banks, as most MFIs operate from less than glamorous offices. In order to remain profitable, streamlining operating costs becomes one of the best mechanisms to reach financial sustainability (Khan and Ashta, 2012). Technology plays an important part here both in terms of reducing the cost of servicing clients as well as managing the administrative costs. Though traditionally a major challenge for MFIs because of the high cost, recent innovations in cloud computing hold out the promise of access to more affordable and high quality back office systems. Still, this market is very new and the risks need to be carefully managed (Barnett, 2011). For example, the bigger the MFI is, the more variety of personnel it hires, and the more it will require a system that can cater to a diversity of needs. The personnel costs are particularly difficult to manage because social security and payroll taxes as well as reporting requirements vary from country to country. Finally, the loan loss provisioning depends on the MFIs risk appetite, delinquency management, and decision to write off loans. This element, therefore, can also be streamlined. For example, by adjusting the policies on what stage of arrears is a provision made for loan loss and when is it actually written off. Cost Category

2003

2005

2007

2009

2011

Financing costs

2.73%

3.70%

4.21%

4.76%

4.94%

Operating costs

15.78%

14.14%

13.69%

13.78%

13.62%

Provision for loan loss

0.96%

1.00%

0.92%

1.20%

0.87%

Total

19.46%

18.84%

18.81%

19.74%

19.43%

Table 6: Cost breakdown for a median MFI, as a % of their Total Assets. Source: MIX Market, 2013. Table 6 outlines the various costs for a median MFI. The total cost as a percentage of total assets seems to remain fairly steady at around 19% over the period, indicating that MFIs have not become dramatically costefficient in their operations over the last decade. However, there are interesting inter-heading changes with operating costs decreasing (which may have reduced owing to improvement in efficiency, perhaps through economies of scale or the use of technology), and the cost of borrowing

The Evolution of Microfinance

15

has almost doubled (perhaps indicating that the sector is evolving from low cost / concessional sources of funding to more commercial sources of funding). Loan lost provisioning has slightly decreased, indicating that MFIs are became more adept in managing their risks and are becoming more resilient, particularly in their credit risk management (Khan and Ashta, 2013).

Profits are increased Although the cost of running a MFI is high, the high interest rates permit most MFIs to make a positive return on assets. With the effect of financial leverage, the Return on Equity (ROE) is much higher. 2003

2005

2007

2009

2011

5.90%

9.30%

10.80%

7.10%

8.30%

MFIs with ROE > 0%

67%

74%

76%

71%

78%

MFIs with ROE > 10%

43%

49%

52%

41%

45%

MFIs with ROE > 20%

22%

30%

30%

22%

23%

MFIs with ROE > 30%

12%

18%

16%

12%

12%

MFIs with ROE > 40%

8%

11%

10%

7%

6%

Median ROE

Table 7: Return on Equity for MFIs. Source: MIX Market, 2013. Table 7 provides statistics on Return on Equity (ROE) for MFIs in the 2003-2011 period. It shows that the median MFI earned about 8.3% ROE in 2011, a 41% increase since 2003. In fact in 2011, 78% of all MFIs reporting to MIX Market have a positive ROE! 45% of MFIs provide a return of more than 10% to shareholders, almost a quarter are providing an ROE of more than 20%, and a small percentage is providing Returns on Equity of more than 40%. It is this high profitability and the large number of sustainable institutions which is driving the growth of the sector.

Managing risks becomes crucial We have already indicated that the provision for loan impairment is a small proportion of total assets (about 1%, Table 6).The standard

Chapter One

16

indicator for monitoring credit risk in the microfinance portfolio is the Portfolio at Risk (PAR) ratio. This is the value of all loans outstanding that have one or more installments overdue over a certain number of days. This is usually expressed as a ratio of the Gross Loan Portfolio (GLP) to the overdue principal amount, thus PAR>30d is a ratio of all overdue more than 30 days to the GLP. 2003 Median PAR 30 days

2005

2007

2009

2011

3.40%

2.70%

2.90%

4.70%

4.00%

MFIs with PAR > 1%

71%

69%

71%

82%

79%

MFIs with PAR > 2%

60%

58%

59%

73%

68%

MFIs with PAR > 4%

46%

43%

42%

56%

49%

MFIs with PAR> 5%

40%

35%

34%

48%

42%

MFIs with PAR > 10%

20%

18%

16%

25%

21%

MFIs with PAR > 20%

8%

7%

5%

9%

9%

Table 8: Portfolio at Risk (PAR) greater than 30 days for MFIs. Source: MIX Market, 2013. Table 8 shows the PAR>30d for all MFIs, obtained from MIX. It is interesting to note that the median MFI in 2011 had a PAR>30day of about 4% which has not changed significantly since 2003. Although the PAR>30day peaked in 2009 and has gone down in 2011, there are still a lot of MFIs exposed to high levels of risk with PAR>30day more than 10%. This increase may be owing to saturation of some markets, multiple MFIs lending to the same borrower in these locations, as well as the effect of the economic slowdown which has manifested itself at all strata of the socio-economic pyramid in most countries. This increase in risk may lead to higher loss provisions and higher write-offs in future years. Coupled with the fact that the amount of provisioning for loan losses has been decreasing over the years, this might indicate a structural problem emerging in the industry where lending is becoming riskier, but MFIs are failing to put in adequate mechanisms to cushion against this risk.

The Evolution of Microfinance

17

Product Diversification is taking place Initially, microfinance was synonymous with microcredit. However, MFIs and researchers alike found that providing microcredit alone does not cater to the financial needs of the poor and like everyone else, the poor also require a number of complementary financial products. These financial services include savings, insurance, guarantees, remittances and equity for their businesses. The first step in the process was to provide savings products in countries where the legislation permits it—more than 25% of MFIs do not take deposits.

Number of small MFIs (< 10,000 depositors) % of small MFIs Number of medium MFIs (10,000 to 30,000 depositors) % of medium MFIs Number of large MFIs (> 30,000 depositors) % of large MFIs

1999

2002

2005

2008

2011

97

376

718

768

813

82.9%

83.2%

71.9%

66.4%

68.1%

8

38

144

160

133

6.8%

8.4%

14.4%

13.8%

11.1%

12

38

136

229

247

10.3%

8.4%

13.6%

19.8%

20.7%

Table 9: Number and categories of deposit taking MFIs. Source: MIX Market, 2013. As we can see from Table 9, in 2011 nearly 70% of all deposit taking MFIs had less than 10,000 depositors. The number of small deposit taking MFIs has decreased since 1999; however, the corresponding number of large deposit taking MFIs (with more than 30,000 depositors) have more than doubled in that timespan, clearly showing an increasing uptake of deposit products by clients. Similarly, other financial products such as micro-insurance and microremittances are also being adapted for delivery through the distribution network of MFIs.

18

Chapter One

Regulatory pressure is increasing The popular image of a typical microfinance institution (MFI) used to be of a non-profit NGO, or a cooperative funded by donors, reaching out to a segment of the poor to whom no banks would lend. However, as the scale increased and profitability proven, donors sought to withdraw and let the market mechanisms take over. So, there has been a shift from NGO to forprofit institutions entering the microfinance market. Depending on local regulations, they could be microfinance banks, non-bank financial institutions (NBFIs), savings and loans companies, and even venture capital companies. As the NGO institutions became larger, some even opted to become regulated financial institutions. As the sector proved to be profitable, some commercial banks decided to ‘down scale’ their operations in order to compete in this market—no doubt partly prompted by public policy of promoting inclusive finance. Cooperatives have always been important in some regions, such as West Africa, and rural banks have mostly been State development banks. In many countries, such as China, Laos, Vietnam and India, the lack of microfinance-specific legislation often results in the MFIs adopting inappropriate charter types. They adopt the best fit offered by the local regulatory regime which allows them to serve the poor whilst being able to carry out other functions such a mobilize deposits, or borrow capital from the open market. Table 10 provides an overview of the different charter types (legal forms) of MFIs in 2011. The table shows the different legal forms that MFIs operate under and is roughly indicative of the regulatory regime in each region. Together, NGOs and Cooperatives still account for 50% of MFIs. Overall, 34% of the MFIs are NGOs, with predominance in the MENA region where 67% of the MFIs are NGOs. Surprisingly, in subSaharan Africa only 21% of MFIs are NGOs, banks provide 20% of the microfinance in this region although overall, banks are only 10% of the financial service providers in microfinance. Rural Banks, accounting for 4% of the total are mostly present in East Asia and the Pacific. NBFIs are the second most important group of MFIs and constitute 30% of the total population. They are dominant players in East Europe and Central Asia, and Latin America where the loan size is largest. Credit unions or cooperatives are the third most prevalent form of MFIs, constituting 16% of the total, and 32% in sub-Saharan Africa.

The Evolution of Microfinance

19

Sub-Saharan Africa

21%

Credit Union / Coop. 32%

East Asia & Pacific

39%

10%

13%

3%

18%

3%

14%

11%

16%

54%

15%

0%

3%

1%

42%

15%

34%

7%

0%

1%

1%

67%

0%

16%

5%

0%

7%

5%

South Asia

41%

8%

28%

7%

5%

3%

8%

Overall

34%

16%

30%

10%

4%

2%

5%

Region

East Europe & Central Asia Latin America & Caribbean Middle East and North Africa

NGO MFI

Not Other specifi ed

NBFI

Bank

Rural Bank

24%

20%

1%

0%

1%

Table 10: Breakdown of MFIs by region and charter types in 2011. Source: MIX Market, 2013

The need for MIS For all growing MFIs, the challenges are high: overcoming information asymmetry, diversifying products, increasing client outreach, improving risk management, and supporting new regulatory constraints. Hence, "Gaining business advantage out of a MIS in the microfinance sector requires a unique ability to improve the MIS again and again, as business grows and evolves. This core requirement of continuous improvement changes the way we must envision IT investments, from a project-based, feature-centric and static approach, where the MIS is “finished” once the implementation project is over, to a product-based, team-centric, ever evolving approach, where the “MIS project” actually begins when endusers start using the MIS, and then never ends." (Pierre Pezziardi, private correspondence).

The next chapter provides an overview of the evolution of MIS in the banking and financial services sector in general. This will enable the reader of this book to situate the developments in MIS for microfinance.

20

Chapter One

Bibliography Armendàriz, Beatriz, and Jonathan Morduch. 2010. The Economics of Microfinance. Cambridge, MA: MIT Press. Ashta, Arvind. 2011. Advanced technologies for microfinance: Solutions and challenges. Hershey, PA: IGI Global. Attuel-Mendes, Laurence, and Arvind Ashta. 2013. "The Truth, But Not Always The Whole Truth In Lending Laws." Cost Management no. 27 (2):6-19. Barnett, Bryan. 2011. Cloud Computing and Financial Services for the Poor: Primer and Procurement Guide. USAID. Khan, Saleh, and Arvind Ashta. 2012. "Cost control in Microfinance: Lessons from the ASA case." Cost Management no. 26 (1):5-22. Khan, Saleh, and Arvind Ashta. 2013. "Managing Multi-Faceted Risks in Microfinance Operations." Strategic Change no. 22 (1/2):1-16. doi: 10.1002/jsc.1918. Milana, Carlo, and Arvind Ashta. 2012. "Developing Microfinance: A Survey of the Literature." Strategic Change: Briefings in Entrepreneurial Finance no. 21 (7-8):299–330. Rosenberg, Richard. 2002. Microcredit Interest Rates. In OccasionalPaper 1. Washington, D.C.: CGAP.

Note i

A formal academic literature review is provided by Milana and Ashta (2012).

CHAPTER TWO INTRODUCTION TO MIS FOR FINANCIAL SERVICES SUNDER ANNAMRAJU

Introduction Investment in information systems and technology by the financial services sector far outweighs any other industry segment. It is estimated that financial services firms will spend between US $270 billion and US $460 billion on Information Technology (IT) in 2013 globally1. Being early adopters of systems and technological innovation, financial institutions manage a mixed portfolio of mainframe-based legacy applications and technologically advanced sophisticated systems, whose current state caters to the challenges faced by the industry today. Microfinance Institutions (MFIs), as distinct providers of financial services to a defined target population, are a recent addition to the financial services sector. As relatively new entrants, they are well positioned to harness current technologies to avoid the mess of IT systems prevalent in banks. Also the MFI sector is not subject to the same set of regulations that govern traditional banks. This means that MFIs can implement simpler information systems capable of delivering results at significantly lower operating costs. Nevertheless, a study of the factors behind the evolution of information systems in banks is relevant for MFIs in order to manage their own future systems roadmap. The paper is organized as follows. We first look at the characteristics and key differences between traditional banks and microfinance institutions. Next, we look at the evolution and historical growth of some of the large banks today and simultaneously, the advances in, and the adoption of technology within banks. We will analyze the reasons behind the proliferation and mix of information systems used by the large banks

Chapter Two

22

and in conclusion, provide a set of recommendations for microfinance institutions embarking on their own information systems journeys. This paper serves as an introductory section and is meant for both types of audiences—those who are technologically aware, as well as those who are new to information systems and technology. It is based on the author’s experience of three decades in the field of banking systems and a review of research publications in the field of MIS in financial services.

Banks and Microfinance Institutions Rank 1 2 3 4 5 6 7 8 9 10

Bank

Country

Deutsche Bank HSBC BNP Paribas Industrial and Commercial Bank of China Mitsubishi UFJ Financial Group Crédit Agricole Barclays Group Royal Bank of Scotland JP Morgan Chase Bank of America

Germany United Kingdom France China

Total Assets ($m) 2,799,977 2,555,579 2,542,738 2,456,287

Japan

2,447,950

France United Kingdom United Kingdom United States United States

2,431,796 2,417,327 2,329,726 2,265,792 2,129,046

Table 1: World’s 10 Biggest Banks 2012 (Measured by Total Assets). Table 1 lists the world’s largest banks measured by their total assets as of August 2012 published by Global Finance2. A look into their operations, ownership and governance structures and range of products reveals a broad similarity. All these banks are publicly listed corporates, often on multiple stock exchanges. They have a diversified set of operations both at a product level and at the geographical level and are accountable to their shareholders for their performance. In a globally interconnected financial environment, they rely on their information systems to streamline their operations, reduce costs, and achieve competitive advantage by monitoring financial markets and being quick to respond to market movements. Almost all these banks exhibit a similar growth pattern. Originally founded to respond to a particular business or commercial requirement in their respective locations (usually to finance trade operations or agriculture), they grew their business by expanding and diversifying,

Introduction to MIS for Financial Services

23

adapting to the changing needs of the times—sometimes dictated by historical events. In order to survive in the face of competition, they sought to increase their range of products and services by either acquiring providers of new and niche services, or merging with other banks to broaden their base of operations and serve a wider client-group. This pattern of inorganic growth placed an enormous strain on their existing information systems and led to the implementation of new systems to cater to the requirements of the evolved organization. However, over the period of their existence, banks have developed a well-defined, formal organizational structure and standardized operational procedures. In relative terms, present-day microfinance is a new industry, although the underlying concept of creating a banking system to help the poor can be traced back to the 1700s. Over the years, the scope of microfinance has evolved beyond providing a loan to finance a microenterprise: “No longer limited to investing in microenterprises, microfinance now encompasses all financial services and how to provide them in a way that improves the quality of life of poor women and men.” (Ledgerwood & Gibson, 2013)

In 1976, Professor Muhammad Yunus, of the University of Chittagong, launched an action research project to examine the possibility of designing a credit delivery system to provide banking services targeted at the rural poor in Bangladesh. This experiment subsequently led to the establishment of Grameen Bank in 1983. In 2007, Forbes estimated that there were over 12,000 microfinance institutions issuing loans worldwide, with an ecosystem of lenders, donors, investment funds and technology service providers fast developing in this space. The MIX Market (www. themix.org), a source for microfinance performance data and analytics, lists over 120 technology service providers serving the microfinance industry. While these are still early days for the industry, the first signs of structural problems—such as concentrated market competition, overstretched systems and controls, and a relaxation of credit standards that are leading to deterioration in portfolio quality—are beginning to show, pointing to a consolidation in the sector. This will have implications for the information systems developed for use by these organizations. Quadri et al. (2011) have articulated some of the challenges faced by microfinance institutions in implementing effective and sustainable information systems. These challenges arise due to differences in the way microfinance institutions are organized vis-à-vis banks, characterized by a lack of standard processes and procedures and constraints imposed by poor infrastructure, lack of adequate maintenance and support for systems,

Chapter Two

24

and a lack of appropriate staff training on information systems and technology. They propose an MIS framework for microfinance institutions, drawing upon case studies where it has been successfully implemented. Table 2 lists some key characteristics and differences between banks and MFIs. Each parameter has an influence in the development of information systems to support operational, strategic, analytical and reporting requirements. Parameter

Banks

Microfinance Institutions

Structure & Ownership

“For profit” organizations. Publicly listed. Responsible to shareholders.

Processes and Procedures Policies

Standardized procedures structured and well-defined. Documented (enables staff to be trained.)

Staff and Training

Skilled staff. Undergo formal training on policies, procedures and systems.

Nature of Clients

Individuals (retail & high net worth), small & medium enterprises (SME), corporates, funds. Clients subject to a “KnowYour-Customer (KYC)” review. Financial records and other reference and transactional data retained. Net effect is a mitigation of financial risk.

Typically "Not-for-Profit" organizations, with some "For Profit" entities. Ownership: private sector entities, members, governments and in some cases with shareholders. MFIs primarily have a social objective coupled with financial sustainability. Unstructured, non-standard and evolving. Partially documented or undocumented policies and procedures. “Employees are predominantly not well educated” (Quadri et al., 2011). Staff training is often unstructured and ad-hoc. Poor and low-income people who do not have access to other formal financial institutions. No formal KYC process. Detailed customer information gathered, including socio-economic background and household details. Inherently, subject to higher financial risk.

Nature of Customer Data

Introduction to MIS for Financial Services

25

Parameter

Banks

Product Offering

Diverse set of products covering savings, loans, trusts, investment management, wealth & asset management, trade finance, foreign exchange, funds transfers, etc.

Small loans with little or no collateral. Product set could include savings, credit, insurance, money transfers and services provided by others but targeted at poor and low-income people.

Risk Management

Characterized by high values. Employ sophisticated risk-modeling techniques to manage risk.

Scale of Operations

Regulatory Environment

Large banks typically have global operations. Others could have country-level or regional operations. Characterized by good infrastructure availability. Excellent availability of technical support for information systems – typically have in-house IT department(s). Can adapt systems to changing business requirements. Strong level of regulatory oversight.

Characterized by low values. Risk of default is managed through social and behavioral practices of sanctions, rewards and access to future credit. Usually localized operations, typically in rural and remote areas. Characterized by poor infrastructure availability.

Influence of External Factors

Impacted by global events affecting financial markets.

Technology Support

Table 2: Comparison of Banks vs. MFIs.

Microfinance Institutions

Low level of support available for information systems and technology (due to sustainability issues). Cannot adapt systems easily due to change in requirements. Low level of regulation. May be licensed or supervised by a banking authority. Relatively insulated as a result of their funding pattern.

26

Chapter Two

Growth of Banks Large banks such as the ones appearing in Table 1 have expanded over a period of time to become global institutions with complex corporate structures, delivering a multitude of products through a variety of channels to a diverse client set. Technology has played a part in fuelling this growth, with historical events and legislation contributing to the shape and structure of these institutions. A literature survey of the history of the banking institutions shows some revealing trends. Until after World War II banks, or their predecessor institutions, experienced significant changes, periods of growth and expansion, as well as disruption and dislocation, due to economic and political factors. Engaged in financing trade, the development of industry and infrastructure, lending to government and enterprises, and reconstructing economies in post-war periods, banks grew by expanding into new geographies, forming alliances with other financial institutions and acquisitions. In the second half of the twentieth century, and continuing up to the present day, banks shifted gear, moving from being large regional players to becoming global players with distinct strategies, products (strengths), identities and branding. It is this period that is of interest to us since it corresponds to the time of considerable technological advancement. Box 1 presents an overview of how some of the top-ranking banks have described their own growth. It is interesting to note the specific reference to technology in some statements.

Introduction to MIS for Financial Services

27

“By the 1970s the bank had firmly developed a policy of expansion by acquisition or formation of subsidiaries with their own identities and expertise. HSBC continued to grow through a number of acquisitions across the globe. In November 1998, HSBC announced the adoption of a unified brand, using HSBC and the hexagon symbol everywhere it operated, with the aim of enhancing recognition of HSBC by customers, shareholders and staff throughout the world.” - Source: HSBC Website - History3 “International business grew far more important in the 1970s. Deutsche Bank began to take shape as a global group. New branches abroad supported this development. The evolution of financial markets, technological progress and the acquisition of major banks in Italy, Spain, the UK, and the United States have all meant that Deutsche Bank has changed more in the last decades than in the preceding century.” - Source: Deutsche Bank Website - History4 “Our leadership had a vision for a bank that could provide service beyond our front door: not only to the nation, but to countries across the globe. Over the past few decades, a series of acquisitions made that vision possible. As a result, Bank of America has increased the breadth of our products and services and made them more accessible to customers worldwide.” - Source: Bank of America Website – Our History and Heritage5 “ICBC provided a wide range of financial products and services to 4.11 million corporate clients and 282 million individual customers through 16,648 outlets across China, 239 overseas subsidiaries and a global network of more than 1,669 correspondent banks as well as Internet Banking, Telephone Banking and Self-service Banking. ICBC established strong presence by its commercial banking operation and rapid expansion to markets worldwide.” - Source: ICBC Website – About Us, Introduction6

Box 1: Statements by Banks on their Growth

Impact of growth We look at growth of banks in two ways: organic and inorganic. When a bank grows by expanding its customer base, its branch network to cover a wider geographical region, or by an enhanced product set, the systems required to support such organic growth are relatively easy to introduce into the bank’s environment. This is because the new systems are selected in conformity with the bank’s existing IT policy7. As technology advanced, the new systems embraced the benefits offered by the technology of their time. These new systems may or may not displace earlier systems already in place at the bank—this decision is usually based

28

Chapter Two

on an analysis of the costs, risk and benefits of a system replacement. Thus, banks grew with a co-existence of old and new systems in their application portfolio. Inorganic growth through mergers and acquisitions adds a different dimension to the systems landscape. In such a scenario, an integration of different technological platforms is often required. Where multiple systems exist for the same function, a decision is required on the system to be retained, or whether to retire both and implement a new one. At best, this exercise is cost-intensive; at worst, the systems are fragmented and the cobbling together of platforms, networks and systems frequently leads to incompatibility between them, causing operational errors and creating adverse business impact until stability is achieved. A side effect of such integration is the growth of silos in operations (and data) and a proliferation of reporting and analytical applications to reconcile data held in multiple systems. Inevitably, inorganic growth rapidly increases the diversity of systems and technology in the (new) bank’s portfolio. There is yet another dimension that adds to the silo effect noted above. Growth of banks during periods of good financial performance has seen a rise in the number of departmental systems introduced. Several of these are developed as algorithms on spreadsheets by highly skilled quantitative analysts (also known as “Quants”). These systems are usually justified on the basis of effectiveness such as the ability to respond to fast-changing market conditions, or enabling product innovation to meet client requirements. Often these systems create multiple data repositories without formal controls and as a result there is a lack of transparency, and thus the quality of data held by the bank deteriorates. This can have a crippling effect on banks, especially during periods of adverse market conditions.

Adoption of Technology Technology is a broad area that includes computer hardware, computer science (governing areas such as operating systems, programming languages, database technologies, software engineering and artificial intelligence), communication networks and related developments in diverse fields such as user interfaces, middleware, network and mobile technologies, the world-wide web (the “web”), virtualization, Service Oriented Architecture, video games and embedded systems. Several of these developments have progressed concurrently and impacted upon the systems used within banks today.

Introduction to MIS for Financial Services

29

The use of information technology and systems by banks is an active area of research and one that has received the attention of academics and practitioners alike. Gupta and Collins (1997) studied the impact of information systems on the efficiency of banks; Smith and McKeen (2008) report on how a financial services firm is leveraging IT to transform itself into a “process-centric” organization where applications are constructed according to service oriented architecture (SOA) principles. Davamaniranjan et al. (2006) explored the effects of IT at the individual process level within a bank and developed models to study how systems characteristics enhance the output(s) of a process, relating the process performance to the economic performance of the bank. Studies have also been made to determine the effectiveness and efficacy of aligning information systems strategy to the business strategy of the (financial services) firm. Ferguson, Jr. (2000) identified three main reasons that banks invest heavily in systems and technology. These are primarily to: (a) Achieve reduction in operational costs by streamlining processes and eliminating manual intervention, (b) Provide better customer experience by offering new products and services, or by enriching existing products, (c) Implement sophisticated risk management and analytics (business intelligence) techniques for exercising business control. Complementing these is a fourth compelling reason: (d) Be able to respond rapidly to changing market conditions and the regulatory environment (i.e., seek competitive advantage). The development of automated systems, technology and software for use by banks dates back to the 1950s. Bank of America asked SRI, a Stanford-based research institute, to conduct a feasibility study on an electronic bookkeeping machine to streamline their cheque handling process. The study resulted in two innovative developments: ‘Electronic Recording Machine, Accounting’ (ERMA) and the font for the ‘Magnetic Ink Character Recognition’ system (MICR) that appears on cheques even today. Working round the clock, the first ERMA computer processed 33,000 accounts per hour, 792,000 accounts in 24 hours and 5.5 million accounts per week. "ERMA was the absolute beginning of the mechanization of business." – Thomas Morrin, SRI Director of Engineering

30

Chapter Two

Automated Teller Machines (ATMs) using magnetically encoded plastic cards were first introduced in 1969 at Chemical Bank in New York, although the concept of a “hole-in-the-wall” machine that would enable customers to conduct financial transactions goes back to 1939. Citibank installed a predecessor to the ATM in 1960, called the Bankograph, which allowed customers to pay utility bills and obtain receipts. Barclays Bank in London installed the first cash dispensing machine in 1967. Over the years the implementation of systems in banks has paralleled the availability of new technology. Cash et al. (1992) demarcate the evolution of the IT environment into three eras. The period up to the early 1970s (Era I) saw the development of centralized systems on mainframe computers. The applications had an organization-wide focus, such as accounting and payroll. New applications were approved for development if they would either reduce or displace costs. There was a strict control on access to computing power in a central IT function. With the availability of minicomputers, personal computers and developments in network technology since the early 1970s (Era II), individual departments within banks had access, knowledge and resources to operate computers and develop software for their own use. Departmental and business effectiveness were considerations that drove the implementation of systems in banks minimizing the requirement to deal with a central IT function. Although a strict timeframe is not attributed to Era III, this marks the period when banks shifted from operational and tactical modes of employing technology to utilizing it to gain strategic and competitive advantage. The advances in technology supported such a shift from the operational to the strategic mode of usage. Not only has there been an increase in computing power packed into small physical devices, but also, developments in the area of programming languages, operating systems, software development methodologies, near-field communications technology (NFC) that facilitates electronic payments, networks, storage devices, data warehousing, user interfaces, security and encryption technologies, and concepts such as cloud computing (enabling the delivery of software as a service). ‘Big Data’ (referring to large data content in unstructured formats such as e-mail, texts, video and audio in addition to traditional databases) and ‘gamification’ (using the engaging medium of games for business purposes) all change the manner in which financial services are delivered, pressurizing banks to stay abreast of shifts in technology for their survival.

Introduction to MIS for Financial Services

31

Current Trends In this section we review some current innovative uses of technology in the domain of financial services and providers. The list is by no means exhaustive and only serves to highlight the capabilities of new technologies to bring about business changes and thereby, changes in the systems landscape. Mobile Payments using Near Sound Data Transfer (NSDT™) technology. This technology, patented by Tagattitude, encrypts transactional data into sounds that are transmitted securely between two mobile phones. The technology can be integrated into multiple channels such as ATMs, Point Of Sale terminals and others. It has a low deployment cost and does not require special hardware. Another technology enabling cashless payments is Near Field Communication (NFC), which is a standards-based connectivity technology that through touching permits electronic devices such as smartphones to exchange financial transactions and other digital content. Gamification – Banks are deploying games (called Serious Games) in their digital channels as an educational tool to inform their customers about products and services and enabling transactions within a game environment. Serious Games seek to use the game paradigm to address a business matter and they do this in very much the same manner as entertainment games through awarding points, virtual currencies, badges and publishing player rankings, etc. The game design is such that by playing, an underlying business issue gets addressed. Such games can be used for educational or training purposes, such as promoting sound financial management principles among the players, through which financial organizations can promote their products and services. Crowd Funding – uses social networking, usually over the Internet, to enable individuals to pool their (financial) resources to support the efforts of other individuals or organizations. An example of such a model in microfinance is Kiva, a non-profit organization that works with microfinance institutions on five continents to provide loans to people without access to traditional banking systems. Cloud Computing – The National Institute of Standards and Technology (NIST) defines Cloud computing as “a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction” (Mell and Grance, 2011). Among the essential characteristics of Cloud

32

Chapter Two

computing is an on-demand self-service capability without the need for human interaction with the service provider(s); a capability of rapidly increasing or decreasing resources based on need (elasticity); and a metered provisioning of computing services (pay for what you use). Cloud computing may be implemented as a Software-as-a-Service (SaaS), Platform-as-a-Service (PaaS) or as an Infrastructure-as-a-Service (IaaS) model. It enables banks to shift resources from maintaining existing systems, and through using the flexible, on-demand service model offered by the cloud, invest in building innovative services and driving business growth. This model is of particular relevance to microfinance institutions as it offers the potential to keep operating costs low. Big Data – traditionally, banks store customer and transactional data in relational databases, subsequently processing these for generating operational and management information reports used for making business decisions. There is yet another vast data set that does not get stored in such traditional systems. These data sets are contained within e-mail messages, weblogs (blogs), chats, photographs, audio / video, corporate documents and social media content, which can be mined using special tools to yield useful information. This is the concept behind ‘Big Data’ and is characterized by the 4 V’s of data: Volume, Velocity, Variety and Value. Large banks certainly have all four, and typically process anywhere between 50 Terabytes and 1 Petabyte of data daily, with unstructured data accounting for an increasing share of this volume in recent years. “Big Data provides opportunities for business users to ask questions they never were able to ask before. How can a financial organization find better ways to detect fraud? How can an insurance company gain a deeper insight into its customers to see who may be the least economical to insure? How does a software company find its most at-risk customers—those who are about to deploy a competitive product? They need to integrate Big Data techniques with their current enterprise data to gain that competitive advantage.” (McKendrick, 2012).

Data in banks tend to end up in different silos making it difficult for decision makers to access them and discover hidden information contained therein. The challenge is to integrate data that is rapidly generated from a variety of sources and process and store the data into highly available storage, using that data in analytical and visualization applications. One such solution is Hadoop, an Apache project that brings large quantities of unstructured data into a manageable file system that can be used by analytical applications across the organization.

Introduction to MIS for Financial Services

33

Changes in the Systems Landscape

Figure 1: A High Level Model of Processing Systems in Banks.

The areas of investment in information systems by financial service providers to support business are best visualized by considering a model of processing systems in banks. A high-level model is presented in Figure 1. Each layer—client interaction systems, front-office systems, middle-office processing systems, back-office processing systems, emergent systems— comprises of a set of systems which could be distributed across business units, as well as across different locations. The choice of systems and

Chapter Two

34

technology at each layer, and even among systems in a given layer, depends largely on the benefits that will accrue as a result of the investment made.

Impact of Technology To understand how technology has influenced the emergence of systems in banks, we refer to the diagram in Figure 1 and note that: -

-

-

Systems at the periphery and having an external (to the bank) interface are most likely to be using current technologies. Increasing the network security, enriching the customer experience by providing mobile applications and online engagement, reducing costs by providing self-service channels to the customer, responding to market movements (using specialized algorithms) and enabling product innovation fall in this category. Systems that consolidate data from diverse sources for analytical purposes, business intelligence, and the computation and modeling of risk also have a high probability of using current (or nearcurrent) technologies. These are shown in the model as emergent systems. Internal and back-office processing systems such as accounting, payroll and other human resources systems are least likely to use new technologies and will continue on legacy platforms unless there is a threat of technological obsolescence or incapability to meet current business requirements. Replacement of these systems is cost intensive and fraught with risks of disturbing existing operational efficiency and efficacy.

Conclusions We have looked at the factors resulting in the current systems landscape in large banks by looking at growth and technological aspects. Both inorganic growth and advances in technology have shown to be disruptive factors, causing great diversity and complexity in the systems portfolio of banks and transforming the way in which financial services are delivered. Relatively speaking the microfinance industry is at a starting point in its systems’ journey. Riggins and Weber (2013) say that technology is not an area of strength for MFIs and suggest that they outsource internal operations and IT capabilities to emerging service providers. They also

Introduction to MIS for Financial Services

35

predict that MFIs will face competition from mobile service providers and commercial banks, which are better equipped with access to public funds, technology and staff. A survey conducted by MIX into the burden of reporting by MFIs in Africa shows that over 40 external reports are prepared requiring an effort of 60 man-days. Further analysis concludes that technology can help reduce the reporting effort for MFIs by offering a common format that incorporates data required by diverse stakeholders8. In addition to these factors, we see evidence of consolidation in the microfinance sector to achieve financial stability. Banks have trodden this path before, but they were structured differently, and had access to deposits from the public which the MFIs do not have. Unlike banks, investment in technology is not a route that MFIs can take to reduce their operating costs. They can, however, purchase the necessary services at a fraction of the cost of owning the asset. Cloudbased service providers of software and infrastructure are examples. However, MFIs should take care to ensure that they remain the owners of their data. This is their only asset, and important to maintain during system upgrades and consolidation within the industry. Use of mobile and near field communication technology for effecting payments is an established method in the microfinance sector, helping to keep the operational cost low. We need to realize though that technology will change and systems implemented today will become ‘legacy’ platforms in the future, and microfinance institutions, like banks, will eventually end up with a mixed portfolio of legacy and technologically advanced systems.

Additional Reading Gartner Worldwide IT Spending Forecast. Accessed April 3, 2013. http://www.gartner.com/technology/research/it-spending-forecast Rural and Microfinance Institutions: Regulatory and Supervisory Issues http://www.imf.org/external/pubs/ft/fsa/eng/pdf/ch07.pdf Computer History Museum. Accessed April 9, 2013 http://www.computerhistory.org/timeline/ Tagattitude. Accessed April 22, 2013. http://www.tagattitude.fr/en/applications/microfinance 5 Banks Leading the Way in Gamification. Accessed April 22, 2013. http://www.banktech.com/business-intelligence/5-banks-leading-theway-in-gamification/240148098

36

Chapter Two

Bibliography Cash, Jr., James I., McFarlan, Warren F., McKenney, James L., & Applegate, Lynda M., “Corporate Information Systems Management: Text and Cases” (Boston, MA: Irwin, 1992), 10-11. Davamaniranjan, Prabhu, Kaufmann, Robert J., Kriebel, Charles H., & Mukhopadhyay, Tridas, “Systems Design, Process Performance, and Economic Outcomes in International Banking”, Journal of Management Information Systems, Fall 2006, Vol. 23 Issue 2, 65-90. Ferguson Jr., Roger W.: Information Technology in Banking and Supervision, October 20, 2000. http://www.bis.org/review/r001023a.pdf Gupta, Uma G., & Collins, William, "The impact of information systems on the efficiency of banks: an empirical investigation", Industrial Management & Data Systems, Vol. 97 Iss: 1 (1997): 10 – 16, doi: 10.1108/02635579710161296 Joseph McKendrick: “Big Data, Big Challenges, Big Opportunities: 2012 IOUG Big Data Strategies Survey.” Sponsored by Oracle. September 2012. http://www.oracle.com/us/corporate/analystreports/infrastructure/iougbig-data-survey-1912835.pdf Ledgerwood, Joanna, & Gibson, Alan: “The Evolving Financial Landscape.” 2013. In “The New Microfinance Handbook: A Financial Market System Perspective”: Edited by Joanna Ledgerwood with Julie Earne and Candace Nelson, 15 – 48. Mell, Peter, & Grance, Timothy: “The NIST Definition of Cloud Computing.” NIST Special Publication 800-145. September 2011. http://csrc.nist.gov/publications/nistpubs/800-145/SP800-145.pdf Quadri, S. Mohd. Najmullah, Vikas Kumar Singh and Kishen Parthasarathy Iyengar. 2011. "IT and MIS in Microfinance Institution: Effectiveness and Sustainability issues." In Advanced Technologies for Microfinance: Solutions and Challenges, edited by Arvind Ashta. Rankin, David, “Simplicity and transparency: Returning from the land of the bland”, Journal of Financial Services Marketing, Vol. 9 (2004): 2 172 – 178 Riggins, Frederick J., & Weber, David M., “The Impact of ICT on Intermediation in the Microfinance Industry”, 2013 46th Hawaii International Conference on System Sciences, 4246-4255, doi:10.1109/HICSS.2013.521

Introduction to MIS for Financial Services

37

Smith, Heather A., McKeen, & James D., “Creating a Process-Centric Organization at FCC: SOA from the Top Down”, MIS Quarterly Executive, 2008, Vol. 7 Issue 2, 71- 84.

Notes 1

Deutsche Bank Research. “IT in Banks: What does it cost?” December 20, 2012. http://www.dbresearch.com/PROD/DBR_INTERNET_ENPROD/PROD0000000000299039/IT+in+banks%3A++What+does+it+cost%3F+H igh+IT+costs+call+for+an+eye+on+efficiency.pdf 2 World’s 50 Biggest Banks 2012, Accessed November 21, 2012 http://www.gfmag.com/tools/best-banks/11986-worlds-50-biggest-banks2012.html 3 HSBC’s History. Accessed April 6, 2013. http://www.hsbc.com/about-hsbc/ history/hsbc-s-history 4 Deutsche Bank History: “Chronicle – from 1870 until Today”. Accessed April 5, 2013. https://www.db.com/en/media/DB_geschichte_meilensteine_120dpi_en.pdf 5 Bank of America: “Our History and Heritage”. Accessed April 7, 2013. http://about.bankofamerica.com/en-us/our-story/our-history-andheritage.html#fbid=aQH3txc-WGb 6 ICBC. Accessed April 6, 2013. http://www.icbc-ltd.com/ICBCLtd/About%20Us/Introduction/ 7 Though we make this statement here, the concept of a ‘centralised’ IT policy in banks and the manner of its implementation has undergone a change with the advancement of time and technology. 8 Information Overload: can technology address MFIs' reporting burden? April 2012. http://www.themix.org/publications/microbanking-bulletin/2012/04/technologyaddress-MFI-reporting-burden

PART ONE INFORMATION REQUIREMENTS FOR MICROFINANCE GLÒRIA ESTAPÉ-DUBREUIL

To begin with, I would like to point out the complementary roles of microfinance and information and communication technology (ICT) in economic development and the reduction of poverty in developing countries. The overwhelming number of articles studying microfinance procedures, institutions, financial and social results, especially those published in the last ten years, are eloquent testimony to the interest in this sector and the role of the Microfinance Institutions (MFIs) as drivers of key development initiatives in the financial sector (Armendariz & Morduch, 2005). As for ICT, there is growing evidence that information technology coupled with knowledge management holds much potential for propelling the development process (Okpaku, 2003; Kamel, 2005). According to the Word Bank (2003) report on ICT and the Millennium Development Goals, ICT reduces transaction costs per customer and enables financial institutions to provide small loans and services to a large number of rural customers. Compared to the dynamics showed by the microfinance sector in promoting innovation and embracing new challenges in fulfilling its social mission, adoption of ICT has lagged behind (Kauffman & Riggins, 2012). Although MFIs face considerable internal information processing, storing and sharing challenges, they have been reluctant to adopt and invest in ICT and introduce comprehensive Management Information System (MIS). Mia (2005) suggests that the main reason for introducing ICT in MFIs in Bangladesh has been operational efficiency rather than to improve their capability to reach and better serve a broader customer segment. However, there are a number of different and complementary reasons to take into account. Together with reasonable fears for the digital divide barriers, there is the perspective that limited funds could be better used by providing loans directly to the needy (Kauffmann & Riggins, 2012).

40

Part One

Furthermore, one of the biggest challenges facing MFIs, particularly in rural areas of the developing world, is implementing an Information System (IS) that can interface with a large number of clients across a region with unreliable physical infrastructure (Tapan, 2006). The costs faced by such implementation are not particularly welcomed by the external donors supplying funds to the microfinance industry, who would rather support independent impact assessments (IA) of the benefits of their funds. Some voices have been raised against such practices: ‘Building internal impact monitoring capacities (...) means helping the MFI develop its MIS (...) to collect readily available data (outreach, repayments, dropout rates, etc.) (...) Strengthening these systems and occasionally verifying them—rather than financing complex impact assessments by visiting consultants—is probably the best way to achieve the ''improving'' goals of IA.’ (Hulme, 2000)

The introduction of ICT in organizations has also exposed concerns regarding cultural aspects. A software package may “impose its own logic” conveying culturally inappropriate imposition of IS and homogeneity as the norm, instead of understanding and valuing locally meaningful ways of doing things (Bada, 2002; Walsam & Sahay, 2006). It is therefore imperative that end users are able to appropriate the technology. Wherever the MFIs are capable to state clearly their core issues, and these put forward the client’s key problems in their given context, the technology is able to achieve development impact (De’ & Ratan, 2009). The three chapters in this part discuss several aspects related to the requirements and use of MIS at the MFI level. To some extent, all the chapters deal with some of the most relevant research questions proposed by Kauffmann & Riggins (2012) on how ICT impacts upon the microfinance industry. Concerning the adoption of MFIs practices, the balance between the outsourcing of their ICT needs and the development of skilled MIS intellectual property is one of the topics drawing a lot of attention. The different considerations in the three chapters of this first part of the book certainly make some interesting contributions to this question. In chapter Three Frances Sinha, Rupal Patel and Nitin Madan cleverly discuss the main generic requirements that a MIS needs to fulfill to qualify for useful reporting of social data in microfinance. The authors first ponder the importance and key elements of Social Performance Management, as well as the relevance of the MIS as a facilitator in the measurement of the MFI’s social goals. The chapter offers both useful and practical insights as to the basic requirements of the MIS. First, a simple but often ignored

Information Requirements for Microfinance

41

requirement: when the aim is not only tracking loan repayments, the MIS should assign a unique ID to each client! Next, the chapter summarizes other desirable features, paying special attention to the integration of clients’ key indicators and characteristics (usually available to the MFI but often not captured) needed to profile client outreach and measure poverty and progress out of poverty. The section discussing poverty profiling is particularly enlightening, since it includes a list of key points ICT and MIS providers should take into account in implementing the Progress out of Poverty Index (PPI). Other critical features of a MIS are also considered, related to its usefulness in the monitoring of the MFI’s clients and particularly to the MFI responsiveness to the clients’ needs. Flexibility is one of the major features discussed throughout the chapter: allowing changes in product design and loan repayment structures; allowing interface with third party providers; or flexibility regarding portfolio reports, to allow for customized and segmented reports that facilitate useful data for the appropriate management decision makers. Chapter Four also dwells on the information requirements of an institutional MIS for the reporting of social data. Glòria Estapé-Dubreuil, Consol Torreguitart-Mirada and M.Rosa Rovira-Val take, nevertheless, a different and complementary approach, considering the disclosure of social performance in a more comprehensive way by providing information on the triple bottom line (TBL) of the microfinance institutions. The authors selected the most widely recognized as guidelines for TBL disclosure, developed by the Global Reporting Initiative (GRI), and studied the specific requirements that an institution’s MIS must fulfill to implement the GRI. First, the chapter revises the actual reporting recommendations and practices in the microfinance industry, both related to financial reporting and to social and, up to a point, environmental reporting. Next, it outlines the GRI reporting framework, with the three levels of reporting settled to enable a progressive adaptation to any organization wishing to adopt TBL reporting. Finally, the authors discuss whether the level of disclosure of a MFI following the microfinance industry standards is enough to fulfill international standards on corporate reporting. They conclude that, per se, there is no obvious need of a specific customization of the MIS software for a MFI to be able to report on TBL, at least in the two first application levels (C and B). They also encourage TBL reporting adoption, being “a very useful exercise in transparency and accountability to the stakeholders of the MFI.” In Chapter five, Djamchid Assadi, Sharam Alijani and Satchidananda Sogala consider the design and operational characteristics, both of the network infrastructure and the MIS suitable for MFIs. Their summary of

42

Part One

strategic options and operational characteristics that the MFIs should address is worth considering. They further illustrate their point by studying the pilot implementation in rural India of a customer intelligence system to collect data (personal details as well as income and other assets details) from customers using a variety of ICT devices and networks. The system further allows documentary evidence, as well as interviews storage and processing, with the aim of delivering credit rating lists and other pertinent information in a cost efficient digital form. The authors argue that such use of digital IS for customer intelligence and relationship management constitute a major step in reducing the cost for the MFIs of building robust MIS, and consequently the cost of transaction in both rural and urban areas of developing countries.

References Armendariz, B. and Morduch, J. (2005). The Economics of Microfinance. MIT Press, Cambrigde, MA. Bada, A.O. (2002). Local adaptation to global trends: A study of an ITbased organizational change program in a Nigerian bank. The Information Society, 18(2), 77–86. De’, R. and Ratan, A. L. (2009). Whose Gain Is It Anyway? Structurational Perspectives on Deploying ICTs for Development in India’s Microfinance Sector. Information Technology for Development 15(4), pp. 259-282. Hulme, D. (2000). Impact Assessment Methodologies for Microfinance: Theory, Experience and Better Practice. World Development, 28(1), pp. 79-98. Kamel, S. (2005). The use of Information Technology to Transform the Banking Sector in Developing Nations. Information Technology for Development 11(4), pp. 305-312. Kauffman, R.J. and Riggind, F.J. (2012). Information and communication technology and the sustainability of microfinance. Electronic Commerce Research and Applications 11, pp. 450-468. Mia, B. (2005). ICT in microfinance: a Bangladesh perspective. In S.Mathison (ed.) Electronic Banking with the Poor: Increasing the Outreach and Sustainability of Microfinance through ICT Innovation , Brisbane: Foundation for Development Cooperation. Okpaku, J.O. (2003). Information and communication technologies for African development. An assessment of progress and challenges ahead. United Nations Information and Communication Technologies Task Force (Eds.), Series 2. New York: United Nations.

Information Requirements for Microfinance

43

Tapan Parikh, P.J. and Sasikumar, K. – Ghosh, K. (2006). Mobile Phones and Paper Documents: Evaluating A New Approach for Capturing Microfinance Data in Rural India. Proc. CHI 2006, ACM Walsam, G. and Sahay, S. (2006). Research on Information Systems in Developing Countries: Current Landscape and Future Prospects. Information Technology for Development 12(1), pp. 7-24. World Bank (2003). ICT and MDGs – A World Bank Group perspective, Washington, D.C.: World Bank Group.

CHAPTER THREE MIS AS A POTENTIAL CATALYST FOR SOCIAL PERFORMANCE MANAGEMENT FRANCES SINHA, RUPAL PATEL AND NITIN MADAN1

Introduction This chapter seeks to address how social performance data collection and reporting, in particular client level data, can be integrated into the software for microfinance reporting. A practitioner paper, it draws on the ongoing experience of EDA Rural Systems, and the MFIs and MIS providers that they work with, to implement social performance reporting alongside financial performance reporting. The requirements and indicators for social performance reporting are derived from the underlying framework developed globally for social performance management (SPM). The following sections, therefore, reflect the groundwork and key initiatives around social performance in microfinance that have taken place in the past few years. These initiatives include: the Imp-Act consortium, in setting out the parameters of SPM2 (Campion et al., 2008); the work of Cerise, a microfinance network that has developed a social audit tool for MFIs3; the work of the microfinance specialist rating agencies in developing social ratings4; Microfinance Information Exchange featuring social performance data reported by MFIs alongside financial data5; and the Grameen Foundation in developing the Progress out of Poverty Index6. EDA is a part of these initiatives. In the course of our work to support SPM with MFIs, MIS has emerged as one of the key roadblocks to effective reporting of social performance. As social data managed by the MIS is critical for effective monitoring and benchmarking of social goals and values in microfinance, MIS systems, if properly implemented, can be a catalytic vehicle to support double bottom line reporting and

MIS as a Potential Catalyst for Social Performance Management

45

management. Specifically, there are two key concerns: 1) capturing, analyzing, and reporting of client level data; and 2) harmonizing software architecture in line with SPM requirements. As a precursor for IT to facilitate SPM, MFIs and IT providers require an understanding of social performance data and its intended use in order to design appropriate software and relevant reporting formats. Currently, MFIs who are interested in adapting their MIS systems are independently engaging with IT providers. We believe that these issues can be seen as somewhat generic, and it will be useful for both MFIs and IT providers to have a systematic reference for designing software and reporting processes now required by investors and other stakeholders. This chapter reviews the relevance and key elements of SPM for microfinance, then discusses the MIS and IT requirements to support a client data reporting system, with reference to the conceptual framework of three social goals of outreach, appropriate services and client progress. The discussion includes some specific details of the Progress out of Poverty Index as a relevant tool to benchmark poverty, both at entry and over time. In the final sections we reflect on effective reporting to enable a feedback loop of client level information into Board and management decision making.

The relevance of SPM Microfinance, which began with the aim of facilitating poverty reduction and women’s empowerment, now faces issues of reputational risk and mission drift. MFIs are under scrutiny, particularly in countries where there has been rapid growth of the sector. Recent publications have highlighted these concerns (for example, Chen et al., 2010; Sinha, 2011)7. The experience has brought to the fore questions on how MFIs, whilst being financially sustainable, can be more clearly client centric, provide services in a responsible way, and provide value to the low income, poor, and often women clients, who are the target markets for most microfinance. In terms of risk, the catchily-titled Microfinance Banana Skins Report identifies the risks of mission drift, client over-indebtedness and [lack of] client management among the top 11 perceived risks in the sector (Lacelles and Mendelson, 2012). SPM is an institutional process that directly addresses these questions. It represents a strategic approach to support the ‘effective translation of an organization’s mission in line with accepted social values.’8 This is the definition of social performance developed in the microfinance industry. The dissemination of this management concept along with practical

46

Chapter Three

guidelines and training resources is coordinated at the global level by the Social Performance Task Force (SPTF), a network of more than 1,300 members from all over the world, representing every stakeholder group committed to promoting the double bottom line in microfinance. The SPTF has defined standards for SPM (Universal Standards of Social Performance Management, USSPM). The Standards are organized in six sections as follows: Section 1: Define and Monitor Social Goals, Section 2: Ensure Board, Management and Employee Commitment to Social Goals, Section 3: Treat Clients Responsibly, Section 4: Design products, services, delivery models and channels that meet clients needs and preferences, Section 5: Treat Employees Responsibly, Section 6: Balance financial and social performance. Implementation of these Standards requires, among other management capabilities, the collection and use of relevant information. Selecting this is a normal part of business intelligence for any organization. As Dean Spitzer (2007) states in his recent book, an organization cannot improve its performance without measuring it. He further adds that if the measurement indicators are incorrect, then an entity will likely move away from its mission and strategic goals. For SPM the ‘right things to measure’ include indicators that capture client level issues and results. The SPM literature categorizes client level results in terms of outreach (reaching target clients), quality of products and services (meeting target clients' needs), and benefits for clients (progress). As a part of the ‘correct’ measurement, performance management assumes an inherent and continuous process of data collection, analysis and reporting. The social performance pathway, Figure 1, demonstrates how social data for client level results can be part of the information that feeds into monitoring, and aligning institutional strategy in adherence to an organization’s mission or intent.

MIS S as a Potential Catalyst for So ocial Performannce Managemen nt

47

Figure 1: Soccial Performance Pathway. Sou urce: Social Perrformance Task k Force, www.sptf.info fo.

Evidencee of achievinng an organizzation’s misssion and adheerence to accepted soccial values is a critical com mponent for m microfinance today t and will help to address somee of the reputaation risks thaat have emerg ged in the sector as weell as providinng the inform mation to whicch an MFI can n respond to improve iits value for cllients. Client leevel data captture and mark ket research aare in fact gen neric and common business activitiies, not only fo or the microfiinance sector, but in all sectors. Clieent level dataa, which assissts an organizzation in know wing and understandinng its client base, b makes business b sensee and helps to o address evident riskss. The challenge and thhe costs lie in n adapting thee reporting sy ystem and skills, whichh are convenntionally focused on creditt account tran nsactions. Though the cost may seeem rather hiigh or intangiible at the ou utset, the ability to haave ‘social’ accounts a along gside ‘financiial accounts’ is i critical for a doublee bottom line sector. s As SPM M is a deliberrate and inten ntional processs, this also applies a to planning andd budgeting forr the MIS. A key k requiremennt for implemeentation is assessing annd planning prroactively, ratther than haviing to react to o external scrutiny or reporting requests. The benefits off incorporatin ng social performancee in ensuring value for clieents and mitiggating reputattion risks justifies the requisite invvestment in orrganizational systems, inclu uding IT. The investm ment also contrributes to glob bal and nationnal standardizeed double bottom line reporting.

Soocial goals framework f k rting point of SPM is to deefine social gooals. Prior to initiating The start changes in tthe MIS, an MFI M needs to thoroughly deefine their soccial goals (target outreach, appropriate servicess, intended bbenefit to clieents) and define the inndicators that will enable th hem to measurre achievemen nt of their social goals.. In relation too the three soccial goals, the questions are:

48

Chapter Three

1. Outreach – are you reaching the clients you intend to reach? 2. Appropriate products and services – are you meeting client needs? 3. Progress – what are the benefits that you aim for in the lives of clients and their households, and are you contributing to such benefits? Traditional research methods such as sample surveys, focus group discussions and client case studies, can be applied to monitor these goals. Additionally, SPM looks at ways to monitor social goals across the microfinance portfolio, as part of routine capture and analysis of data about all its clients. These include profiling data at entry, accessing different products, transaction details and client retention, as well as any changes in profile indicators over time. Outreach profiling is essential, not only to monitor details of clients but also to enable market segmentation. In turn this is key to providing the basis for product design and analysis, as well as the analysis of potential changes over time (which in turn depend on a client’s starting point as well as their use of services). In this way, the information for all three social goals is interlinked with measurement of each depending on measurement of the previous goal. For this, the portfolio MIS is a vital and ongoing source of client level information that can be used for regular SPM reports to management and to the Board, alongside financial and other organizational reports. These can usefully be supplemented by periodic research (focused surveys, interviews) to follow up on issues that arise from analysis of the MIS data. Thus, indicators and targets for internal management can be effectively linked to these three social goals, and integrated with IT software to develop reporting on client level outputs and outcomes. Box 1 outlines various information components to track social goals, identifying the data that can be tracked through the portfolio MIS, as part of a client data reporting system. In the rest of this chapter, we focus on the IT requirements to operationalize client level reporting in relation to the social goals framework. We start with the cornerstone—having a unique client ID.

MIS as a Potential Catalyst for Social Performance Management Data sources

MIS client data

Examples of other methods, research

Outreach– target clients -Area – rural/urban -Client profile: key indicators for individual and household

Social Goals Appropriate products/services -Access to products -Transaction analysis -Client retention/exit

Client feedback/ satisfaction surveys. Client exit surveys/interviews. Case study interviews

49

Client progress - Client retention and transaction analysis at 5 years plus - Change in entry level indicators Graduation to formal banking - Client survey, focus groups, case studies

Box 1: Information to track social goals.

Cornerstone of reporting: Keeping track of your clients The first MIS requirement is to assign a unique ID to each client. This is the foundation of a well-designed software system as it enables MFIs to ‘keep clients in the system’, maintain historical data, link social indicators to financial data, measure client retention, and track indicators of clients over time. Unfortunately, as in banking, software systems in microfinance have too often been designed to maintain data on accounts rather than clients. One result of this is that once a particular loan is repaid, that account is closed and the ID disappears from the system. With this type of system, it is neither easy nor always possible to analyze a client’s financial history or current access to multiple products. MFIs and IT providers are now increasingly aware of the issues of having a unique ID. Initially a number of issues within the software need to be verified by the IT provider, as well as operationally by the MFI. The client ID should remain the same, even if a client moves to the next loan cycle or to another credit product, takes multiple financial products, moves to another center or branch, or returns to borrow from the MFI after a hiatus.

50

Chapter Three

As a secondary precaution, the software should also provide a check either through an exception report or a validation during input which would remove any duplicate client IDs.

Outreach profiling The next step is to begin integration of key indicators and/or tools for data collection. Profiling of client outreach, beyond the numbers of clients, can be integrated as part of routine operations. Often the type of data is already part of an existing format (membership form, loan application); however, this data is not usually captured and reported. Profiling of clients is seen as key to monitoring outreach in line with outreach targets, to profiling market segments, and to analyzing market intelligence. It can also be used as a baseline for clients at entry to track socio-economic changes over time. Key indicators to profile client outreach include those related to geographical area (rural/urban, less developed regions/provinces); client characteristics (e.g. gender, age, occupation, marital status), household characteristics (e.g. main livelihood, number of income sources, total family members, earning family members, income/economic level, household assets, children in school), financial access (e.g. access to other MFIs, other credit, savings) as well as details of a financed enterprise (e.g. economic sector, whether start-up, women client’s involvement, employment of non-family workers). These indicators reflect those that are relevant to stakeholders, for external reporting. They should be adapted by the MFI to ensure relevance to local context. This is data that an MFI often has as part of its different forms (membership form, loan application form, loan utilization form). The catch is that though the data is used for loan appraisal or checking, it may not always be automatically captured and recorded, and is therefore not available for routine reporting and analysis. If captured and reported, these profile indicators provide rich information about market outreach and segmentation. It can also be used by the MFI as a basis for tracking progress over time. The experience of MFIs in capturing this data highlights a number of features. Firstly, questions must be selected on the basis of their intended use. Questions that will not be used should be removed. These questions should be framed in the clearest possible language. They need to be closeended, with clearly defined options. Answer choices should be designed with consideration of how the results can be codified and analyzed. Finally, these questions can be integrated into a comprehensive format. An example of this is provided in the Annex. The example includes ten

MIS as a Potential Catalyst for Social Performance Management

51

indicators used in the Progress out of Poverty Index (PPI) to measure household poverty. This tool is discussed separately below. In terms of the MIS, these indicators can be captured in designated fields. Many systems already have such fields for additional indicators or they can be added for a low cost. However, the IT provider needs to conduct a number of quite basic checks. These include: assessing limitations in the software; checking the order in which indicators are to be inputted matches the order in which the data is collected in the client forms (for ease of data entry); appropriate designation of fields for numeric responses; having the default answer as a blank or read as ‘Select’; ensuring data validation checks is integrated so that staff must input data for all required questions; and ensuring that the software maintains historical data. Experience has shown that these issues, rather than an IT limitation, can be resolved by clear planning and consistent and appropriate management review by the MFI. They do though require careful attention to detail by the IT provider and the MFI working as a team to ensure collection and collation of meaningful data.

Poverty profiling Given the quite widespread intent of microfinance to serve low income and poor people, there has always been interest in measuring the poverty of clients, both when they start borrowing and longitudinally after several loan cycles. Some MFIs have developed their own indicators or poverty score cards.9 A recent development in the microfinance sector is the application of tools not only to measure economic poverty of clients but also benchmark poverty with reference to national and international poverty lines. One such tool is the Progress out of Poverty Index (PPI) developed for 45 countries (as of November 2012) by Mark Schreiner with the support of the Ford and Grameen Foundations.10 We note that the PPI is really a Poverty Index (PI) to measure poverty. When it is used to capture change over time then it may indeed serve as a PPI. In this section we discuss the PI as a means to capture poverty outreach. The (P)PI is a country-specific index based on the country’s national household income and expenditure survey. It is made up of ten questions that are designed to be easy to understand and answer. (Much easier, that is, than the task of capturing household information on income or expenditure for typical microfinance clients working in informal enterprises or seasonal agriculture related activities). Indicators are chosen for their statistical strength in predicting poverty and their potential to change.

Chapter Three

52

These indicators match quite closely the type of indicators that MFIs may already be interested in capturing as part of profiling a client household (such as number of household members, quality of housing, ownership of assets) and can be integrated into a client form and collected relatively easily. The tool is very useful as a means of benchmarking poverty outreach, assessing the extent to which the MFI is actually serving the poorer 30% or 50% of a population. The index should be supplemented with key household profile indicators that are significant in the local context. Prior to implementation of the PI, three foundational questions need to be addressed: from whom will the data be collected, for which products, and how frequently? PI data can be collected from all clients, routinely with every loan disbursement; or it may be collected for all clients at entry and exit to track changes in selected loan cycles. If collected for all loan disbursements, it is important to identify the ‘primary’ products for which the PI is collected, omitting any additional, short-term loan products (such as emergency or festival loans) so as not to collect multiple poverty scores for the same client. Finally, as (P)PIs are updated over time, to adapt to new national survey data, the MIS software will require an option that allows for activation/deactivation of different (P)PI versions and allow the opportunity to define for which products and loan cycles the (P)PI is to be collected. Apart from these settings, key points for the IT provider to remember in implementation of the PPI are listed in Box 2. 1.

2.

3. 4. 5. 6.

The index is a composite index and therefore poverty levels can only be generated if all ten questions are fully answered. A validation needs to be integrated to ensure this in the software. Each answer choice has a corresponding numeric value that is static and is statistically linked to the applicable country’s national household survey. If any questions, answer choices, and associated answer values are changed, the statistical link will be broken and the PPI is no longer valid. A PPI score is the sum of all ten answer choices. PPI scores range from 0-100 and must be converted to poverty likelihood calculation to be of value. The poverty rate of a set of clients is calculated by averaging the likelihood scores with reference to specified poverty lines. The MIS should be able to collect multiple instances of the PPI and maintain historical data (as noted earlier, this is the case for all data inputted). Reporting needs to be linked to context variables (geographic, client profiling, loan product and cycle) and benchmarked to country and regional poverty rates.

Box 2: IT implementation issues for the Poverty Index.

MIS as a Potential Catalyst for Social Performance Management

53

Monitoring and responding to client needs and preferences Gathering client feedback and acting upon it is a core component in any business. This type of data is important for maintaining and increasing business, contributing to product innovation, improving customer service, and designing appropriate processes. MFIs, traditionally, may have relied on a single type of (credit) product along with measurement of growth in client numbers and on-time repayments to represent market satisfaction. As perhaps a natural evolution of the sector, as well as a response to more diverse and competitive markets, MFIs are increasingly looking at client retention, as well as options to adapt and diversify their product offering in response to client feedback. For this social goal, the portfolio MIS provides the means to monitor client retention, as well as to provide the market segmentation context for designing client feedback mechanisms and analyzing the feedback that is collected. Monitoring borrower retention is an important tool since high retention rates are a proxy for client satisfaction, whilst low retention rates are a warning sign and should prompt follow up research into the reasons for exit. Client retention is also important to be able to contribute to client progress over time. There has been considerable debate in the sector on how to define a ‘dropout’ and the appropriate formula to use to track the exit rate. See, for example Pawlak et al. (2004), Waterfield (2006), M-CRIL (2007). A degree of consensus around the following formula for client retention: total borrowers at end of period/total borrowers at beginning of period plus new borrowers during the period. This is the formula currently applied by the MIX. MFIs can usefully apply this formula, or adaptations of this formula, in their portfolio MIS to monitor client retention by product, by region, by branch, even by loan officer—provided there is a unique client ID and the ability to identify new clients. Deeper analysis of the exit clients can also be undertaken by loan cycle, as well as by PPI score and other client profile indicators. In addition to regular monitoring of client retention, there are a number of different quantitative and qualitative methods that an MFI can employ to obtain and analyze client feedback. Once the emphasis is put on the client and the inherent value in this is recognized, systems and policies will flow from that focus. Quantitative methods include sample surveys of client satisfaction on different services and issues, reasons for client exit, specific market research, and aggregation of client complaints. To integrate such surveys in reporting, flexible software is required that allows each instance of the survey to be housed under a unique ID. In

54

Chapter Three

addition, these surveys will most likely need to be adapted as new questions evolve. Ideally the software will provide an in-built framework for the survey and will allow the MFI to change questions and answer choices accordingly. ‘Mifos’, an open-source microfinance software, has been able to provide such flexibility to its customers. The software allows the MFI to create a ‘questions group’ which can be used for collecting client level surveys, social indicators, and poverty data. (The software provides a built-in question group for the PI.) The user is allowed to enter their own questions and even designate what type of answer choices they would prefer such as multiple option answer choices, single select answer choices, and open-ended answers. In addition, the questions group can be integrated into different processes such as creating a client or creating a new loan.11 The primary benefit of housing these surveys in the software rather than hard copies is the ability to generate periodic reports for management and the board of directors. The MIS is critical then, not only to routine monitoring of market feedback, but also to enabling the MFI to respond effectively to the findings. Any changes in operations will also require changes in the MIS software. Therefore, it is crucial that software systems can allow for some flexibility in product design and repayment structure. This need for flexibility in products has emerged as a major learning point for the sector from the household financial diaries explored in the book, ‘Portfolios of the Poor’: “Of all the commonalities, the most fundamental is that the households [living below $2/day] are coping with incomes that are not just low, but also irregular and unpredictable, and that too few financial instruments are available to effectively manage these uneven flows. It is a “triple whammy”: low incomes; irregularity and unpredictability; and a lack of tools.” (Collins et al., 2009)

If microfinance is to be able to provide the right sort of financial instruments that can help the poor to manage uneven cash flows, then one of the major challenges will be to have software that enables an appropriate level of flexibility (along with the necessary controls). One example is whether the MIS can cater to variation in frequency of loan repayments (not only weekly but fortnightly or monthly). To cater for this, the software has to be able to calculate the appropriate interest due and overdue regardless of the frequency. Other examples of flexibility include: allowing for pre-payments, partial repayments and rescheduling of loans.

MIS as a Potential Catalyst for Social Performance Management

55

In addition to flexibility in credit repayments, a range of financial services to meet different needs is a recognized component of the drive for financial inclusion. Increasingly, new products and services may be provided in partnerships with third party providers—not only insurance companies, but banks, banking correspondents, and mobile banking facilitators. For this, the MFI software requires a way to interface with third party providers such as an Application Programming Interface (API) to effectively synergize the systems. If the products/services are independent (such as health insurance) then this needs to be a standalone item in the system, not attached (to the loan product) as a supplementary fee for the term of the product/service. Once software systems have these options available, the MIS can support the design of new products such as agricultural loans, emergency bullet loans, and Grameen II-style loans which allow for customized loan repayments. Ultimately, however adequately modified, effective software is only a facilitator, and in the end it is the MFI’s intent and strategy that ensures diverse products catering to client needs.

Tracking client progress over time For MFIs (and their investors and other funders) who seek to contribute to a positive change, the portfolio MIS can support the collection and analysis of data to provide evidence of progress for clients. This approach differs from ‘impact’ which technically attributes any change to the intervention (microfinance program) by comparing the difference with a control sample of non-clients without access to microfinance. But, given the complexities and costs of impact evaluations, tracking progress for clients on selected indicators is a practical alternative that can provide evidence about benefits, or otherwise, for clients. An important element of the analysis is to include transactions data so as to track inputs to outcomes, as well as the client profile data, previously discussed, serving to segment the client market to explore more specifically what is happening to whom. Supplementing this analysis with information on client exit (dropouts and ‘flyouts’), and additional questions and case studies to explore reasons for any change (positive or negative), strengthens the evidence and provides intelligence that can feed back into service design. The MIS features in terms of providing transactions data can be linked to client profile data. Basic transactions analysis includes looking at client distribution by loan cycle and/or loan size categories, and at each loan cycle or product. We typically find significant variation with some clients

56

Chapter Three

increasing their loan size, others staying at a lower level, even after several loan cycles. Clearly, the application of credit and the expected results are going to be very different and following up the reasons for these differences is useful information for tracking change. A similar method can be used to track savings over time, with the ability to accumulate savings and manage savings being an important emerging indicator of client capability. To measure progress, indicators used to profile client and household characteristics at entry to the MFI can be tracked for the same clients (who stay with the program) over time. MFIs using the PI at entry can apply the same index as a PPI to measure change in poverty levels, along with other relevant indicators. A realistic timeframe to track such progress is a minimum of at least three years. To do this the MFI can collect follow up data from all clients or from a sample of clients. For example, data from clients in their 4th loan cycle (after a 3 year period) can be compared with the data an MFI has for these clients at entry (removing data for dropouts during the 3 years, from the overall entry level base line). The progress may also be measured in terms of dropouts (or rather ‘flyouts) who may have graduated say to a formal bank for their financial needs. Such information should be included in a client exit survey. As noted earlier, the portfolio MIS can support the analysis of client progress by linking the findings to different market segments, and client access to different services or a range of services. Some clients may have accessed non-financial services, for example, which would affect the results. Careful analysis of the data can be used to explore the relevance of different combinations of services, how results may vary for different client segments, as well as variations around average results. This in turn can lead to follow up research, including case studies, to explore more deeply the results, contributing to future strategies or product design.

Client Data Reporting and Use As illustrated in the social performance pathway (Figure 1), there is an important feedback loop whereby the client information feeds back into management systems and is used to improve operations. Reports for social performance should be viewed alongside those for high level financial review by the management and the board. Client profile and retention reports should be quarterly as well as annual. Reports related to client feedback may be annual. Client progress reports would be less frequent.

MIS as a Potential Catalyst for Social Performance Management

57

In terms of the IT for MIS portfolio reports, it is recommended that reports are not static but rather flexible which allow the user to choose basic parameters such as: (i) To and from dates for report periods; (ii) Operational units; (iii) Rural and urban area selection; (iv) Women/men clients; and (v) Loan cycle or number of years with the MFI. The client data discussed in this chapter reflects the information that is relevant for reporting to external stakeholders, such as the country network, investors or the MIX. For external reporting, average results across the MFI program are reasonable. For internal reporting for SPM, the MFI needs more segmentation of the information, to identify areas of high and low results, as an effective basis for follow up and improvement. Internal reports should be concise, more like a dashboard than a research thesis, with data presented and tabulated in ways that management are used to seeing. To the extent that the MIS can be programmed to generate customized and segmented reports for client level data, this will help facilitate the regular use by MFI management and board of this social information. Apart from customized reporting, a ‘data dump’ in the MIS is recommended. This is a file that provides an extraction of all client level data available in the software, both social and transactions data. This file allows MFIs to conduct in-depth cross analysis beyond the basic reports available. Maintaining this report provides the MFI with the flexibility to have a full data set on hand in case any additional analysis is necessary. Box 3 illustrates how three MFIs in India are using client level data. These are MFIs who attended EDA workshops on MIS requirements for social performance reporting during 2012. Example 1: Janalakshmi has a dedicated analysis department called the Client Insights Group. The group undertakes periodic research surveys, including use of the PPI, to explore urban poverty issues, to segment the client market, and to track change for the MFI’s clients over time. The management uses client segmentation information to assess client risk and the appropriate product mix. Profile data is collected and maintained for all clients along with transaction data across all products and services they use. This data is regularly mined to analyze the client mix and pattern of access to financial services. Example 2: Grameen Koota uses data collected through the PI to identify clients for their Cook Stove Loan. The organization deep-dives into client level data to identify areas where there is high usage of firewood and charcoal as a

58

Chapter Three

source of cooking energy. The data is used as a management tool to estimate demand and appropriately target branches. Example 3: Ujjivan found from its client exit interviews that a majority of clients were dissatisfied with weekly repayments and center meetings. In response to this feedback, management decided to offer customers the option to switch from weekly to monthly loan repayments and center meetings. This repayment flexibility resulted in a reduction in the dropout rate from 22% to 14%. Ujjivan is now working on reducing loan disbursement turn-around-time, based on client feedback, to further reduce the dropout rate.

Box 3: Using client level data as part of SPM.

Managing improvements in the MIS As with any organizational change, a proper assessment is necessary prior to transitioning. The first step is to discuss with the IT provider the changes required and the limitations, if any, of the software. Below are some key questions the MFI management and the IT provider should discuss: 1. How much alteration is required in the software? What are the associated costs? Are there alternatives to these changes? 2. Does the MFI need to upgrade to a newer version of the software? How much cost is associated with the upgrade? 3. Is it beneficial for the MFI to migrate to a different software? What is the time required and costs associated with migration? As these questions are cleared, the IT provider and MFI can then begin to plan and budget accordingly. Prior to budgeting, an assessment should be made to understand whether MIS and SP staff are available and whether they have the capacity to oversee the project. Other options may be to outsource this function. This discussion should be based on each MFI’s internal assessment. The development of a social reporting system represents time and cost for an MFI. The costs not only include those for the MIS adaptation and time spent to collect the data, but also the costs associated with training staff and conducting data quality audit checks. As a consequence, an MFI should think through the use of each data point-question, and the appropriate method to collect this information and check its quality. Collecting data that is of doubtful quality, or cannot be clearly reported, is inefficient and can be expensive.

MIS as a Potential Catalyst for Social Performance Management

59

Thus, though implementation requires adequate software, it is also crucial that all MFI departments work harmoniously to ensure effective implementation: operations incorporate data collection formats as part of routine client forms; staff are trained in data collection and quality issues; data quality checks are included as part of operations and as part of internal audit; and finally, reports are integrated with management decision making and sent to the Board periodically. Strong coordination will support collection of high quality data and development of appropriate reports for social performance reporting.

Conclusion This chapter has tried to throw some light on the basic IT and MIS adaptations and reporting required for client level analysis in the context of social performance management. The basic tools and indicators for client level reporting are now quite well known and agreed within the microfinance industry. Information is essential to monitor social goals in microfinance. Ultimately, the push for client level information depends on the orientation of the board and senior management, and how they apply the data. Externally too, other stakeholders, including investors/funders and networks are asking for this type of client level information. MFIs are increasingly putting systems in place to track their mission and beginning to see that client level analytics makes business sense. To obtain these insights, a client reporting system needs to be put in place which effectively collates the valuable data collected into a set of reports that can be used for management level decision making. An MIS that can support this is a service to the industry.

Annex: Example of format to collect social profile data on MFI clients The following form outlines examples of client and household profile social indicators, PI indicators (based on the current India PPI), and questions related to loan use and the financed enterprise (if the loan is used at least partially for enterprise investment). The aim of the format is to demonstrate a concise but comprehensive method for data collection. The indicators represent those which are likely to be relevant for profiling client households and tracking change over time. Specific indicators should be changed and adapted to suit different microfinance models, and country contexts.

Chapter Three

60

Member Profile Branch ........................................................................................................... Village or Town ............................................................................................. Date filled on Filled by Checked by A Client Information / House Hold Information Name

ID

Female/ Male Residence

Age (yrs)

Client Education

01. No/few years’ formal schooling 02. Completed primary 03. Middle 04. Secondary 05. Diploma/certificate/ Higher Secondary /graduate 01. Single 02. Married 03. Widow/divorced/separated/Widower

Marital Status

18-25, 26-35, 36-45 >45

01. Rural 02. Semi-urban 03. Urban

Phone number

Client Occupation (multiple responses possible)

01. Agriculture 02. Agri Allied – livestock 03. Non-farm business (HH) 04. Non-farm business (Own) 05. Casual labor 06. Skilled Labor 07. Salaried employment 08. Home-based piece rate work 09. Housework

Household livelihoods/ sources of income (multiple responses) Religion

01. Agriculture 02.Agri Allied – livestock 03. Non-farm business 04. Casual labor 05. MNREGA 06. Skilled Labor 07. Salaried employment 08. Home-based piece rate work 09. Remittance 10. Pension

Number of members …………. Members with disability? No/Yes Number of Girls between 5-17 years Ownership of House 01. Rented 02. Own 03. Parental If own house, is the house in your/joint name? Y / N Drinking water

Adults more than 17 Working or Earning members years: If yes: 01.Self 02.Spouse 03.Child 04.Parent Number of girls 5-17yrs) going to school regularly Type of housing? 01. Kuccha 02. Semi Pucca

Toilet?: 01. None

02. Public/Shared

03. Pucca

03. Own

1. Pond, public well 2. Shared hand pump/well 3. Own hand pump/tap connection 01. No 02. Boil 03. Use a water purifying or filtration method

Do you purify water for drinking? Does the household have cultivable land ?

Y /N

Do you have land registered in your/joint name? Y/N

MIS as a Potential Catalyst for Social Performance Management B PPI Indicators 1 How many household members are 17 years old or younger? >=4 3 2 1 0 2 What is the general education level of the male head/spouse? A. No male head/spouse B. Not literate, no formal school, or primary or below C. Middle D. Secondary or higher secondary E. Diploma/certificate course, graduate, or postgraduate 3 What is the household type (based on main source of income)? A Labor (agricultural, casual, or other) B. Self-employed (agric, non-agric), C. Regular wage/salary

61

5 Does the household possess any casseroles, thermos, or thermo ware? A. No B. Yes 6 Does the household possess a TV& a VCR/VCD/DVD player? A. No B. Yes, TV C. Yes, both 7 Does the household possess a mobile handset & a telephone instrument (landline)? A. No, neither one B. Yes, only a mobile C. Yes, a landline, regardless of mobile 8 Does the household possess a sewing machine? A. No B. Yes 9 Does the household own an almirah/dressing table? A. No B. Yes 10 Does the household possess a bicycle, motorcycle/scooter, or motor car/jeep? A. None B. Yes, bicycle only C. Motorcycle/scooter D. Motor car/jeep (regardless of others)

4 What is the primary source of energy for cooking? A . Firewood & chips, dung cake, kerosene, charcoal, coke or coal, gobar gas B. LPG or Electricity C. No cooking arrangement C Access, participation Do you have an individual Y/N Are you member of a Y/N savings account (Bank/PO)? chit fund? Has the Y/N If yes, for what purpose did you borrow? household 01. Agric 02. Business 03. Medical borrowed from 04. School/educ 05. Other household needs a moneylender 06. Social events (wedding, funeral, festival) in past one year?

Bibliography Campion, Anita, Chris Linder and Katherine Knotts, 2008. “Putting the ‘social’ into performance management: A practice-based guide for microfinance”. Institute of Development Studies, UK Chen, Greg, Stephen Rasmussen, and Xavier Reille. 2010. "Growth and Vulnerabilities in Microfinance". CGAP Focus Note No. 61, February. Collins, Daryl, Jonathan Murdoch, Stuart Rutherford, and Orlanda Ruthven. 2009. Portfolios of the Poor. Princeton: Princeton University Press. Grameen Foundation USA. 2011. "Case studies & Papers". Progress out of Poverty. Accessed December 2012. http://progressoutofpoverty.org/casestudies.

62

Chapter Three

Lascelles, David and Sam Mendelson, 2012. Microfinance Banana Skins: Staying Relevant. Centre for the Study of Financial Innovation. Microcredit Ratings International. 2007. Estimating client exit rate. Technical Note, Gurgaon: M-Cril. Pawlak, Katarzyna, and Selma Jahic. 2004. Promoting Client-focused Organization- Partner's Exit Monitoring System. MFC Spotlight Note, Microfinance Centre. Sinha, Frances; 2006. Social Rating and Social Performance Reporting in Microfinance: Towards a common framework. Washington D.C.: SEEP Network. Spitzer, Dean. 2007. Transforming Performance Management: Rethinking the way we measure and drive organizational success. New York: American Management Association. Waterfield, Chuck. 2006. The Challenges of Measuring Client Retention. Putting client assessment to work: Case study #2, Washington D.C.: SEEP Network.

Notes 1

The authors are with EDA Rural Systems, Gurgaon, India. This paper draws directly on EDA’s experience in introducing SPM with MFIs in India and other countries of S Asia, SE Asia and Africa. Our recent initiatives have included direct engagement with IT providers in India. We would like to acknowledge and thank the MFIs that we work with, the social investors who support us, EDA colleagues present and previous, as well as the IT providers who have begun to look more closely at ways to include social performance reporting. 2 The Imp-Act consortium represents 12 organizations, networks, MFIs, investors, and TA providers, working across all global regions. They are: AZMJ, CARD and Microfinance Centre of the Philippines, CRS – Mission Project (LAC, Africa), EDA Rural Systems (Asia, Africa), Freedom from Hunger, Grameen Foundation, IDS, University of Sussex (UK), Microfinance Centre EECA, Oikocredit, Pro Mujer (LAC), and Sanabel (MENA). Information about the consortium and materials on SPM is available at www.imp-act.org and www.spmresourcecentre.net 3 http://www.cerise-microfinance.org/spip.php?page=article&id_article=263 4 There are four specialist microfinance rating agencies: Micro-Credit Ratings International Ltd (M-CRIL), MicroFinanza Rating (MFR), Micro-Rate and Planet Rating. Through a working group of the Social Performance Task Force, the rating agencies proposed the framework for social performance reporting. This is documented in Sinha, 2006, available at: http://www.seepnetwork.org/socialrating-and-social-performance-reporting-in-microfinance--towards-a-commonframework-resources-309.php 5 http://www.mixmarket.org/social-performance-data 6 www.progressoutofpoverty.org

MIS as a Potential Catalyst for Social Performance Management

63

7 Since 2006, the microfinance sector has faced crises linked to very rapid growth of service providers in a number of countries, particularly Nicaragua, Bosnia and Herzegovina, Morocco, Pakistan, and India. 8 Whilst different organizations have different missions, there is broad agreement on social values in microfinance as including: increasing financial access for the poor and excluded, improving the quality and appropriateness of services, reducing vulnerability, and alleviating poverty. For more details refer to: http://sptf.info/what-is-social-performance 9 Examples of this include the Cashpor Housing Index developed by Cashpor in India, and the Client Assessment and Monitoring System developed by Buusaa Gonofaa in Ethiopia. http://www.cashpor.in/chi.htmlhttp://www.bgmfi.com/spm.htm 10 For more details refer to: http://progressoutofpoverty.org/. Another similar tool is the USAID Poverty Assessment Tool (PAT) developed by the IRIS Centre at the University of Maryland. Other tools and score cards have been developed by MFIs and these are often applied for targeting. If the data is captured, the data can also be used in the same way (profiling, market analysis, and change over time). The difference between those tools and the PPI and the PAT is that the latter tools enable benchmarking to specific poverty lines which is relevant for in-country and for international comparison. 11 For more details refer to: http://mifos.org/functionalspecifications/retrievinginformation/question-groups.

CHAPTER FOUR MIS AND REPORTING IN MICROFINANCE IN THE FRAMEWORK OF THE TRIPLE BOTTOM LINE GLÒRIA ESTAPÉ-DUBREUIL, CONSOL TORREGUITART-MIRADA AND M. ROSA ROVIRA-VAL

Introduction Corporate reporting has experienced a gradual evolution from conventional financial reports and related accounting methods to more comprehensive reports providing information on intangible assets and nonfinancial issues of the organization (Skouloudis, Evangelinos and Kourmousis, 2009). Public information on topics related to environmental and social performance has been progressively demanded by investors and other stakeholders, including customers, suppliers, employees, communities and other social groups, in order to make accurate decisions and to have a more comprehensive description of corporate impacts, risks and performance (Logsdon and Llewellyn, 2000; Repetto, 2005; Rasche and Esser, 2006). The International Survey of Corporate Responsibility Reporting, conducted by KPGM every three years, shows the increase in corporate attempts to account for their social and environmental impacts. The KPMG (2005) survey reveals an increase from 13% to 33% in the number of top companies that provide separate “sustainability reports”. In its 2008 edition, the KPMG concludes that corporate responsibility reporting has become quite common, since nearly 80% of the largest 250 companies worldwide (G250) issued reports, and so did, on average, 45% of the largest 100 companies in 22 countries (Mintz, 2011). The KPMG (2011) survey stresses the increasing numbers of leading companies that have

MIS and Reporting in Microfinance

65

combined their corporate responsibility report and financial reporting, often by merging the two into their annual report. Within the G250, slightly more than 25% have incorporated Corporate Responsibility (CR) reporting into the directors’ report, or have a dedicated sub-report on the topic. Reporting on non-financial issues is neither universal nor standardized, with no enforcement by the accounting standards-setting organizations (Christofi, Christofi and Sisaye, 2012). The different names given to such reports (triple bottom line reporting, social and environmental report, sustainability report or corporate social responsibility report) hinder clarity. Some consider the term "sustainability report" as problematic because of the difficulty of applying sustainability at an organizational level (Gray and Milne, 2004), preferring the use of the all-encompassing term, coined by Elkington (1997), of triple bottom line (TBL) reporting (Archel, Fernández and Larrinaga, 2008). The most widely recognized guidelines for TBL disclosure were developed by the Global Reporting Initiative (GRI), a proposal rooted in the US non-profit organizations, the Coalition for Environmentally Responsible Economies (CERES) and the Tellus Institute. Under the guidance and support of the United Nations Environment Programme (UNEP), the first set of guidelines was launched in 2000, and it is now in its third edition, the G3 guidelines. While compliance with the GRI’s Guidelines is voluntary, its use is widespread within large companies (Sherman, 2012). Reporting within the Microfinance industry is also widespread and can arise from a variety of internal and external requirements, including accountability to donors and other stakeholders with diverse expectations and goals, responsiveness to external governance pressures, or progressive attaining of an organization’s mission. Evidence of Microfinance Institutions (MFIs) reporting is available at Microfinance Information Exchange (MIX), a non-profit organization established in 2002. Through its web-based platform MIX Market, it provides performance information on more than 2,700 MFIs worldwide1. The information is supplied voluntarily by the MFIs, who are also encouraged to provide some form of third-party validation. In about two-thirds of the cases, the MIX data for 2010 is supported by audited financial statements or by an external ratings report (Gaul, 2012). At least two different dimensions, the size of the Microfinance Institution (MFI) and its commercial orientation (for-profit organizations vs. non-profit NGOs), have been shown to influence the level and type of disclosure. Concerning their reporting on the Internet, Gutiérrez-Nieto et

66

Chapter Four

al. (2008) concludes that for-profit MFIs disclose more financial data on their web sites, while non-profit NGOs reveal more social information. Such a trend may nevertheless be changing, as according to Pisteli et al. (2011), who, after reviewing 405 reports of MFIs reporting to the MIX Market during 2009 and 2010, concluded that the Microfinance industry as a whole perceives social reporting as part of the disclosing behavior in any institution, regardless of its status. The controversies concerning social disclosure remain nonetheless unresolved (Hes et al., 2012). While financial statement transparency alone may not have a role in the enhancement of the MFI’s performance, according to the study of Hartarska (2009), an increasing number of MFIs perceive the need to report their social and, more recently, environmental performance (Allet, 2012). However, there is a lack of a truly common standard for disclosing relevant information in the microfinance field. Stephens (2012) summarizes the review of 18 different reporting formats, finding that many of them asked for widely different information, and, when asked for the same information, did it in different ways. The GRI guidelines are among the formats considered by some MFIs when preparing their annual reports. The database of the GRI contains recent reports of diverse Microfinance Institutions, such as Banco Fie in Bolivia, Banco Solidario in Ecuador or XAC Bank in Mongolia. An evaluation of GRI as social audit tool issued within the Microfinance Industry (SPTF User Review, 2009) highlights GRIs guidelines as providing worldwide comparability for the reports of its users, not just with other MFIs. However, it is also pointed out that “implementing the tool requires well-developed reporting systems, including a flexible management information system (MIS)” (op. cit. p.2). In addition, the financial cost may be significant if an institution’s MIS has to be upgraded to implement the GRI. In our opinion, the actual predisposition to disclose on social performance will eventually lead to more MFIs choosing TBL reporting. This opens the question of whether the actual reporting practices of the MFIs are very different from those expected for TBL reporting. Furthermore, deeper insights on the specific requirements that an institution’s MIS must fulfill to implement the GRI are required. This chapter seeks to contribute to this discussion. Our starting point is a brief study of the actual reporting recommendations and practices in the microfinance industry. It is followed by a brief description of the Global Reporting Initiative. The background provided by these two sections allows us to answer the first question mentioned above by comparing the actual reporting standards in the

MIS and Reporting in Microfinance

67

microfinance industry with those of the TBL reporting. In turn, it leads to a discussion on the requirements that a MIS must fulfill to provide TBL standardized reporting capabilities. A few final remarks conclude the chapter.

Reporting in Microfinance: an overview Since the late 1990s, there have been increasing efforts in the microfinance industry worldwide to set and implement reporting standards. The objective centered primarily on financial information, gradually extended to offer disclosure guidelines on social performance. More recently, environmental performance has also been considered for inclusion in the reporting framework. We shall consider these complementary issues successively in this section.

Financial reporting The differences in the spread and development of microfinance among countries, as well as the legal status and the funding sources of the MFIs, has not favored regulation that could mandate an international standard basis for financial reporting. In their review of MFIs’ regulatory framework, van Greunin et al. (1999) observed that “few MFIs are legally required to disclose their financial condition with the rigor demanded of licensed banks, which makes meaningful comparisons of MFI financial statements rather difficult” (p.33). The review concluded that the collection and provision of operating and financial information under a well-organized and orderly reporting system was “one of the foundations for a viable prudential regulatory framework”. Such information reports should be useful both for internal management and for donors and other stakeholders, whether the MFI is legally an NGO or a for-profit institution. They also recommended that independent private institutions provide further insights on the development of minimum standards for the microfinance industry, similarly to those provided by credit rating agencies in global capital markets. Indeed, in 2000, the Consultative Group to Assist the Poor (CGAP), an independent policy and research center hosted by the World Bank and supported by many development agencies and private foundations, started a process to develop a set of disclosure guidelines for financial reporting by MFIs. The final version of these guidelines (Rosenberg et al., 2003) were developed in consultation with microfinance practitioners, taking into account the results of field-tested inputs from the Small Enterprise

68

Chapter Four

and Education and Promotion Network (SEEP), a consortium of microfinance practitioner and microenterprise development networks. CGAP guidelines specify information that should be included in MFI's financial reporting. The suggested disclosure should be considered as the minimum set of financial information needed to understand the condition of an MFI’s operations to assess its financial sustainability and loan portfolio. CGAP disclosure guidelines are based on the International Financial Reporting Standards (IFRS) and the International Accounting Standards (IAS). However, there is no explicit recommendation of any particular format for disclosing the information, thus allowing its use in any country, regardless of its accounting standards and methods of financial presentation. Table 1 summarizes the main disclosure items recommended by the CGAP guidelines, as well as some relevant details on the type of disclosures expected in each of the statements. The guidelines also ask for details on the methods used to account for most of the items on the statements, as well as the particulars of the policies underlying the practices used for provisions. Furthermore, detailed information regarding all loans made to the MFI, including terms, guarantee mechanisms, interest expenses and, if applicable, full details of any arrears, must be provided. In 2005, the SEEP Network published an update on microfinance reporting standards (SEEP, 2005) that included a set of 18 ratios designed to measure MFI performance in four areas: (i) sustainability/profitability, (ii) asset and liability management, (iii) portfolio quality, and (iv) efficiency / productivity. This set was further enhanced in 2009 with the proposal of eight new ratios, some of them focusing on savings to fill the gap within the first publication, which was tailored to credit-only institutions (Tulchin et al., 2009). In 2010, the Microfinance Reporting Standards Initiative, a subcommittee of the SEEP Network, compiled and revised the ratios to offer an updated set with standard definitions (SEEP, 2010). Designed for use by all kind of MFIs (NGOs, non-bank financial institutions, commercial banks, rural banks, credit unions, cooperatives), one of its primary goals is to ensure that microfinance financial performance ratios are calculated in a consistent manner to facilitate comparability and to improve decision making for a range of microfinance industry stakeholders.

MIS and Reporting in Microfinance

Statements

Relevant details

Balance Sheet

Showing separately: x Any type of deposit account tied to the ability of the MFI clients to obtain loans x Long-term deposits x Equity investments required from clients, if applicable

Income (profit and loss) Statement

Particularly, the income statement should clearly show: x The amount of current period donations x Provision expenses related to loan losses x The amount of the allowances for loan losses x The amount of loans written off during the period x Income from investments, separately from interest, fees or other loan income collected from borrowers

Portfolio Report

Including a table reconciling the accounts affecting the loan portfolio: x Loan portfolio at the beginning and end of the period x Loan loss allowance at the beginning and end of the period x Loan loss provision expenses recognized during the period x Write offs of uncollectable loans during the period

69

It should also show: x The extent of late payment on loans for the current reporting period x Outstanding loans (to members of the MFI’s management, or governing body) at the beginning and end of the period x Voluntary savings accounts at the beginning and end of the period, if applicable

Table 1. Main disclosure items and details recommended by the CGAP financial disclosure guidelines.

Table 2 summarizes these revised ratios, grouped in five categories (profitability, capital adequacy and solvency, liquidity, portfolio quality and efficiency and productivity). Three quarters of these are “core ratios”, while the remaining ones should be reported only if applicable (e.g., if MFIs accept savings). The table also includes the ratios that can be found on the MIX Market, either in exactly the same form or as a closely related ratio.

Chapter Four

70

Areas

N/C

Ratio

MIX

Portfolio yield ł Net interest margin ł Return on average assets (ROA) ł Return on average equity (ROE) Financial expense ratio /ł/ Impairment expense ratio ł Operating expense ratio Capital Debt to equity ratio ł Adequacy Equity to assets ratio /ł/ and * Capital adequacy ratio (CAR) Solvency * Uncovered capital ratio (UC) /ł/ Liquidity Cash ratio * Savings liquidity * Loans to deposits ratio /ł/ Portfolio Non-Performing Loan (NPL) 30 days past due /ł/ Quality ł Write-off ratio NPL30 + write-offs ratio Efficiency Portfolio to assets ł and Cost income ratio /ł/ Productivity Cost per active client /ł/ Borrowers per loan officer ł Active clients per staff member /ł/ Client drop out ł Average outstanding loan size Average loan disbursed * Average deposit account balance ł * Average deposit account balance per depositor ł Legends: * non-core ratio; ł raƟo disclosed on MIX; /ł/ a similar raƟo found in MIX Profitability

Table 2. Financial ratios recommended by Microfinance Reporting Standards Initiative (SEEP, 2010).

Social reporting Compared to financial reporting, social reporting is not yet as widespread among MFIs. On the average, only 30% of the MFIs report to MIX on social performance (unevenly distributed, since it attains almost 51% in Latin America2 and only 16% in sub-Saharan Africa). Although the social mission of most MFIs cannot be doubted, financial sustainability, and therefore financial reporting, have been the main concern of the Microfinance industry. Moreover, MFIs were encouraged to

MIS and Reporting in Microfinance

71

be more “business-like” so they could access investment funds rather than continue depending on donor grants (Sinha, 2006). Performance Categories

Process Indicators

Mission and Social Goals

Type of client targeted Development objectives specifically pursued Poverty level of the clients targeted Board members trained on social performance management Formal committee monitoring of social performance Credit products / Savings products / Insurance products / Other products Enterprise services / Education services / Health services Women empowerment services Consumer protection principles: 1. Evaluation of borrower repayment capacity and loan affordability 2. Internal audits to prevent indebtedness increase 3. Internal valuation of portfolio quality 4. Full disclosure to clients of the conditions of all financial products 5. Training staff to effectively communicate with customers 6. Code of ethics or similar with clear rules on debt collection practices 7. Value and rewarding of ethical behavior in the corporate culture 8. Active mechanism to handle customer complaints 9. Explicit rules to ensure clients’ knowledge of the use of their data Transparency of costs of services to clients Transparency on salary / Benefits (medical insurance, etc.) / Protection at work / Equality (equal pay for men and women in equivalent levels) Staff incentives related to: the ability to attract new clients/ outreach to remote communities / outreach to women / portfolio quality, etc. The questionnaire asks for 6 different policies, such as: rising clients’ awareness about environmental impacts / training clients / identifying enterprises with environmental risks / offering special lending products Poverty measurement tool used (if any) Typology of clients surveyed for poverty measurement Specifically target products offered

Governance Range of Products and Services Social Responsibility to Clients

Human Resources and Staff Incentives

Social Responsibility to the Environment Poverty Outreach

Table 3. Social Performance Process Indicators as reported to the MIX Market (MIX, 2011).

72

Chapter Four

Efforts aimed at defining and measuring social performance in MFI can be traced to the Social Performance Indicators Initiative project established in 2002 by the Argidius Foundation, CERISE (Comité d’Echange, de Réflexion et d’Information sur les Systèmes d’Epargnecrédit), a platform of French-based leading Microfinance support organizations, and CGAP (Zeller, 2003). In 2005, the leaders from various social performance initiatives agreed on a common framework to develop social performance measurement (CGAP, 2007). The Social Performance Task Force (SPTF), today consisting of over 1,300 members from diverse microfinance stakeholder groups, was then formed to “develop, disseminate and promote standards and good practices for social performance management and reporting” (http://www.sptf.info/sptaskforce). In 2011, after a pilot project carried out during 2009 and 2010 involving over 400 MFIs reporting to the MIX specifically on social performance, the SPTF and the MIX developed a set of social performance indicators on which data would be collected (MIX, 2011). Tables 3 and 4 provide an overview of the indicators currently reported to the MIX Market, differentiating between process and result indicators3. Performance Categories

Result Indicators

Mission and Social Goals Governance

Number of active borrowers / Percent of female borrowers Borrower retention rate / Active clients Number of board members / Percent of female board members Personnel / Percent of female staff / Staff turnover rate Number of managers / Percent of female managers Loan officers / Percent of female loan officers Number of microenterprises financed Number of start-up microenterprises financed Percent of financed microenterprises that are start-ups Number of jobs created Number of clients surveyed for microenterprise data Number of microenterprises surveyed for employment data Number of clients surveyed for poverty measurement First poverty line considered / Clients below first poverty line Second poverty line considered / Clients below second poverty line

Human Resources Enterprises Financed and Employment Creation

Poverty Outreach

Table 4. Social Performance Result Indicators as reported to the MIX Market (MIX, 2011).

MIS and Reporting in Microfinance

73

Although included within the social reporting recommendations, one of the items listed in Table 3 actually refers to the third dimension of the triple bottom line reporting, i.e., reporting on environmental issues. This is nevertheless an undeveloped area for the microfinance industry, in which until now only the proposals of Allet (2011, 2012) have shed some light.

The Global Reporting Initiative The Global Reporting Initiative (GRI) was created in the late 1990s as a temporary project to provide an international accountability tool to report on environmental performance. From its inception, the scope extended to social, economic, and governance issues. The tool was developed in a rapid continuous improvement system that produced three generations of reporting guidelines in the short period of one decade (the first version was launched in 2000, the second in 2002, the third generation, G3, in 2006 and an extended version of this, the G3.1, in 2011). The next generation, G4, is expected in 2013. As of December 2012, GRI provides a database of up to 4,800 organizations and more than 10,500 GRI Reports (http://database.globalreporting.org/). The success of the GRI can be explained by six different and complementary traits. First, the process used to establish its guidelines, for GRI has been one of the first organizations to use a multi-stakeholder international consultation process from the beginning, with all standards or guidelines versions developed by an international group of experts. The resultant work was then exposed to public comment for a period, thus ensuring that the resulting standard is as appropriate and credible as possible for all sectors and organizations. The second aspect relates the GRI’s multi-stakeholders network, with a global network of more than 600 core supporters, its “organizational stakeholders”, with representation from civil society, business, mediating institutions, academia, labor and public agencies. Thirdly, the setting of “focal points”, GRI offices in particular countries, with the mission of encouraging the reporting standard adoption in five key countries (USA, Australia, Brazil, China and India), has also contributed to the widespread use of the GRI guidelines. Fourthly, GRI has also established strategic alliances with the key standard-setting bodies (OCDE, UNEP, UN and ISO) to ensure a good complementarity and to avoid competition among the different standards. Fifthly, GRI’s guidelines, besides a general reporting framework, include sector guidelines for several industries, to obtain more appropriate reporting for that sector and sector-specific performance indicators. Last but not the least, GRI offers a growing list of services to provide support

74

Chapter Four

to its users and networks (coaching and training, software certification, guidance for small and medium sized enterprises in initial reporting, and certification of completed reports), including educational and other dissemination publications and a biannual conference on sustainability and transparency. The GRI reporting framework is a tool designed to improve organizations’ transparency and accountability. The use of this international standard ensures that reports will correspond with the topics that any interested user would expect on sustainability dimensions (economic, environmental, social and governance). Among the core principles of the GRI reporting tool, both the stakeholder engagement principle and the materiality principle should be highlighted. The stakeholder engagement principle requests the organization to undertake dialogue with its main stakeholders to garner their opinions and interests on the major material impacts, risks and opportunities that the organization should face, manage and improve upon. This is truly a big change compared with the previous traditional reporting culture based on the organization’s self-interest. Stakeholder interest should be combined with the focus of the organization and this mixture will determine the relevant impacts of its activity on the different dimensions of sustainability (materiality principle). Finally, the management, measuring and reporting of the performance follows the commonly-used international standard. Since the voluntary exercise of preparing a sustainability report is a challenge for any organization, the GRI standard has been designed as a progressive learning system to smooth adoption by any kind of organization. In fact, the GRI has a specific program, the Business Transparency Programme, “to develop reporting capacity within first-time reporting companies to foster responsible management and transparency regarding economic, environmental, social and governance impacts”4. A similar goal is achieved by offering a progressive adaptation to the GRI guidelines through a system of three levels of reporting, from level C (the simplest or least complete one for beginners) to level A (for advanced reporters). Furthermore a “plus” (+) indicates at each level that external assurance has been used for the report. The GRI reporting guidelines ask for three types of standard disclosures: strategy and profile (setting the overall context for understanding organizational performance), management approach (to disclose how the organization addresses a given set of topics) and performance indicators (to elicit comparable information on the economic, environmental, and social performance of the organization) (GRI, 2011). Therefore, they are configured to report both on the context (strategy and analysis, governance, commitments, engagement, and management approach) and on the results

MIS and Reporting in Microfinance

75

(economic, environmental, labor practice, human rights, society, and product responsibility) of a given organization. It is the number of performance indicators disclosed, as well as its distribution across the different categories, which determines the level of the resultant reporting. The next section provides further details on the nature of the disclosure items for each type.

TBL reporting in Microfinance compared to international standards In this section, we shall discuss whether the level of disclosure of a MFI following the microfinance industry standards would be enough to fulfill international standards on corporate reporting, and what additional information should be added, if necessary. The TBL approach taken in this paper assumes the use of the GRI guidelines as the international standard reference. Since GRI does not entail a closed form of reporting, disclosure of sustainable performance can be combined with annual or financial reports in an integrated report. The identification of the disclosed elements is made through a “GRI Content Index”, listing every guideline’s disclosure addressed in a given report. We therefore use such index lists as guidelines to the discussion. As for the actual reporting practices of the MFIs, we use the reporting information showed in the MIX Market. For each MFI, such information is divided into three parts. The first part provides a general profile of the MFI (addresses and contact information, background, mission, products and services, main funding sources, etc.) as well as a social performance profile, and partnerships and related organizations. The second provides data reporting on various financial indicators, products and clients, outreach indicators as well as social performance indicators. Finally, files submitted by the MFI containing audits, annual reports, and social performance reports, can be accessed. For the purpose of this discussion, only the systematic information provided in the first two categories are considered. Tables 5A and 5B, considered together, contain the list of the 42 items recommended under the first of the three types of standard disclosure in the GRI report guidelines, concerning Strategy and Profile. Ticks on the last column indicate either that the information is currently available (provided that the MFI has made it available) on the MIX Market or that it should be known by the MFI in order to report to the MIX. Under the information known by the MFI, we have included items such as the governance structure of the organization (4.1), its operational structure (2.3), the process for defining the report content and its boundaries (3.5, 3.6), or significant changes occurred during the reporting period (2.9, 3.11).

Chapter Four

76

Strategy and Profile disclosure items

MIX

1. Strategy and Analysis 1.1 1.2

Statement presenting the overall vision and strategy for the short, medium and long term, including strategic priorities; broader trends affecting the organization; key events, achievements, and failures, etc. Description of key impacts, risks, and opportunities

9(P) *

2. Organizational Profile Name of the organization Primary brands, products, and/or services Operational structure of the organization, including main divisions, 2.3 operating companies, subsidiaries, etc. 2.4 Location of organization's headquarters Number of countries where the organization operates, and names of 2.5 countries specifically relevant 2.6 Nature of ownership and legal form Target audience and affected stakeholders. Markets served (including 2.7 geographic breakdown, sectors served, and types of affected stakeholders/customers/beneficiaries) 2.8 Scale of the reporting organization Significant changes during the reporting period regarding size, structure, or 2.9 ownership 2.10 Awards received in the reporting period 2.1 2.2

9 9 a 9 a 9 9(P) 9 a a

3. Report Parameters 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13

Reporting period (e.g. fiscal/calendar year) for information provided 9 Date of most recent previous report (if any) 9 Reporting cycle (annual, biennial, etc.) 9 Contact point for questions regarding the report or its contents 9 Process for defining report content a Boundary of the report (e.g. countries, divisions, subsidiaries, etc.) a Specific limitations on the scope or boundary of the report (if any) a Basis for reporting on related entities that can significantly affect a comparability from period to period Data measurement techniques a* Explanation of the effect and the reasons of any re-statements of a information provided in earlier reports Significant changes from previous reporting periods in the scope, a boundary, or measurement methods Table identifying the location of the Standard Disclosures in the report Policy and current practice with regard to seeking external assurance for a* the report Legends: * not required for Level C / 9available in the MIX Market / (P) only partial availability / a known by any reporting MFI

Table 5A. List of guidelines disclosure under the GRI guidelines 3.1. Part I: Profile Disclosures (Sections 1 to 3 of 4).

MIS and Reporting in Microfinance

Strategy and Profile disclosure items (cont.)

77

MIX

4. Governance, Commitments, and Engagement 4.1 4.2 4.3

4.4

4.5 4.6 4.7

4.8

4.9 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17

Governance structure of the organization, including committees under the highest governance body responsible for specific tasks, such as setting a strategy or organizational oversight Indicate whether the Chair of the highest governance body is also an a executive officer For organizations that have a unitary board structure, state the number and gender of members of the highest governance body that are independent a and/or non-executive members Mechanisms for internal stakeholders (e.g. members), shareholders and employees to provide recommendations or direction to the highest governance body Linkage between compensation for members of the highest governance body, senior managers, and executives, and the organization's performance a* (including social and environmental performance) Processes in place for the highest governance body to ensure conflicts of * interest are avoided Process for determining the composition, qualifications, and expertise of the members of the highest governance body and its committees, including * considerations of gender and other indicators of diversity Internally developed statements of mission or values, codes of conduct, and principles relevant to economic, environmental, and social * performance and the status of their implementation Procedures of the highest governance body for overseeing the organization's identification and management of economic, environmental, * and social performance Processes for evaluating the highest governance body's own performance * Explanation of whether and how the precautionary approach or principle is * addressed by the organization Externally developed economic, environmental, and social charters, * principles, etc. endorsed by the organization Memberships in associations and/or national/international advocacy 9* organizations List of stakeholder groups engaged by the organization Basis for identification and selection of stakeholders with whom to engage Approaches to stakeholder engagement, including frequency by type and * by stakeholder group Key topics and concerns raised through stakeholder engagement, and * responses of the organization Legends: * not required for Level C / 9available in the MIX Market / (P) only partial availability / a known by any reporting MFI

Table 5B. List of guidelines disclosure under the GRI guidelines 3.1. Part I: Profile Disclosures (Section 4).

78

Chapter Four

Considered together, and at the expense of writing down the specific background information required, the list covers almost all relevant aspects needed to qualify for Level C. The whole list of the guidelines for Level C disclosure under the GRI guidelines in the three first sections (Table 5A)—strategy and analysis, organizational profile and report parameters—can be easily completed by any MFI actually reporting to the MIX. Nonetheless, it should be noted that three additional items related to the MFI governance are needed to qualify for Level C. Those three items, related to the stakeholders of the MFI (written in italics in Table 5B) are not considered either explicitly or implicitly by the actual standard reporting practices of MFIs. Disclosure levels A and B under the GRI guidelines would also require the design (if not already made) of several mechanisms and processes related to governance, as well as the description of key impacts, risks and opportunities faced by the organization. The corresponding items have been marked with a star in Tables 5A and 5B. The second obligatory part of the TBL sustainability report consists of the disclosure of performance indicators. The list of performance indicators is organized in three dimensions: economic, environmental and social. Social indicators are further categorized by labor, human rights, society and product responsibility. Totally, the list contains 82 indicators, 55 of them considered as core indicators. It also includes a few additional industry specific indicators. Since there are no specific recommendations for microfinance, we have considered those specific to the finance industry as well as those elaborated by the NGO sector. Advanced GRI reporting organizations would also need to report on its management approach to disclosure (part II of the GRI guidelines) in each aspect of the six resultant performance indicator categories. Level C requires reporting on at least 10 core performance indicators, at least one of them for each of the three main dimensions. Level B requires full reporting on at least 20 performance indicators, 14 of them core indicators, and at least one from each of the six final categories. Tables 6A and 6B lists a total of 27 performance indicators that are either on the list of indicators actually reported in the MIX Market, or could easily be fulfilled by the MFIs under the Standards for Social Performance Management (SPTF, 2012)5. Note that the number and distribution of the selected indicators largely meets the GRI requirements for Level C reporting, and that Level B could be easily reached by any MFI actively engaged in social performance management.

MIS and Reporting in Microfinance

Performance Indicators

79

MIX

Economic EC01

c

EC04 c EC05 EC06 c EC07 c EC09 NG07 NG08

Direct economic value generated and distributed (revenues, operating costs, employee compensation, donations and other community investments, retained earnings, and payments to capital providers and governments) Significant financial assistance received from government Range of ratios of standard entry level wage by gender compared to local minimum wage Policy, practices, and proportion of spending on locally-based suppliers Procedures for local hiring and proportion of senior management hired from the local community Understanding and describing significant indirect economic impacts, including the extent of impacts Resource allocation Sources of funding by category and five largest donors and monetary value of their contributions

9F 9F a a a 9SP 9F 9F

Environmental Initiatives to reduce indirect energy consumption and reductions 9SP achieved Initiatives to mitigate environmental impacts of products and 9SP EN26 c services, and extent of impact mitigation Others Financial or NGO specifics Monetary value of products and services designed to deliver a specific social benefit for each business line broken down by 9F FS07 purpose System for program monitoring, evaluation and learning, (including measuring program effectiveness and impact), resulting a NG03 changes to programs, and how they are communicated Measures to integrate gender and diversity into program design, implementation, and the monitoring, evaluation, and learning 9SP NG04 cycle Legends: c core indicator / 9F Financial Disclosure Guidelines / 9SP Mix market Social Performance Indicators / a easily known by any reporting MFI EN07

Table 6A. List of GRI guidelines Performance Indicators in the Economic and Environmental categories that can easily be fulfilled by MFIs.

80

Chapter Four

Performance Indicators (cont.)

MIX

Social: Labor Practices and Decent Work Total workforce by employment type, employment contract, and 9SP LA01 c region, broken down by gender Total number and rate of new employee hires and employee 9SP(P) LA02 c turnover by age group, gender, and region Benefits provided to full-time employees that are not provided to 9SP LA03 temporary or part-time employees, by major operations Average hours of training per year per employee by gender, and a LA10 c by employee category Composition of governance bodies and breakdown of employees 9SP(P) LA13 c per employee category according to gender, age group, minority group membership, and other indicators of diversity Ratio of basic salary and remuneration of women to men by 9SP(P) LA14 c employee category Social: Society Access points in low-populated or economically disadvantaged a FS13 areas by type Initiatives to improve access to financial services for 9SP FS14 disadvantaged people Social: Human Rights Operations and significant suppliers identified as having a HR06 c significant risk for incidents of child labor, and measures taken to contribute to the effective abolition of child labor Operations and significant suppliers identified as having significant risk for incidents of forced or compulsory labor, and c a HR07 measures to contribute to the elimination of all forms of forced or compulsory labor Social: Product Responsibility Life cycle stages in which health and safety impacts of products and services are assessed for improvement, and percentage of a PR1 c significant products and services categories subject to such procedures Type of product and service information required by procedures and percentage of significant products and services subject to such 9SP PR3 c information requirements Practices related to customer satisfaction, including results of 9SP(P) PR5 surveys measuring customer satisfaction Initiatives to enhance financial literacy by type of beneficiary 9SP FS16 Legends: c core indicator / 9F Financial Disclosure Guidelines / 9SP Mix Social Performance Indicators / 9SP(P) Partially included in Mix Social Performance Indicators / a easily known by any reporting MFI

Table 6B. List of GRI guidelines Performance Indicators in the Social category that can easily be fulfilled by MFIs.

MIS and Reporting in Microfinance

81

MIS and TBL reporting With more details on the type and quality of the information needed for TBL reporting, we can now consider the question of whether specific MIS software is needed to fulfill the requirements of the GRI guidelines for disclosure purposes. At this point, it is worth mentioning that the main distinct feature of a computerized MIS is its ability to provide crossreferenced information that would be compiled separately in conventional systems. For instance, information regarding the MFI’s clients can be recorded in a simple spreadsheet storing—for example, both personal data and data related to the particular products and services used by the client. However, the tabular form of a spreadsheet is not the best way to store information regarding the installment repayments of a microcredit, or the diverse amounts versed in a voluntary savings account the client could have. Instead, such information should be recorded by the accounting system of the organization. However, without the information integration provided by some MIS software, obtaining the cross-reference information required to track down missed repayments, for instance, is not an easy task (Ivatury, 2004). Therefore, the above question can be reformulated in terms of the amount of cross-referenced information that should be disclosed using a TBL approach. A review of Tables 5 (profile disclosures) and 6 (performance indicators) reveals that the most of the information required for quantitative reporting can be obtained without cross-referencing. By its nature, the requested disclosure on the profile of the organization is mostly qualitative, with little quantitative information. Indeed, only the scale of the reporting organization (2.8 in Table 5A, detailing number of employees, net sales/revenues, quantity of products/services provided) is required for Level C reporting, whereas more experienced reporting organizations should include details related to the stakeholder engagement (4.16 also in Table 5B, requiring the frequency of engagement broken down by type and stakeholder group). In both cases, those particulars can be easily reported without a computerized MIS. Quantification of specific performance targets, including targets for the next period and mid-term objectives and goals, would be required within the Strategy & Analysis section (1.2 in Table 5A). The specifics of this quantification are, nevertheless, up to the organization6, and should depend on the mission and goals of the MFI. That is, the MFI should define its mission and goals, and design the system to ensure this is addressed, measured, audited and reported.

82

Chapter Four

In turn, a number of performance indicators, included in Tables 6A and 6B, can be disclosed without the use of specific MIS software. Among them, we can mention most of the economic indicators, which the accounting system of the organization should be able to easily provide. The exception is EC09, which is non-quantitative. Similarly, the two environmental indicators in Table 6A are largely descriptive, although some quantitative report should be available. Again, the need for cross-referencing is very limited. If, for instance, the institution offers lending lines linked to alternative energies (one of the questions included in Table 3, social performance process indicators reported to the MIX Market), quantitative details related to the number and amount of loans disbursed should be easily available and straightforward to report. Only if there are exceptions, and if there are many, is a new “field” required for this purpose. A similar argument exists for the need of MIS for social performance disclosure. Of the six labor practices indicators included in Table 6B, LA03 is qualitative, whereas LA01, LA02 and LA13 simply require the basic data on human resources, including the governance bodies of the organization. LA14 would require the salaries broken down by category and gender, and LA10 cross-references with the organization’s training programs for employees. In both cases, information is relatively easy to report. This is also the case for the FS13 society indicator, which requires specific data concerning the points of access. A number of them could be in low-populated or economically disadvantaged areas (broken down by region and by type of access), compared to the total number offered by the MFI and changes during the reporting period. Once again, this can be obtained without cross-referencing data from diverse sources. Finally, it is sufficient to say that the remaining indicators, on society, human rights and product responsibility, are non-quantitative in nature, thus not requiring the MIS of the organization. We can therefore conclude that, per se, there is no obvious need of a specific customization in the MIS software for a MFI to be able to report on TBL. Indeed, using the GRI guidelines, both Levels B and C can be obtained by disclosure of basic quantitative information regarding the MFI’s economic, social and environmental performance. Such quantitative data should be accompanied by an informed account of the mechanisms and processes established by the organization to check on its mission and goals, and to regularly update them using an informed decision making process. Nonetheless, advanced TBL reporting would also benefit from accurate cross-referencing of information. Therefore, it would be advisable to consider international standards of TBL reporting when designing or customizing the organization’s MIS, to take into account details regarding

MIS and Reporting in Microfinance

83

human rights or environment, for instance, that the MFI considers significant and would later be willing to disclose.

Final remarks Although social and TBL reporting is not yet widespread in the microfinance industry, our study shows that the actual practices and standards so far established are well in line with international standards, not only in financial reporting, but also with regard to the corporate social responsibility report or TBL report. What is also important is that the more recent developments in the sector, such as the social reporting jointly proposed by CERISE and MIX (see footnote 1) adds to the MFIs’ ability to easily report on more TBL performance indicators following the aforementioned mentioned international standards. All in all, MFIs should seriously consider TBL reporting: not only because they can easily provide this information, but also because they can offer further reflection on the MFI’s own management, on terms that may be helpful in enhancing its performance according to its mission, and to be transparent and accountable to its stakeholders. Our study has also shown that, for non-advanced levels, new MIS software is actually not required for TBL disclosure. Nonetheless, MFIs should be encouraged to have an efficient tool for information management, and therefore a computerized MIS. Although the main use of information systems is the enhancement of the MFI’s operative efficiency, its reporting capabilities can save precious time that can be more fruitfully invested in information analysis for general improvement. As a final conclusion we shall say that TBL reporting is in fact a very useful exercise in transparency and accountability to the stakeholders of MFIs. For that reason, its adoption as a standard in the microfinance industry should be seriously considered in the near future. Besides, the lack of advanced MIS software should not be a barrier to effective TBL reporting, although training, audits or consultancy support on how to record, analyze and report the additional data could be needed to help initiate the work.

References Allet, M. (2011). Microfinance & Environment: Why do microfinance institutions go green? Paper presented at the Second European Research Conference on Microfinance, Gröningen.

84

Chapter Four

www.rug.nl/research/Globalisation-Studies-Groningen/Research/ ConferencesAndSeminars/Conferences/EUmicrofinconf2011/Papers/8. Allet.format.doc Allet, M. (2012). Measuring the Environmental Performance of Microfinance: A New Tool. Cost Management, 26(2), pp. 6-17. Archel, P., Fernández, M. & Larrinaga, C. (2008). The Organizational and Operational Boundaries of Triple Bottom Line Reporting: A Survey. Environmental Management, 41, pp. 106-117. Christofi, A., Christofi, P. & Sisaye, S. (2012). Corporate sustainability: historical development and reporting practices. Management Research Review, 35(2), pp. 157-172. CGAP (2007). “Beyond good intentions: measuring the social performance of microfinance institutions”, CGAP Focus Note No. 41, CGAP, Washington, DC. Elkington, J. (1997). Cannibals with Forks: The Triple Bottom Line of 21th Century Business, New York Society Publishers, USA. Gaul, S. (2012). Diamonds no longer forever. MicroBank Bulletin, 4. Available at: http://www.themix.org/publications/microbanking-bulletin /2012/04/diamonds-no-longer-forever Gray, R. & Milne, M. (2004). Towards Reporting on the Triple Bottom Line: Mirages, Methods and Myths. In Henriques, A. (Ed), Triple Bottom Line: Does It All Add Up? Assessing the Sustainability of Business and CSR. Earthscan Publications, London. GRI (2011). RG. Sustainability Reporting Guidelines. Version 3.1. GRI. Available at https://www.globalreporting.org/resourcelibrary/G3.1Guidelines-Incl-Technical-Protocol.pdf Gutiérrez-Nieto, B., Fuertes-Callén, Y. & Serrano-Cinca, C. (2008). Internet reporting in microfinance institutions. Online Information Review, 32(3) pp. 415 – 436. Hartarska, V. (2009). The impact of outside control in microfinance. Managerial Finance, 35 (12), pp. 975 – 989. Hes, T., Srnec, K., Drasarová, M. & Neradová, A. (2012). Proposal for establishing an environmental, social, and governance (ESG) groundwork: Creating a closed system within the microfinance sector. Journal of Internet Banking and Commerce, 17(1) (http://www.arraydev.com/commerce/jibc/) Ivatury, G. (2004). Harnessing technology to transform financial services for the poor. Small Enterprise Development, 14(4), pp. 25-30. KPMG (2005). KPMG International Survey of Corporate Sustainability Reporting 2005. KPMG, Holland.

MIS and Reporting in Microfinance

85

KPMG (2011). KPMG International Survey of Corporate Sustainability Reporting 2011. KPMG, Holland. Available at: http://www.kpmg.com/PT/pt/IssuesAndInsights/Documents/corporateresponsibility2011.pdf Logsdon, M.J. & Llewellyn, P.G. (2000). Expanding accountability to stakeholders: trends and predictions. Business and Society Review, 105, pp. 419-435. Mintz, S.M. (2011). Triple Bottom Line Reporting for CPA’s. Challenges and Opportunities in Social Accounting. The CPA Journal, December 2011, pp. 26-33. MIX (2011). MIX Social Performance Indicators. Available at http://www.themix.org/social-performance/Indicators Pistelli, M., Simanowitz, A. & Thiel, V. (2011). Does Social Performance Data Support the Industry’s Social Mission Claims? A survey of 405 MFIs reporting to MIX in 2009-2010. MicroBanking Bulletin. http://www.themix.org/publications/microbankingbulletin/2011/07/does-social-peformance-data-support-industryclaims#ixzz2HhDtH6z5 Rasche, A. & Esser. D.E. (2006). From stakeholders management to stakeholder accountability. Journal of Business Ethics, 65, pp. 251267. Repetto, R. (2005). Protecting investors and the environment through financial disclosure. Utilities Policy, 13, pp. 51-68. Rosenberg, R., Mwangi, P., Christen, R.P. & Nasr, M. (2003). Microfinance Consensus Guidelines. Disclosure Guidelines for Financial Reporting by Microfinance Institutions, 2nd ed., Consultative Group to Assist the Poorest (CGAP), World Bank Group, Washington, DC. SEEP (2005). Measuring Performance of Microfinance Institutions, A Framework for Reporting, Analysis and Monitoring. SEEP, March. —. (2010). Pocket Guide to the Microfinance Financial Reporting. Standards Measuring Financial Performance of Microfinance Institutions. SEEP, October. Sherman, W.R. (2012). The Triple Bottom Line: The Reporting of “Doing Well” & “Doing Good”. The Journal of Applied Business Research, 28(49, pp. 673-681. Sinha, F. (2006). Social Rating and Social Performance Reporting in Microfinance. EDA/M-Cril, Argidius, and the SEEP Network, Washington, DC. Skouloudis, A., Evangelinos, K. & Kourmousis, F. (2009). Development of an Evaluation Methodology for Triple Bottom Line Reports Using

86

Chapter Four

International Standards on Reporting. Environmental Management, 44, pp. 298-311. SPTF (2012). Universal Standards for Social Performance Management. SPTF, http://www.sptf.info/images/designed%20usspm%20manual%2010%2 015%2012.pdf SPTF User Review (2009). User Review: Global Reporting Initiative (GRI), 1, no. 5, July. Available at http://www.microfinancegateway.org/p/site/m//template.rc/1.9.42855 Stephens, B. (2012). Information Overload: can technology address MFIs' reporting burden? MicroBanking Bulletin, www.themix.org/publications/microbankingbulletin/2012/04/technology-address-MFI-reporting-burden Tulchin, D., Sassman, R. & Wolkomir, E. (2009). New financial Ratios for Microfinance Reporting. MicroBanking Bulletin, 19, pp. 30-38. van Greunin, H., Gallardo, J. & Randhawa, B. (1999). A Framework for Regulating Microfinance Institutions. Policy Research Working Paper 2061. The World Bank, February. Zeller M., Lapenu C., Greely M. (2003). Social Performance Indicators Initiative (SPI). Final Report. Available at http://www.cerise-micro finance.org/publication/pdf/impact/SPI-summary.pdf

Notes 1

As at November 2012. According to Pisteli et al. (2011), Latin America is the most active region reporting social performance data: 51% of MFIs reporting financial data to MIX Market in 2009-2010 also reported social data. 3 Reporting on Social Performance indicators is by no means a case closed. Building on the MIX SP indicators recommended by the SPTF and using an SPI audit tool originally developed by CERISE, a new tool has been updated in collaboration with the MIX Market in mid-2012. It allows the MFIs to further report on social performance, adding new questions and performance categories to the previous existing ones. For further details, see http://www.cerise-micro finance.org/spip.php?page=article&id_article=302. 4 https://www.globalreporting.org/reporting/reporting-support/support/Pages/ default.aspx 5 Performance indicators in italics relates to data that would be available using the new form of SP produced jointly by MIX market and Cerise (see note 1). 6 In fact, 1.2 in Table 5A deals with a very difficult exercise on risk management that very few big companies can afford. 2

CHAPTER FIVE AN EXPLORATORY ASSESSMENT OF CUSTOMER INTELLIGENCE INFORMATION SYSTEM IN MICROFINANCE TRANSACTIONS: EVIDENCE FROM INDIA DJAMCHID ASSADI, SHARAM ALIJANI AND SATCHIDANANDA SOGALA

Introduction There are about 2.5 billion people worldwide who lack access to banking services. More than 80% of adults in sub-Saharan Africa, 67% in the Middle East, 65% in Latin America, 59% in East Asia and the South East, 58% in South Asia, 43% in Eastern Europe and Central Asia and 8% in advanced countries are considered as unbanked (Chaia et al., 2009). The economically-active poor are often denied access to a broad array of financial services that are provided by the institutional banking system. In an increasingly global and networked economy, many unbanked individuals are denied traditional banking services, being considered high risk, low return or both. The size of loans, the cost of loan management and the risk associated with the loan (Akula, 2010) are often mentioned to explain the reluctance of conventional banks to engage in microfinance. Also, the inability of the economically-active poor to provide sufficient collateral is viewed as a risk factor. Most studies have demonstrated that the cost of administrating small loans in informal markets is higher owing to higher repayment risks (Bottomley, 1964, 1975). Critics refer to the prohibitive cost of providing microloans in distant rural areas as an impediment to economic development.

88

Chapter Five

Consequently, the economically-active poor and the unbanked populations are pushed to operate outside formal or official banking networks. In informal money markets, moneylenders provide loans at rates well above those offered by the commercial banks. By setting a high interest rate to cover the insolvency risk associated with the economically-active poor, unofficial moneylenders operate within quasi-monopolistic economic markets. Beyond moneylenders, some informal credit markets favor social group-based lending practices. The key contribution of the Grameen Bank has been to popularize the practice of group lending without a requirement for material collateral. By contrast, informal moneylenders make no qualms about generating hefty profits by exploiting the poor and the unbanked, situated as they are at the bottom of the social pyramid (Prahalad, 2009). The combined effects of network exclusion, monopolistic market structure, and credit layering tend to increase the transaction costs of banking operations in what may be described as informal and underdeveloped financial markets (Adams and Vogel, 1986). Social collateral, group-based loans and associated technologies provide alternative approaches to counterbalance informal monopolies. As poor borrowers have few assets it may be considered as impractical or immoral to request material collateral from them. Therefore loans to the poor usually cannot be secured. However, social collateral can be considered to be the collateral of the poor and used as a substitute for asset-based collateral. The example of the Grameen Bank—founded in 1976 and later transformed into a multimillion member organization— clearly demonstrates how group-based lending practices can be used to lower microcredit moral hazard risks, and turn micro ventures into lucrative operations for stakeholders. Indonesia’s Bank Rakyat and Bolivia’s Banco-Sol are further examples of microfinance institutions that have been successful in fighting poverty through market-organized microcredit in areas of high poverty (Robinson, 2001; Khandker, 1998). The use of information and communication technologies (ICT) could also help build network access that would, in turn, ensure affordable financial services to the poor. In this regard, technology can help lower the cost of financial services and, by the same token, accelerate the social integration of the economically-active poor. While ICT can be used to gather customer intelligence, the expansion of financial networks is likely to empower the poor to engage in socially rewarding activities.

An Exploratory Assessment of Customer Intelligence Information

89

This paper argues that the development of customer intelligence databases lowers the transaction cost of financial intermediation for both regionally and nationally-based microcredit organizations.

Microcredit and the Transaction Cost Conundrum Opponents of microcredit programs argue that institutional programs divert poor clients’ economic actions and financial resources into a host of micro-business activities that increase their usage of, and dependency on, high cost credit (Bouman and Hospes, 1994). A number of studies have underlined that the transaction costs of microloans tend to be higher due to the complex bureaucratic procedures of microfinance organizations (Adams et al., 1984, Ahmed, 1989; Floro and Yotopoulos, 1991). As a consequence, only a small percentage of the borrowers at the bottom of the social pyramid have access to the full range of loans, such as those used for self-employment and small entrepreneurial activities. The question of whether or not profitable microfinance institutions (MFIs) can fully replace informal moneylenders seems to have been closely associated with the extent of credit delivery programs and customer intelligence. In this regard, the example provided by the Reserve Bank of India (RBI) is particularly noteworthy. The growth of commercial banking in rural India has significantly reduced the dependence of the economicallyactive poor on the local moneylenders, as is evidenced by the decline in the use of informal lenders in last two decades. The dependence on informal moneylenders in India dropped from a 68% peak in 1971 to a 36% low in 1991. Today, approximately one fifth of the population in rural India has access to commercial banks’ saving and funding services. In most cases, credit delivery procedures take into account the size of the loan, customer creditworthiness of the new and repeat borrowers, as well as the professional and family groups and other categories to which borrowers belong. Like the traditional funding markets, microcredit markets are characterized by information asymmetry. This means that a microloan market is likely to be subjected to economic rationing due to the fact that lenders are not able to assess the risk associated with each borrower. As a result, moneylenders are inclined to formulate loan contract terms designed to induce the borrower to take actions that are primarily in the interest of the lender (Stiglitz and Weiss, 1981, p. 381). In markets that have been abandoned by the commercial banks and dominated by monopolistic moneylenders, high interest rates are imposed by the lender

90

Chapter Five

and often accepted by the borrowers, usually because of an absence of any alternatives.

The Use of Customer Intelligence in Microfinance Operations The use of technology in building effective customer-supplier relationship is often viewed as a cost-reducing mechanism in the financial services sector. One should be reminded that the use of telegraph technology in the last quarter of the nineteenth century and the use of the telephone at the turn of twentieth century considerably increased the New York Stock Exchange’s capacity for price quotation at a lower cost (Domowitz and Steil, 2002). The development of scalable databases has enabled the banking system to improve its overall institutional oversight of lending and borrowing activities and boost financial transactions through enhanced customer knowledge and through peer-to-peer information exchange. Electronic intelligence systems would further provide support for implementing customer tracking and customer segmentation models. Such systems would also allow differential product positioning and provide programs and tools for measurement of customer delinquency. It is considered good practice for microfinance operators to evaluate the impact of information technology (IT) when designing, implementing and evaluating microcredit information systems (MCIS) (Ashta, 2011). Among the numerous cost-reducing factors for microfinance institutions (MFI), one may refer to the pivotal role of mobile telephone technology in reinforcing transaction traceability and reliability. Assadi and Cudi (2011) have demonstrated how mobile telephony can be used to reduce the cost of distributing financial services in developing countries through support for voice and payment services to the unbanked in rural and urban areas. Similarly, electronic communication networks enable microfinance institutions to match and limit orders at the highest bid and lowest offer price. This has traditionally been the case in stock exchange markets where trading costs have dropped substantially in both over the counter (OTC) and listed stocks as a result of automated operations (Domowitz and Steil, 1999; Domowitz et al., 2001; Domowitz and Steil, 2002). The use of the Internet as a general purpose technology in banking intermediation for balance inquiry, funds transfers, cash management, bill payments, and stock brokerage has constituted a major factor in boosting both productivity and organizational efficacy. Similarly, the implementation of ICT within the microcredit sector can help to monitor and enforce contracts, while offering control mechanisms for peer pressure

An Exploratory Assessment of Customer Intelligence Information

91

and group-based borrowing oversight (Stiglitz, 1990; Varian, 1990). In this way, MCISs can strengthen trust-based relationships and transactions among a host of group-based lenders and borrowers. Information systems (IS) have developed considerably since the 1990s owing to “front office” technology, for sales, marketing and customer service (Ryals et al., 2000). More recently, Internet technology has lowered the costs of electronic data interchange, and advanced the notion of the extended enterprise and its partners within the supply chain (Ovans, 2000; Lawrence, 2001). In this regard, Daniel et al. (2003) believe that the most fundamental benefits of IT result from the gathering and integration of intelligence about the customer from diverse sources to enable tailoring of products and services. Previous piecemeal systems have been progressively replaced by customer relationship management (CRM) suites to provide a unified view of the customer. This trend towards customer databases has been substantially influenced by relationship marketing theory. Reichheld and Sasser (1990) showed how customer retention influenced profitability. Later studies showed not only that customer retention costs less than customer acquisition, but also the fact that longer relationships with customers increase profitability more significantly (Reichheld, 1996). Managing a relationship necessitates sharing knowledge of customers amongst all those who deal with them, and hence leans heavily on IT. For Daniel et al. (2003) the ultimate objective is to define markets: quantifying the value of each customer segment, communicating this value within the organization in order to capture it correctly, delivering this value to customers, and measuring the value actually delivered. By designing customer intelligence systems to store and retrieve scalable information, microfinance institutions can develop their core competences through which they can build and sustain competitive advantage (Prahalad and Hamel, 1990; Teece et al., 1997). By developing a knowledge-based approach (KBV) to microcredit services, microfinance institutions can effectively build new strategic market resources whose exploration and exploitation would enable them to reduce further the cost of their loans. This would in turn allow MFIs to build the dynamic capabilities needed to cope with change, while keeping the cost of change low. Dynamic capability is defined as the capacity purposefully to create, extend, and modify the firm’s resource base (Helfat et al., 2007; Wernerfelt, 1984; Barney, 1991). In the case of microcredit institutions, transaction costs are likely to be reduced through an ongoing process of acquisition and exploration of critical data. Thus, MCIS can enable microfinance institutions to exploit effectively a wide array of data

92

Chapter Five

regarding microcredit customers, their prior experience, loan repayment record, age, gender, and family group activity.

Designing, Building, and Maintaining a MCIS A MCIS involves the design and implementation of the appropriate network infrastructure in order to collect, store, and retrieve a wide array of financial and marketing data. The infrastructure may be used by different operators including, but not limited to, commercial banks, cooperative institutions, financing agencies, public and private donors, regional, national and international agencies. The MCIS should reflect information integrity, traceability, and reliability for all stakeholders involved in microcredit operations (Laudon and Laudon, 2009). Many financial intermediaries that operate in rural areas have sought to improve their decision-making capabilities by using some sort of MCIS and data interchange. The MCIS rely on a growing volume of customer data storage and retrieval as well as large stocks of investment and financial data to facilitate the process of loan selection and delivery. For this purpose, microcredit institutions are increasingly using online analytical processing and support tools for selection and delivery purposes. Descriptive analytics refer to system functions such data visualization, OLAP (Online Analytical Processing), published reports, scorecards, and SQL (Structured Query Language) queries. Predictive analytics provide decision support tools such as decision trees and neural networks. Finally, optimization analytics are used for mathematical and simulation programming for scientifically-oriented organizations. The adoption of a flow approach can help improve the workflow between different categories of lending institutions and borrowers. When building a MCIS platform, a number of strategic options and operational characteristics need to be addressed and implemented by the financial institutions: 1) Peculiarity: The platform should offer idiosyncratic strategic objectives and appropriate action plans. 2) Universality: The system needs to be operational across the firm’s boundary so that all stakeholders (institutional banks, donors, microfinance institutions) clearly know the rules of engagement. 3) Interoperability: The proposed MCIS should provide multiple technological options such as personal computers, mobile systems, offline and online access modes, and relational databases to ensure system sustainability.

An Exploratory Assessment of Customer Intelligence Information

93

4) Comprehensiveness: The system should support the customer life cycle including profiling, enrolment, financial monitoring and service options. The system must adopt a holistic market approach market whereby customers (e.g. farmers) are evaluated on the basis of their lifecycle customer value. The proposed system should be able to register and measure the data relative to all transactions at both front and back-ends including fingerprint-based biometric ATMs (Automated Teller Machine). 5) Adaptability: As an agile system, the proposed MCIS must be scalable to support dynamic adaptation to market contingencies. 6) Cost effectiveness: Once implemented, the MCIS must reduce overall transaction costs to achieve a low-cost and effective solution for the largest possible group of users. 7) Security: the MCISs should support full delivery and comprehensive payment systems checking for multiple risk scenarios including measures against identity theft, financial fraud, credit user information and credit rating risks. When designing and implementing a MCIS platform, the following questions on system redundancies and dysfunction should be asked: 1) What resources and processes would have to be put in place so that data quality can be maintained over time? 2) In which ways will the implementation of the MCIS help enhance the overall cost effectiveness and organizational capabilities through cross-pollination of user groups? 3) How could microcredit institutions optimize the cross-functional tasks and the flow dynamics within and across their different constituencies? The following diagram sheds further light on the design and implementation of a MCIS. The system uses customer information as an input to deliver microcredit facilities and financial services at minimum cost to a versatile group of economically-active poor operating in rural and urban areas.

94

Chapter Five

Figure 1: MC CIS and Custom mer Intelligence Capability. Souurce: Authors

A MCIS M for Rural R Indiaa ISs occuupy a central place in the production an and delivery of o a wide array of finaancial servicees. They enablle the bankingg system to streamline its product offering annd provide a dynamic ppricing mechaanism to customers, tthus loweringg the informattion asymmettry between borrowers b and lenders,, and reducingg the risks and d the transactioon costs in miicrocredit distribution.. More speecifically, th he implemenntation of customer intelligence systems that select custom mers and buildd customer rellationship managemennt constitutes the cornersto one of an effficacious MC CIS. The collection oof data on rurral population ns, for governnance and commercial purposes, coonstitutes onee of the many well-establishhed features of o MCIS. In the casee of India, considering c th he size and diversity of the rural population, the collectioon, processing g, and analyysis of custom mer data constitutes a daunting taask. This papeer provides ann MCIS mod del whose principal feaatures would enable a microfinance opeerator to colleect, store, update, conssolidate and process p data for f customizedd offering. Th he system would allow w the use of PDAs, P laptopss and smart phhones to captu ure, store and retrieve voice, imagee and text dataa. The use of a digital cameera would further facilitate the identtification of bo orrowers and validate the process p of information handling annd treatment. The borrow wers’ voice may be recorded at tthe time of reggistration for authenticationn. The system m manages a data centeer wherein all collected datta are stored aand synchronized with the central sserver (Microssoft SQL Serv ver Compact E Edition).

An E Exploratory Asssessment of Cusstomer Intelligeence Informatio on

95

CIS Server and Network N Infrasttructure. Sourcee: Authors Figure 2: MC

Credit raating of the ruural population n constitutes another important step when designning a microcrredit IS. The general g inform mation infrastructure is separated frrom the transactional datab base in orderr to make datta centers distinct from m the manageement of the rural infrastrructure. The proposed system will act as a onee-stop custom mer informatioon center thrrough the Internet for assessing thee credit-worth hiness of the cclient. The sy ystem has the ability tto operate ussing scalable bandwidth, aas well as monitoring m screens, withh ‘save and suubmit’ button ns. The generaation of reportts in local languages oor dialects and a the use of a text-too-voice conveerter and synthesizer constitute anoother important feature of the MCIS thaat is used to create a nation-wide information database. d Thee implementaation of a comprehenssive database with detaileed informatioon on more than t 700 million ruraal dwellers will w allow ban nks, public auuthorities, co ooperative institutions, planning annd developm ment agenciess, private bu usinesses, international agencies, ass well as oth her service annd utility prov viders, to customize thheir value offfering. Figure 2 provides a detailed pictu ure of the PDA appliccation and SQ QL server useed for microccredit servicess in rural India.

96

Chapter Five

A Pilot Study in the Bangalore Area In order to test the model of customer intelligence-based IS, a pilot study was implemented in selected villages of the backward Honavar block, about 500 kilometers away from Bangalore, the state capital of Karnataka in India. Prior to the field study, data about demography (population, earning power and social standing), infrastructure (transport, banks, cooperatives and other extension services) and patterns of crop production were collected and analyzed. A sample of rural adults was composed according to a wide array of personal details (name, age, address, date of birth and other), educational qualification, overall household income and income tax, land ownership and liabilities and assets details. The interviewee’s consent for the study was obtained against her/his signature as well as left thumb impression in all cases. Interviews were stored in digital form. The pilot project study was funded by Microsoft and sponsored by a consortium of technology companies and service providers such as the State Bank of India, Syndicate Bank, local government authorities, local non-governmental agency, and the self-help groups and co-operatives. The study was conducted by the Centre for Banking and Information Technology (CBIT) of the International Institute of Information Technology in India. AC Nielsen (ACN) was chosen for the sourcing and validation of the data and the hosting of the solution. As a measure of quality assurance, a group of local bankers led by the head of the bankers’ forum in the district, leading bank officers and a representative of the regulatory and development authority were chosen to review and verify the data and indicate any deficiencies or errors for remediation. Banks' representatives were further consulted for the identification of the data and the documents submitted by the borrowers. The answers were corroborated through documentary evidence (photo ID, driving license, etc.). The interviews were stored and processed in the system for delivering credit rating lists and other pertinent information. Figure 3 provides a model for a mobile data center for rural India. A data collection agent collects and stores the data collected in the field using a variety of networks (PSTN, ISDN, satellite, etc.) and networking technologies including wireless systems. Bank officers will have the possibility of using the data when verifying and authenticating data, credit rating, and preparing and validating microcredit applications.

An E Exploratory Asssessment of Cusstomer Intelligeence Informatio on

97

R India. Sou urce: Authors Figure 3: Mobbile MCIS in Rural

Imp plications foor the Custo omer Intelliggence Model The centtral idea of thhe pilot study was to test th the ability of an MCIS model to prrovide basic information i and a automatedd credit rating lists in digital form m in a cost-eefficient mann ner. The creddit rating mo odel used contained paarameters succh as personal assets and iincome, bank payment history, creddit history, overdue amountts, outstandingg loans and crredit type. These param meters have beeen further broken down innto a number of o critical factors. Forr example, the t payment history covvered the nu umber of installmentss paid, numberr of defaults in i the past annd the period of o default and other neegative publicc records. The credit historyy included tim me periods since the acccount was oppen and the tiime since lastt activity. In the t credit type, the totaal number of accounts a and types of accouunts held by individual borrowers w were captureed. The ratin ng system uused was th he expert judgment-baased rating. Thhe purpose was to arrive att a credit scorre that the banks couldd use to evaluaate the overalll creditworthinness of each individual borrower annd the interest rate to be charged c in acccordance with h the risk level of thaat borrower. As A discussed earlier, custom mer intelligen nce-based ISs can draamatically redduce transactiion costs in ddelivering miicrocredit products andd services. Hoowever, for im mplementing ccost-efficient ISs, I MFIs

98

Chapter Five

should take into account four key factors: ‘leadership’, ‘infrastructure’, ‘value’, and ‘skill’. Leadership pertains to organization planning and strategic alignment. Skill underscores the need for user competencies and training. Infrastructure encompasses a wide range of system variables including hardware, software, data standards and testing. Value shows the degree to which organizations are committed to dedicating their financial resources to build and sustain a robust IS. It should be noted that in the case of MCIS, customer intelligence constitutes a key factor in sustaining a firm’s competitive advantage (Moss and Atre, 2003; Turban et al., 2010; Gray, 2010; Gonzales, 2004).

Conclusion and Future Research Axis This chapter has aimed to demonstrate that the use of digital ISs for customer intelligence and relationship management constitutes a major step in reducing the cost of building robust MCIS, and consequently the costs of transactions in both rural and urban areas. The example provided in this study was based on a pilot project in rural India. It was argued that the development of banking services to rural populations constitutes a precondition for accelerating growth and increasing income in povertystricken areas of India. Access to utility services that increasingly use automated payment systems represents another extension of customer intelligence-based models. Similarly, the delivery of education and healthrelated services to the economically-active poor could be facilitated using similar databases. More empirical work would be needed to study various cost-reduction mechanisms. The suggested MCIS needs to be scaled-up by integrating more advanced functions for the MFIs, institutional banks and social lenders.

References Akula, Vikram. 2010. A Fistful of Rice: My Unexpected Quest to End Poverty through Profitability. Harvard Business Press. Assadi, Djamchid, and Anaïs Cudi. 2011. "Le potentiel d’inclusion financière du ‘Mobile Banking’ Uneétudeexploratoire." Revue Management & Avenir, décembre, no. 46. Assadi, Djamchid, and Meredith Hudson. 2011. “Marketing Analysis of Emerging Peer-to-Peer Micro-lending Websites.” In Ashta, Arvind (Ed.). Advanced technologies for microfinance: Solutions and challenges, Hershey, PA: IGI Global.

An Exploratory Assessment of Customer Intelligence Information

99

Barney, Jay B. 1991. “Firm Resources and Sustained Competitive Advantage.” Journal of Management no. 17(1): 99-120. Chaia, Alberto, Aparna Dalal, Tony Goland, Maria Jose Gonzalez, Jonathan Morduch, and Robert Schiff. 2009. Half of the World is Unbanked. October, Financial Access Initiative Framing Note, The Financial Access Initiative (FAI). Daniel, E. Wilson H. and McDonald M. (2003), “Towards a map of marketing information systems: an inductive study”, European Journal of Marketing, Vol. 37 No.5/6. Gonzales, Michael L. 2011. “Success Factors for Business Intelligence and Data Warehousing Maturity and Competitive Advantage.” Business Intelligence Journal no. 16: 22-29. Gray, Paul. 2010. “Competitive Intelligence.” Business Intelligence Journal no. 15: 31-37. Helfat, Constance E., et al. 2007. Dynamic Capabilities: Understanding Strategic Change in Organizations. New York: Wiley-Blackwell. Laudon, Kenneth C., and Jane P. Laudon. 2009. Management Information Systems: Managing the Digital Firm. New York: Prentice Hall. Moss, Larissa T., and Shaku Atre. 2003. Business Intelligence Roadmap: The Complete Project Lifecycle for Decision-Support Applications. New York: Addison-Wesley Professional. Ovans, A. (2000), “E-procurement at Schumberger”, Harvard Business Review, May/June, pp. 21-2. Prahalad, Coimbatore K. 2009. The Fortune at the Bottom of Pyramid. Pennsylvania: Wharton School Publishing. Reichheld F. (1996). The Loyalty Effect, Harvard Business School Press, Boston, MA. Reichheld F. and Sasser W. E. Jr (1990), “Zero defections quality comes to services”, Harvard Business Review, Vol. 68, September/October pp. 105-11. Ryals, L., Knox, S. D. and Maklan, S. (2000). Customer Relationship Management (CRM): Building the Business Case, FT/ Prentice-Hall, London. Teece, David J., Gary Pisano, and Amy Shuen. 1997. “Dynamic Capabilities and Strategic Management.” Strategic Management Journal no. 18: 509-533. Turban, Efraim, et al. 2010. Business Intelligence: A Managerial Approach. New York: Prentice Hall. Wernerfelt, Birger. 1984. “The Resource-Based View of the Firm.” Strategic Management Journal no. 5(2): 171-180.

PART TWO SOFTWARE FOR MICROFINANCE MARC INGHAM

"Software for microfinance as an enabler for fostering competitive advantages, value co-creation and social innovation: the roles and strategies of software suppliers."

In his preface to this book, Arvind Ashta previews the microfinance stakeholders and how they interact. This second part provides complementary views on the roles software vendors play in supporting the activities of the microfinance institutions (MFIs). The reader will discover the number and variety of opportunities that can be grasped or created to foster competitive advantages, support shared value co-creation and contribute to enhancing and leveraging social innovations. All the contributions presented in this part focus on the challenges and obstacles that need to be faced and overcome to produce benefits, for the vendors, their clients (MFIs) and, as a consequence, the final beneficiaries. This focus pleads in favor of a systemic view and invites us to open our eyes and use a wide angle lens when it comes to dealing with the contribution of MIS solutions for microfinance. All the research studies presented in this part provide in-depth analysis of contemporary issues in the field. The journey that the authors invite us to undertake leads to the discovery of new territories. Each chapter is part of a whole, as illustrated in Figure 1, while addressing specific central questions in detail. What are the drivers of organizational buyer behaviors? (Chapter 6). What are the challenges of being an MIS service provider for microfinance? (Chapter 7). How better to understand the evolving industry for microfinance software? (Chapter 8). What about the open source attitude in microfinance? (Chapter 9). To what extent is SaaS a strategic innovation? (Chapter 10). And what are the benefits and risks of cloud computing for microfinance? (Chapter 11).

102

Part Two

Figure 1: Understanding the creation of competitive advantage

Let's glance at these rich research studies. Understanding why, when and how buyers make their decision is a prerequisite for a successful diffusion of technological solutions. Godfrey Supka (Chapter 6) presents and discusses results drawn from an analysis of organizational buyer behavior in the microfinance market. The literature review leads to the conclusion that, while there is considerable available literature on organizational buyer behavior and technology adoption in general, there is very little information available in the field of microfinance. The aim of this research is to contribute to filling this gap. The background of the study focuses on "what makes an organization buy a replacement system, and buy it now" and provides a conceptual framework, built on Kotler and Armstrong (2010), who identified different layers of influence in the purchasing response: environmental, organizational, interpersonal and individual. Semistructured interviews with open-ended questions with twenty key

Software for Microfinance

103

informants enabled, thanks to an interpretive and inductive approach, the identification of the main purchasing drivers (as they are cited by respondents) which were grouped into two groups: "external" (competition, funder, economy, regulation, technology), and "internal" (growth, the taking of deposits, system standardization, new channels, centralization). It is worth stressing that the research also identifies triggering events falling into one of the following categories (new resources, new license; new management, system failure, new entity). Increased competition and the influence of funders, investors, and donors range at the top of external drivers cited by respondents. Growth is cited as the main internal driver. As Godfrey Supka states, "Above all, the results highlight the need for drivers to include a discrete driving event rather than just a gradual build-up of need." After having assessed the drivers of buyer behavior and the events that trigger their decisions, MIS service providers have to be aware of the challenges and obstacles that need to be faced and overcome, arising from the specificities of the contexts in which they operate. Gaurav Sinha (Chapter 7) sheds lights on these issues. Adopting an "ecosystem" approach, the literature review led to the development of a conceptual theoretical framework combining a set of external challenges (government policies and norms, client level issues, market) and internal challenges (operational, resources-related). The three case study analyses conducted in the context of India are documented, thanks to the collection and interpretation of qualitative data (documents and interviews). The chapter presents and discusses how these challenges can be tackled or converted into opportunities by adopting various strategies, such as sector diversification to reduce sector-specific risks, geographic diversification to absorb losses incurred in one region, and managing human resources efficiently and strengthening relationships with clients, especially through timely after-sales service. As Gaurav Sinha writes "The paper provides an opportunity to learn from the experience of MIS vendors in the microfinance sector. It further attempts to build a case for policy or funding support to these vendors, which are playing enablers’ roles in the process of greater financial inclusion." A third, central issue for vendors is the need to provide MFIs with products that meet their various needs. This capability is at the core of a successful strategy. Arvind Ashta, Vitalie Bumacov, Mikhail Cherkas and Dinos Constantinou (Chapter 8) present and discuss findings from research on a

104

Part Two

sample of twenty-five MIS products from the CGAP database. They identify evolutions through the analysis of different versions of these systems. The chapter provides a brief literature review on innovations, focusing on "incremental" ones, which form the bulk of innovations in this industry. It also outlines the main evolutions in the microfinance industry. The findings show that "the comparison of these products brings out the diversity of software owing to the broad needs of microfinance market, as well as that caused by the development stage of software market for microfinance." Among the findings, the research indicates that continued presence in the market allows vendors to complete required functionalities and to distinguish themselves from competition. Moreover, newer solutions have, in general, fewer customers than the older ones. This could be explained by the kind of international strategies employed—more global for the older, local for the newer. Furthermore, differences in pricing strategies and overall costs make the software difficult to compare. Finally, the analysis of correlations between factors provides not only very rich insights on their significant interrelations, but also suggests the variety of vendors' strategies depending on their nature (for-profit or not-forprofit), and the share and scope of microfinance business in their productmarket portfolio. One of the key messages that can be drawn from the research is that "sustaining competitive advantages necessitates incremental innovation and differentiation." The second message concerns diversity in the product offered, the needs to be met, the vendors' strategies and their business models. Diversity offers opportunities for vendors to build and sustain competitive advantages in this industry, which is still in its early stage of development, fragmented and growing rapidly. This enables vendors, depending on their strategic choices and capabilities, to benefit from differentiated and "unique" positioning. They can provide MFIs with products that meet their needs and, as a consequence, be a source of competitive advantage. As the authors stress in their conclusion, "MIS vendors try to follow challenging and diverse requests that come from MFIs, an indicator of their corporate social responsiveness." Incremental innovation and differentiation are common in the market and can contribute to the creation of sustainable competitive advantages. The journey continues, thanks to the discovery of new solutions, which can be viewed as more radical, disruptive or strategic innovations. Such is the case with open source software, software as a service (SaaS), and cloud computing, all of which could help to enhance and scale the

Software for Microfinance

105

contribution microfinance can make as a social innovation. These solutions could help smaller MFIs to achieve their objectives. Vitalie Bumacov, Frederic Lanet and Arvind Ashta (Chapter 9) conducted a case study analysis at AIRDIE, a French microfinance institution, which adopted Octopus (a computerized MIS based on open source software for microfinance). They start with a literature review on the characteristics and advantages associated with such solutions, including reduced or no cost for the software, adaptability to user's needs, and support of the community, as well as the obstacles to their adoption, for example knowledge barriers, integration of legacy systems, forking (the need to maintain customized systems for each customer), sunk costs, and technological immaturity of the new solution. The chapter examines the perceptions and implications of these barriers in the context of an early adoption of this solution. The case study analysis of AIRDIE aims at providing answers to a set of research questions. Did AIRDIE innovate? What drove it to choose an MIS based on open source software? Did it voluntarily reveal its innovation through sharing its experience on its website? Can the Octopus diffusion of innovation compete with diffusion costs of commercial production and distribution? After presenting in detail the organization and its context, its information needs and the reasons why Octopus captured the attention of AIRDIE, the authors identify how AIRDIE was able not only to overcome challenges but also to turn them into opportunities. Forking, sunk costs and legacy integration were not important obstacles for AIRDIE. The lack of technical and business knowledge needed to implement the solution was overcome thanks to cooperation with the creator of Octopus. The case study offers the opportunity to learn how value can be co-created, turning a constraint into an opportunity, thanks to the adoption of an open source attitude. The authors write "the knowledge barrier played a catalytic rather than a restrictive role"(…) "Aware of the ignorance of the MFIs in Europe that OSS-based MIS was available and potentially useful for their needs, AIRDIE decided to tackle this barrier in order to open the way for others and create, with the existing and potential newcomers, a community that would solve the MIS problem by joint effort." Arvind Ashta (Chapter 10) situates his research on SaaS in microfinance in the literature on innovations. Strategic innovation is positioned in a theoretical map based on typologies and taxonomies, both incremental (architectural, modular) and radical (disruptive, sustaining).

106

Part Two

This framework enables the formulation of a set of hypotheses concerning the actors providing SaaS innovation (newcomer or incumbent); the target clients (smaller MFIs who are not being served or large MFIs); requirements (more standardization or more flexibility in financial products); and the business processes (same as or different to those of existing software suppliers). The next set of hypotheses combine several dimensions: Are the same or different customers targeted? Are the same business models and distribution networks used? Is a low pricemargin strategy used or is there no change in the overall economics? The last set of hypotheses focus on incumbents' response strategies, with questions such as whether to permit SaaS operators a business area the incumbents consider to be peripheral or to block them? Should incumbents continue with no introduction of SaaS solutions in their portfolio or improve their products by providing a SaaS option? If the latter, should they start a separate unit for their SaaS offering or keep it within the same organization? The case study research (sample of incumbents and challengers) is based on the collection and interpretation of information from websites, reports and interviews. Findings regarding incumbents' strategies and innovations show (among other factors) that pricing is very complex and price/cost ranges are broad, leading to some confusion for customers. Distribution is usually through commercial partners. For the challengers, pricing policy is often not transparent. The author identifies and discusses three main issues relating to strategic innovation: disruption in the market place, business models and blocks and challenges. As Arvind Ashta suggests, the analysis of both incumbents and challengers indicates that "SaaS may be a radical innovation and may provide a sufficiently different pricing model to permit strategic innovation, either by new entrants or by smaller existing vendors." Software as a service may offer solutions to meet the needs of categories of MFIs, especially smaller ones which are less well served. But these solutions also bear potential risks. Bryan Barnett (Chapter 11) identifies these risks and how they can be mitigated in the case of cloud computing. Starting with a detailed presentation and discussion of four types of risks (internet service interruption, data center compromised, unauthorized access to data, business failure / change of control) the author indicates the likelihood potential impact of each risk. Internet service interruption, somewhat common in developing countries, has potential impact on service levels.

Software for Microfinance

107

Data center stability and security problems are rare but each may have a very important impact with potential loss of data. Unauthorized access to data, which can be difficult to assess and detect, has potential impact in terms of financial losses and loss of customer trust. Business failure / change of control is more likely to occur in smaller, new companies with limited resources, and could lead to immediate loss of access to data and operational software. This analysis enables identification of strategies to mitigate these risks. Two major issues are discussed: due diligence and contract concerns. The chapter stresses the questions to be asked and the criteria to be used to assess risks, plus the key concerns which have to be addressed while establishing a contractual relationship: What are the terms of the agreement? Are they negotiable? Is the agreement enforceable? As Bryan Barnett states, "While the benefits of cloud-based services are genuinely compelling, there are risks as well. And if those risks should not be exaggerated, neither should they be ignored." This enjoyable and interesting journey, which the authors of these chapters invite us to follow, leads us to discover new territories, thanks to the pioneering research on which each chapter is based.

Bibliography Kotler, P. & Armstrong G. (2010) Principles of marketing. 13th edition. Pearson Prentice Hall, New Jersey.

CHAPTER SIX DRIVERS TO ACTION: ORGANIZATIONAL BUYER BEHAVIOR IN THE MICROFINANCE MANAGEMENT INFORMATION SYSTEMS MARKET GODFREY SUPKA FERN SOFTWARE LTD. UK

Introduction MIS is the strategic infrastructure for all activities of the microfinance institution. As commercial banks have found throughout the last 50 years or more, the system you have determines what you can do. All financial products and services offered must be processed and settled. All activities undertaken must be managed and reported to stakeholders and regulators. All risks must be controlled and mitigated. Just as the microfinance movement has reinvented the wheel of banking, so it has with its core management information systems. Many microfinance information systems have grown out of aid and development projects. These have added more and more traditional banking functionality as MFIs have added new banking products and services. At the same time, core banking systems vendors have woken up to the opportunity represented by microfinance and offered their systems to larger MFIs or networks of MFIs. The result of such developments has been the growth of an increasingly mature and sophisticated MIS market, supported by an infrastructure of consultants, local implementation and support partners, trainers, conferences, publications, independent ratings and assessments (Supka and Ling, 2009). There have been many reports on what features and types of systems MFIs should buy, some reports on what they do buy, but to date very little on why they buy, and specifically on the timing of their decisions.

Organizational Buyer Behavior in the Microfinance MIS Market

109

As any marketer will attest, need is not demand. Demand is only proven when contracts are signed and money changes hands. Just because a client needs a system, it doesn’t mean they will actually buy one. Many organizations in all industries and markets struggle on for many years with systems that clearly must be replaced, even when this is clear to the organization itself. Sometimes such inaction defies reason and yet it occurs. Similarly, an organization will suddenly make a new investment seemingly out of the blue. Sales management methodologies—for example, Oracle’s Target Account Selling (Oracle, 2009)—talk about the “Compelling Event”, without which the system replacement project is doomed to an eternity of discussion and delay. So far there has been little, if any, analysis of what these compelling events are in the MIS market. This chapter tries to provide some answers to that question. This question is not only of interest to MIS vendors, but to also MFIs and their stakeholders. MFIs management and staff may waste time on projects that have little chance of going ahead. Funders, regulators, rating agencies and Technical Assistance (TA) providers may struggle without success to motivate an MFI to invest in an MIS improvement that is necessary (or to restrain the MFI when it is not). A clear empirical understanding of the drivers to action would be valuable to all. This chapter examines the issues of MIS purchases using available literature on organizational buyer behavior, and trends in microfinance. It also examines publicly available market research on preferences and priorities expressed by some microfinance institutions with regard to their current MIS system. Gaps in the literature and key issues that need to be researched are identified. Primary research is then conducted on 20 specific MIS purchases to gather data to answer those questions. The research results are then placed in the context of the available frameworks for analysis to infer the main conclusions. Finally the research method is critiqued for limitations, and new questions identified for further research. It should be noted that this chapter does not examine the choices between different classes of systems or between vendors, but simply the drivers towards a decisive action.

Background While there are independent reviews of MIS systems, for example those undertaken by CGAP (MixMarket, 2012), and Accenture (2010), and some records of which MFI uses which MIS system on the MixMarket database (MixMarket, 2012), there is little information

110

Chapter Six

available on the purchasing decisions by MFIs. Closest are the occasional surveys of MIS usage and intentions, for example one conducted by the European Microfinance Network (EMN, 2011) in which 16 European MFIs responded. The EMN (2011) survey had some startling if contradictory results. According to the EMN Survey, “Two-thirds of the organizations completely lack an application to ensure that the data flows automatically from the portfolio management application to the financial accounting software without duplicate data entry.” To the outsider this might seem a critical requirement. And yet the survey also reported that, “Respondents are rather satisfied with their systems” and “Most organizations don’t think that their information systems prevent them from achieving their operational goals.” Clearly need is very subjective. The survey exhibited little evidence of urgency to replace systems. Most MFIs surveyed said they intended, “to improve or introduce technology applications in the next 1-2 and 3-5 years.” There are several guides on how to select an MIS system (Braniff and Faz, 2012; Ahmad, 2005; Mainhart, 1999). The EMN (2011) survey also offers its own guide. However, while many pages or chapters are rightly devoted to functionality, processes, technology and project management, there is very little discussion on the decision whether or not to invest, or how to ensure the decision and selection correctly aligns with external and internal business drivers. Of the group, EMN asks the most direct questions: “What do you need and why do you need it? What results do you wish to accomplish with this effort? What are the business reasons (drivers) for these changes, and how do these changes support the overall needs of the business?” Nevertheless, there is little or no information available on how these drivers affect purchasing decisions, or on which drivers are dominant.

1. Organization Buyer Behavior While there is no specific literature available on microfinance IT systems buyer behavior, much literature exists on organizational technology adoption in general. The Technology-OrganizationEnvironment (TOE) Framework (Tornatzky and Fleisher, 1990) suggests that the key influences on an organization adopting new technology are external environmental factors, such as market dynamics and government regulation, characteristics of the organization itself, such as its size, resources and the formality of its processes, and finally the nature of the

Organizational Buyer Behavior in the Microfinance MIS Market

111

technology being adopted, plus the existing technology within the organization. While the Diffusion of Innovations model (Rogers, 1995) and the Technology Adoption model (Davis, 1989) also explore the factors influencing technology adoption, they are more focused on how a particular new technology becomes adopted by a market over time, rather than the choice between competing technological solutions by a specific organization. Scott and Christiansen (1995) delve more deeply into motivation at the institutional level, going beyond rational goals such as efficiency to include cultural influences, such as desires for status and legitimacy. These drivers include a desire to conform, or follow the market. DiMaggio & Powell (1983) expand on these factors, describing “Institutional Isomorphism”, a form of collective rationalism, as a "constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions.”

Figure 1: Successive layers of influence (Kotler & Armstrong, 2010)

A more general study of organizational buyer behavior, such as that of Kotler and Armstrong (2010), takes a higher level view that encompasses all of these factors. In fact organizational buyer behavior, which can be used to examine all organizational purchases, is well suited to the specific questions of this study, which focus on what makes an organization buy a replacement system, and buy it now, rather than on the choice of a particular system, or type of system. Kotler and Armstrong (2010) explain how the buying process is undertaken by the buying center, which is part of the buying organization.

112

Chapter Six

The buying center may consist of one individual or may require the consensus of a number of stakeholders. In the case of a major system purchase, the number of stakeholders may be quite extensive. Indeed, the boundaries of the buying center may extend beyond the boundaries of the MFI to include external stakeholders, such as funders, regulators, rating agencies and TA providers. As Kotler and Armstrong (2010) explain, the organizational buyer inhabits a world of successive layers of influence, resulting in the purchasing responses of the organization, in terms of supplier and system selected, and price paid. These layers include the wider business environment and trends, the structure, scale, culture, mission, priorities and resources of the buying organization, the dynamics of the relationships in the decision making unit, and the personality, preconceptions, status and culture of each individual in the unit (Kotler & Armstrong, 2010).

2. Environment The market for MIS systems is one of derived demand, where the effect that environmental factors have on the business of the MFI results in changes in MIS system market dynamics. In the microfinance context, Rhyne and Otero summarize the key environmental drivers for microfinance growth as competition, commercial entry, technology and the policy and regulatory environment (Rhyne & Otero, 2006), which we will discuss one by one. In very mature microfinance markets (Bangladesh, India, Kenya, Uganda, most of Latin America) poor borrowers are now in the fortunate position of having competing potential lenders where 20 years ago there were none (Rhyne & Otero, 2006). This has had the effect of driving down interest rates, putting pressure on margins and increasing the range and choice of services. For MFIs to survive the competition on rates, they need to reduce costs. Commercial banks have cost ratios far lower than MFIs, partly because of differences in loan sizes between microfinance and banking, but also because of successive investment in technology to reduce operational costs. Commercial entry, as defined by Rhyne and Otero (2006), is simply another form of competition, but from commercially-motivated institutions entering a market where profitability has been proven by more traditional NGO MFIs. However, this trend is part of a wider trend towards the commercialization of microfinance (Robinson, 2002) including the trend to financial sustainability by traditional MFIs, and the entry of commercially-motivated investors in microfinance (MicroCapital,

Organizational Buyer Behavior in the Microfinance MIS Market

113

2007; CNN Money, 2008). A key significance of commercial entry is the threat that banks pose to the deposit base of MFIs taking (or planning to take) deposits. If an MFI chooses to use deposits as a source of loan funds, the danger is that its clients will deposit more of their funds with banks in order to take advantage of better services (e.g. ATMs) supported by technology offered by the banks. Over time this would erode the funding base of the MFI. There has been much discussion on the role of technology in microfinance (Ashta, 2011; EMN, 2011), but this has usually focused on front-end technologies such as mobile or branchless banking. While Rhyne and Otero describe the exciting potential for such innovative technological delivery channels, they stress that, “Breakthroughs in the use of technology require that microfinance institutions have already incorporated technology thoroughly into their operations... that is, primarily on the back end... (to) produce timely and transparent financial reports, and report on their operations as otherwise needed. These attributes will be sine qua non for future microfinance institutions.” (Rhyne & Otero, 2006). Their message is that robust MIS is the essential foundation for innovation in delivery channels. Rhyne and Otero focus on the degree to which regulatory and policy frameworks are supportive of microfinance, for example in eliminating government subsidized credit and abolishing interest rate limits. In addition to these general drivers of the scale of the industry as a whole, regulation can change the structure of the industry and the need to invest in technology. For example McKee envisages a result of recent financial turmoil to be regulators “pushing consolidation of institutions under their supervision by boosting minimum capital requirements and promoting mergers” (McKee, 2008). Such consolidation in microfinance would increase the number of MFIs with the critical mass for significant MIS investment, and provide them with a need either to combine or replace existing systems. Some form of MIS project would be unavoidable. If new services offered include deposit taking, MFIs in most jurisdictions are required to be regulated, for example acquiring licenses as banks or non-bank financial institutions (NBFIs). Not only does deposit taking require an increase in function beyond what may have been available in a lending system, but the regulators may require more robust MIS systems as a condition of granting a license. The importance of these environmental factors is that they affect the market as a whole, to a greater or lesser extent, and therefore are valuable in predicting the changes in demand for MIS systems at an aggregate level. However, they do not predict how a specific institution will react to

114

Chapter Six

the environmental stimuli. It is also necessary to consider the organizational factors—the objectives, structure, policies, procedures and systems of that organization.

3. Organization Kotler and Armstrong (2010) group the key organizational aspects of the purchasing unit into its mission or objectives, its structure, and its policies, procedures and systems. Mission, structure and systems are all inter-related, and responding to environmental changes. In response to external pressures and to changes in microfinance best practice, a larger proportion of MFIs are trying to increase scale and improve sustainability, rather than focusing on the “poorest of the poor” (Poyo & Young, 1999; Christen & Drake, 2002; Robinson, 2002). Increasing scale requires the ability to process large numbers of clients, to offer a wider range of services, and to access deposit funding. Improving financial sustainability requires increases in revenue from scale while containing or reducing costs. Helms (2007) cites two overall categories of MFI relevant to this study: semi-formal providers such as NGOs, credit unions and cooperatives, and formal institutions such as state banks, agricultural development banks, rural banks, savings and postal banks, commercial banks, and non-bank financial institutions. The former can be memberowned or run by trustees and supported by donors. The latter can be state or privately-owned. If privately-owned, shareholders may be for-profit, social mission-oriented or both (Helms, 2007). The ownership structure is likely to have a major impact on MFI strategy and all major investment decisions, as investors or donors with large or majority stakes may have influence on strategy and major expenditure, effectively bringing them into the buying center. Some of these stakeholders can be considered to be environmental, e.g. governments, ratings agencies and credit bureaus. However, some stakeholders may be so influential as to be considered to be inside the buying organization, for example a tightly-controlled network, a majority investor, or a founding donor. For the system acquisition project to be authorized there must be some resources allocated to it, whether ultimately sufficient or not. These will be influenced by the business case, but also by the availability of resources and the competing needs. Once again, system acquisition may be triggered by an increase in available resources. An increase in resources can be the

Organizational Buyer Behavior in the Microfinance MIS Market

115

driving force, for example in unleashing latent desires for expansion that are only achievable with the new resources.

4. Interpersonal and Individual The dynamics of the relationships in the decision making unit and the personality, preconceptions, status and culture of each individual in the unit will also influence the specific MIS chosen. Kotler and Armstrong (2010) and Webster and Wind (1972) proffer different views of these two layers of influencing factors. Webster and Wind focus on the buying center, the “Decision Making Unit” (DMU), describing the key actors as the users, buyers, influencers, deciders and gatekeepers (Webster & Wind, 1972). Kotler and Armstrong (2010), however, consider the interpersonal issues within the buying center, and between the center and the vendor. Both pairs of authors then converge on the individual characteristics of each actor. Kotler and Armstrong (2010) highlight the demographic characteristics such as age, education and status, plus personality and risk appetite. Webster and Wind’s (1972) individual factors are more subtle, reflecting culture, personality, experience and preconceptions, of each actor, plus the interpersonal issues between each actor and vendor staff. The last of the environmental factors, “Culture and customs” is also relevant here. Many MFIs have been established with a non-profit structure. The culture of these institutions, and the individuals working there, may experience a cultural mismatch with commercial organizations such as MIS vendors. While there is considerable literature available on organizational buyer behavior and technology adoption in general, and some practitioner information that is specific to the banking systems market, there is very little information available of any type on systems buying behavior in the microfinance market. Therefore it is necessary to undertake primary research to fill the gaps in the literature.

Methodology Research was conducted by confidential interviews with 20 people representing MFI management (9), vendor sales staff (8, representing 5 vendors) and MIS industry experts (3). The interviewees responded to questions relating to 20 different MIS purchase events over a five year period of which they had experience. In some cases a respondent had

116

Chapter Six

information or opinions on more than one MIS purchase case studied, in which case responses were aggregated by response type. Interviews were conducted by telephone and lasted 1-2 hours, with the average being 75 minutes. The interviews were all semi-structured, with very open-ended questions on the key issues highlighted in the background above, for example: x What were the key environmental trends that determined a need to change MIS system? x What were the key business needs that MIS investment was designed to address? x What effect did technological changes such as e-channels and improved Internet have on decisions? x How did the organizational aspects of the MFI affect its purchasing process? x How did changes in resource availability affect purchase events? x How influential were stakeholders in decisions, and in what way? The approach taken was interpretive and inductive, in that it investigated the issues in general and then moved to specific issues as they arose. While guided by the questions above, the objective was to get each interviewee to talk about what was important to him or her on the general topic. In keeping with the inductive approach, new interesting lines of enquiry arose as time went on. The annotated interviews were then analyzed for responses, which were then coded and grouped with similar responses. These were collated in the following topics: External drivers, internal driver and events. As the study did not use a formal questionnaire with standardized, weighted responses, no attempt was made to weight the responses in the results tables.

Findings The survey revealed purchasing drivers that could be grouped into two groups—external and internal—each containing five drivers. In some cases it was difficult to categorize. For example, was the availability of new funding external, as it came from outside the MFI, or internal, in that the funder might be part of the decision making unit? In fact, there was considerable overlap between categories and drivers.

Organizational Buyyer Behavior in n the Microfinannce MIS Markeet

117

1. Externall Drivers

Figure 2: Exteernal drivers

The main external driver d cited was w increasedd competition. Seven respondents cited increaased competiition from loocal and inteernational commercial banks. A vaariety of mark ket factors weere seen as drivers d of increased coompetition itseelf, including deregulation oor active prom motion by government, e.g. “South Africa is man ndating microofinance to commercial banks”, “Goovernments inn Latin Ameriica are pressuuring access to o credit”; and banks w waking up too the profitab bility and acccess to cheap deposits available inn the microfiinance markeet: “Major baanks are mov ving into microfinancce, on a for-prrofit basis”, “Big “ banks seee a captive market m for mobilizing ddeposits.” Three reespondents allso cited com mpetition (or potential com mpetition) from non-banking comppetitors: “Theere is a blurrring of lines between banks, phonne companiees and merchants”; “Wee foresee Telcos and supermarketts acquiring baanking licensees.” Increasedd competitionn through con nsolidation waas seen as a long-term l trend: “We anticipate coonsolidation based b on ecoonomies of sccale—e.g. MFIs with 2200,000+ clieents—with oth hers being drriven out of business.” b Such consolidation will be strengthen ned by a conntinuing trend d to high minimum caapital requirem ments in countries such as K Kenya, Tanzan nia, West Africa, Brazzil, Mexico annd Indonesia. Another important exxternal influen nce cited was the influence of either funders, invvestors, lenderrs or donors. Having H said tthat, the influ uence was

118

Chapter Six

general rather than specific. While two respondents cited cases where funders had specifically influenced the systems purchased, four respondents expressed the opinion that funder influence was limited to the decision to invest in MIS rather than system selection: “Major stakeholders, e.g. investors, lenders, networks, may want to influence technology decisions but have had little practical influence to date. In general the message is ‘you need to get yourself a good system’ rather than being specific.” However, the view of the two clients that are networks and therefore also influencing stakeholders is informative: “With those who are whollyowned or majority-owned, especially if they have a board seat, we can ask a potential partner MFI ‘Are you running on my standard system?’.” Another confirms that influence is dependent on investment: “It can be a problem to get cooperation from an MFI if we do not have a majority stake.” Vendor respondents supported the influence of funding stakeholders. Several cited examples of sales won or lost because of the influence of a major investor. One cited a project where an investor bought systems for ten partner MFIs. Enabling growth and improved risk management were the two main funder objectives mentioned for improved MIS. Investors especially want partner MFIs to have systems that are transparent, to prevent fraud, and that are effective in managing credit risk. Three respondents showed some interest by funders in improved social impact statistics, however interest was lukewarm. While some MFI respondents saw this as a useful feature, no vendor respondents reported it as a requirement in the sales. Most vendor respondents reported active involvement by specialist microfinance IT consultants. Often these were introduced by stakeholders. Recent financial turmoil was not cited as a major factor. The consensus was that the microfinance market had suffered less than the commercial banking market from the current economic turmoil because the market is less integrated into the rest of the banking system. In fact, the evidence states that macro-economic changes have been a mildly positive driver. In three cases a key driver for a MIS purchase was recently improved economic and political stability in the country in question, reflecting the fact the current financial crisis is not truly global, and many markets where microfinance is prevalent are thriving. Two respondents felt that regulators would respond to the banking crisis by increasing regulation of their microfinance industries: “Regulators may slow down innovation—MFIs may be required to

Organizational Buyyer Behavior in n the Microfinannce MIS Markeet

119

channel throough traditionnal banks.” Th his trend mighht increase thee number of MFIs thaat take bankingg licenses, or would drive demand towaards those MFIs that allready have liccenses, makin ng them more viable.

2. Internall Drivers

Figure 3: Inteernal Drivers

Perhaps not surprisinngly, a drive for growth was cited ass the top internal drivver. Howeverr this does no ot explain whhy these MFIss wish to grow and w what external factors have prompted p that at drive. As on ne expert pointed out,, “It is a chicken and egg problem, how w to justify in nvestment without scalle, how to coppe with scale without w investtment.” In som me cases, growth is prrompted by annother driver, for example tthe availabilitty of new resources. Nevertheeless, it is cleaar that the typical business ccase is based firmly on increased revenue, rather than on cost savings s for exxample. A grow wth drive was cited byy eight responndents over a wide range. Inn a notable ex xample, a group lendinng institution “Grew from 70,000 7 to 600,,000 clients in n 4 years” following innvestment in a core bankiing system. O One network saw that their investtment in MIS allowed th hem to “Driive up scale through consolidatioon, with a fasteer ramp up tim me.” The nextt most commoon internal driver was the m move to take on savings and depositss. This resultted in three new n requiremeents. First, th he system had to have the necessaryy savings fun nctionality. In some cases MFIs M had grown usingg loan trackingg systems. Sav vings requiredd some form of o system

120

Chapter Six

overhaul. The second requirement was an MIS system that would satisfy regulators. This drove several MFIs towards vendors with global reputations. The third requirement was in the technical configuration. Several MFIs had survived on decentralized systems that required manual or semi-automated updates to the center. They saw a need to upgrade to a centralized, online MIS system to maintain real-time balances available to all branches and channels in case of withdrawals. There was a consensus from clients, experts and vendors that a key emerging trend is the need to connect to electronic delivery channels. However, there was diversity in the driving forces perceived to be behind the trend. One MFI network cited the primary channels under development as 1) POS, 2) Cards, and 3) M-banking. A vendor cited the need for “GSMenabled point of sale (POS) handheld devices and biometrics for client acquisition, loan origination, payments, funds transfer” to increase outreach and offer new services. However, cost is also a driver in the desire to implement e-channels. One network cited a “Vital need to move clients from the banking hall, at $1 to $1.25 per transaction, to ATMs at 30 cents and mobile phone at 8 cents per transaction. In some cases banks will even finance the cost of the mobile phone to accelerate uptake.” The need to reduce reliance on physical premises and on cash was cited as a driver in the trend towards branchless banking: “There is a need to bring to the developing world the cashless technologies and methods of the developed world. It is dangerous for banks to have cash—for both employees and clients.” Moreover, “Increasing outreach into rural areas is more difficult. There is more off-grid electricity and poor connectivity.” However, there was also skepticism of the effectiveness of e-channels in extending outreach. “There has been excitement and some isolated examples of use of handheld devices, but now the trend may be reversing. Biometric as a useful component, but it’s not compelling. They can be very expensive devices, and paper records can be very accurate.” An expert respondent added, “Anyway, loan officers are biometric, especially in group lending,” implying that loan officers generally know their clients well enough to recognize them. Another echoed this skepticism: “While there are loads of pilots, there is very little running live now.” Importantly, experts emphasized the need for robust MIS as a basis for channels. “Channels don’t work without strong MIS. Often MFIs don’t have a good core system at the back end.” The drive towards standardization of systems was cited in the case of the four networks and one apex institution in the sample. This reflects the

Organizational Buyyer Behavior in n the Microfinannce MIS Markeet

121

nature of theeir multi-entitty business, an nd their need ffor aggregatio on of data across the neetwork. A key fi finding of the survey was the t fact that some drivers manifest themselves aas specific evvents, rather th han a graduall build-up, and d these— whether intternal or exxternal in naature—are prresent on most m MIS purchases.

3. Eveents

Figure 4: Eveents triggering purchase p

In the 200 cases studied there were 20 2 triggering events falling g into one of five categgories. Howeever, not all purchases p hadd a triggering event. In three cases tthere were tw wo triggering events—new e llicense plus either new funding (2) or new mannagement (1). In three casses there was no clear obvious trigggering event.. Interestingly y, in each of those non-ev vent cases the driver w was competitiion. The partiicular manife station of com mpetition was that thee client in queestion emulateed a close com mpetitor by pu urchasing the same syystem. Therefoore the system m purchase byy the competiitor could be seen as ann event in itseelf. The most common event e was thee arrival of nnew external funding. These new mand, enab w resources released pent-up dem bled the implementattion of ambitiious expansion n plans, or reqquired new in nvestment in systems too achieve the mission mand dated by key sstakeholders. The nexxt most comm mon event waas the transforrmation from NGO to licensed baank, in orderr to grow th hrough accesss to deposit funding.

122

Chapter Six

Although transformation is an event from one perspective, the question still remains of what prompted an MFI to take a banking license at this point. In some cases studied, the MFI reached the limit of its available funding. In other cases the MFI was prompted to take a banking license by a key stakeholder. Three other events remained with lower incidence. New management was the event in two cases: in one case just a new CTO; in another the whole management team. Clearly the drivers were already in place, but new leadership provided the necessary impetus for action. In two cases the triggering event was the failure of a previous project. Clearly this event masks the driver and event for the original purchase. Finally, in one case the client was a newly created entity.

Discussion: Key influences on buyer behavior The findings agreed with Rhyne and Otero (2006) that the key environmental driver is competition, including the entry into microfinance by non-traditional providers, especially commercial banks. The entry of commercial banks, and non-banking entrants such as Telcos and supermarkets, was viewed as a subset of a more general trend in microfinance towards commercialization, with a greater emphasis on market share or outreach and financial sustainability than necessarily targeting the poorest of the poor. Respondents also saw regulation as an important driver, either in forcing consolidation through minimum capital requirements, or by limiting product innovation to licensed banks. Moreover some respondents saw a connection between competition and policy as some governments are requiring commercial banks to offer some kind of service to poorer clients. Rhyne and Otero’s fourth factor, technology, was viewed more as internal, rather than an external, driver. While a trend towards electronic delivery channels, such as mobile phones, handheld POS devices, mobile branches and branchless banking was recognized, such innovation in the market was not cited as a driver for purchases per se. However, the desire to use such technology to achieve specific business objectives—reduce cost, increase outreach, or real-time processing for deposits—was cited as an important internal driver. It can be inferred that such internally-driven initiatives are themselves driven by the availability of new technology in the market, and its use by competitive or comparative organizations. The consensus agrees with Rhyne and Otero’s assertion that a necessary precondition to deploying e-channels is a stable and robust back office

Organizational Buyer Behavior in the Microfinance MIS Market

123

system. This driver has strong implications for the MIS market as echannels become more widespread. The three sub-categories of organizational factors cited by Kotler and Armstrong are objectives, structure, and policies, procedures and systems. The stand-out internal factor—growth—fits firmly in the first category, objectives. Clearly, the desire for growth has itself many external and internal drivers and it is advisable to dig beneath this widespread but rather obvious driver. What makes the MFI want to grow, and grow now? There was a strong correlation (7 out of 9) between the desire for growth and two events, either new funding (4) or a new license (3). However, even these two factors cannot be seen as source drivers as the decision to obtain a new license or to seek new funding were themselves driven by other unknown factors. In fact, the statements that the new MIS system was needed to fund new growth may have been postrationalization: It is likely that significant growth was required to justify the investment business case. The requirement for a new system to support deposit taking is clearer, for the regulatory and technical reasons explained above. Once again this conceals the external driver, although some inferences can be made. The wish to take deposits could be driven by a desire for a cheaper and more independent source of funds, either to replace existing liabilities or for new growth. It could be motivated by a lack of available wholesale funding in a post-Credit Crunch environment. Lastly, it could be part of a mission to offer a fuller set of financial services to clients, either as part of a financial inclusion mission, or to deepen the relationship with clients to protect against competition from banks. More research would be required on which were the key fundamental drivers. Kotler and Armstrong’s (2010) organizational factors, structure, policies, procedures and systems, seem only to be significantly influential in the driver of standardization. Not surprisingly this driver exists only in the four networks and one apex fund. However it may be more important that this represented 100% of the multi-entity cases studied. They all took action to purchase standardized MIS systems. Less clear is the fundamental motivation. In all cases the network (or fund) used its greater aggregate purchasing power to secure a better price for performance. However, not all of the partner or subsidiary MFIs were actively looking for new systems, and there was limited aggregation of requirements and priorities from the field in advance of the project. The drivers seems to have been genuinely internal and organizational, namely a desire for homogenized systems for better reporting to the center, enforcement of robust systems to prevent fraud and aid improvement credit risk

124

Chapter Six

management, and overall a general sense that the umbrella organization was simply more competent to select, implement and manage MIS systems than the entities in the field. There were many opinions expressed related to the interpersonal and individual factors, and how these affected the purchasing and implementation process, for example on the cultural differences between core banking systems vendors and employees of pro-poor MFIs. However, none of these amounted to significant drivers to the purchase event itself. However, one factor related to the individual within the decision making unit does stand out. In two cases, the MIS purchase followed the appointment of new management. If all other factors were equal (admittedly a big ‘if’ without closer analysis) it can be inferred that the new management took a latent business case that may have existed, and provided the leadership to drive the process forward to a purchase event. Finally, and not surprisingly, new money was a key driver for investment. In fact this was predicted by the EMN study, where 81.3% of respondents cited lack of funding as the main constraint to the investment in technology. This factor was far higher than next highest constraint (unclear requirements, 37%). The key insight here is that unlike most of the other drivers, new money tends to occur as a specific event. It also affects not just resources available, but also business plans (and thus requirements), and sometimes even causes changes in management personnel. Generally it also adds a new voice, that of the funder, to the decision making unit. While the role of funders in the decision making process was not highlighted in the responses, clearly their money has a very loud voice.

Potential future research There were many limitations of the study in its aims to assess the drivers to MIS purchases. These could be improved by further study. At 20 cases, the sample size was quite small. As a result it was not representative in composition either in terms of organizational type, scale or location. A larger sample could remedy this. Respondents were not objective. However, this is an inherent problem of such a study method. More importantly, they were subjective in different ways. Some cases were described by the MFI and some by their vendor, or an industry expert. It would be more accurate to use either just clients or vendors and try to control for a uniform bias, or to compare between the opinions of both clients and vendors in each case.

Organizational Buyer Behavior in the Microfinance MIS Market

125

While opinions were expressed on the influence of each driver, there was no objective evidence of their existence, let alone their influence. A further study could try to get impartial views of, say, levels of competition in Ghana, and then apply this to Ghanaian cases. Evidence for influence could be derived by correlating between objective factor and eventual action, controlling for the other factors. No correlation could be inferred between organizational structure and decisiveness. The nature of the study was of MFIs that purchased systems: 100% were decisive in the end. Further study would be required to study this factor. Analysis of the findings revealed that a sample where all cases resulted in a purchase asks many questions about what happens when MFIs don’t buy systems— for example, projects that are cancelled, or that go on for years without result, or where a need clearly exists but no action is taken. Studying such non-purchase cases would be very difficult, but would answer such questions as the correlation between structure or organization type and decisiveness. For example, are consensus-oriented NGOs less inclined to action than licensed banks (as one might suspect)? Finally, while the study assessed the drivers to action, it did not look at the degree of action, for example in terms of the resources allocated to the project. As the EMN survey states, “Organizations spend an average of 9% of their operational budget on ICT, with two organizations spending 15% and one 30%.” Further study could investigate what factors drive higher and lower proportionate levels of investment. Similarly, further study, with a wider sample group, could also examine the types of system purchased, and how these correlated with the internal and external factors.

Recommendations For vendors, the key recommendation is to focus on empirical evidence of the main drivers identified by the study. Ideally these should be volunteered by the MFI, rather than in response to questions, such as, “Are you planning a major expansion?” or “How are you being affected by competition?” There is a danger that the question will suggest a concrete driver where none exists. The findings suggest that a vendor should drill down below the positive statements of the MFI to identify the business case, and test it for evidence of the key drivers. Moreover, vendors should question whether any of the drivers are manifesting themselves as events. The client may truly need a new system and may have the funds, but maybe this was also true last year, and the year before. This year may bring the same lack of action.

126

Chapter Six

Both of these recommendations echo good sales practice, to identify a solid business case, and preferably a compelling event. The findings show evidence that this good practice brings results and avoids wasted effort. For MFIs and their funders, the key recommendation is to use these findings as a framework with which to assess the business case. The danger here is certainly one of reverse rationalization: “Are we planning a major expansion?” or “Are we being affected by strong competition?” The answer is likely to be ‘yes’. However it can help the MFI, or its funder, to test the business case against some key factors. Moreover, if a system selection process has dragged on for some time, it may be that a key driver, or a key event, is missing. For regulators, the findings may explain why MFIs are or are not upgrading systems. This may help regulators to motivate MFIs to improve their systems, or to regulate and control proliferation of new technology. For consultants, the key recommendations are similar to those for vendors. The drivers and events identified can be used as a framework to identify the true readiness of an MFI for new technology investment.

Conclusion The key external drivers to MIS purchases identified were competition, followed by the availability of new funding, plus the influence towards improved MIS from new funders themselves. Other main external drivers included improved economic stability in the local market, changes in regulation and the availability of new technology, either for electronic channels, or simply improved internet bandwidth for branches. The principal driver of competition was broken down into subcategories. Examples of such included banks offering improved services such as ATMs and bill payments, banks and Telcos providing mobile payments and mobile wallets, and banks and competitor MFIs offering loans at lower interest rates owing to lower cost ratios. The key internal drivers identified were a drive for growth, plus the need for new systems with the necessary functionality, robustness and real-time information needed for deposit-taking. Additional internal drivers included the need to invest in robust MIS as a foundation for electronic channels. A need to move to a centralized, real-time system was cited, but this correlated with a move to take deposits, which is the more fundamental driver of the two. A further internal driver was identified that was specific to multi-entity purchasers in the group, such as networks and an apex fund, namely the need expressed to standardize systems across the group of subsidiary or partner MFIs.

Organizational Buyer Behavior in the Microfinance MIS Market

127

The principal internal driver of growth seemed to be a derived need masking a number of more fundamental drivers. The most fundamental of these involved the mission of the MFI to increase its outreach. However, there was quite close correspondence (50%) between the drivers’ growth and new funding. This may suggest that both new funding and new MIS were sought as part of a drive for growth, or that growth (to disburse the new funds) and new MIS were resultant from the new funding event. Further analysis of correlation between the drivers plus specific interview responses suggests that growth was needed in order to justify a wider business case, for example to take a deposit license, invest in electronic channels, or in fact to justify the investment in new MIS itself. The key finding of the study was the very close correspondence (85%) between an MIS purchase and at least one driver that manifested itself as an event. This concurs with sales management theory of the need for a compelling event to drive the purchase process to a successful conclusion. The principal events cited were the arrival of significant new funding from a microfinance investor, lender or donor, either for lending expansion, or more specifically for investment in technology, and the request for an award of a deposit-taking license. Other driving events were the arrival of new management, the creation of a new entity, and in two cases the failure of the previous MIS project. While these events clearly were a necessary condition for MIS purchases, they equally clearly were not sufficient. Each event masks more fundamental drivers. What drove the previous (failed) MIS project? What drove the institution to obtain a deposit license? Why was new funding sought? Nevertheless, the survey and analysis shows that, in almost all of the MIS purchase cases studied, the external and internal drivers focused themselves into specific events prior to the MIS purchase commitment. A tentative inference for MFIs, MIS vendors, and other stakeholders such as regulators, investors and donors is that need is not sufficient for change. It must manifest itself as an event—a jolt even—to rouse the MFI into action. Without that event the MFI will endure its needs and find workarounds to its problems, and continue business as usual.

128

Chapter Six

Bibliography Accenture (2010), Automation of SACCOs, Assessment of Potential Solution. Accenture Development Partnerships. Ahmad, A. (2005). Management Information Systems (MIS) for Microfinance. Pakistan: First MicroFinanceBank Ltd. Ashta, A. (2011), Advanced Technologies for Microfinance: Solutions and Challenges." IGI Global. Braniff, L. & Faz, X.. (2012). Information Systems. A Practical Guide to Implementing Microfinance Information Systems. Consultative Groups to Assist the Poorest: Christen, R. P., & Drake, D. (2002). Commercialization: The New Reality of Microfinance. In D. &. Drake, The Commercialization of Microfinance; Balancing Business and Developmenet. Bloomfield CT: Kumarian Press. CNN Money. (2008). Nobel winner slams for-profit microfinance. CNN Davis, F. D. (1989), "Perceived usefulness, perceived ease of use, and user acceptance of information technology", MIS Quarterly 13(3): 319–340 Dimaggio, P.J. and Powell, W.W. (1983) The iron cage revisited institutional isomorphism and collective rationality in organizational fields, "American Sociological Review", Vol. 48, No. 2, pp 147-160. EMN (2011), The Use of Technology in Microfinance. European Microfinance Network, IT & Innovation Working Group. Helms, B. (2007). Access for All: Building Inclusive Financial Systems. Washington, DC: World Bank. Kotler, P. & Armstrong G. (2010) Principles of marketing. 13th edition. Pearson Prentice Hall, New Jersey Mainhart, A. (1999). Management Information Systems for Microfinance. Development Alternatives Inc. McKee, K. (2008, December 15). Behind the Headlines: The Credit Crunch and Microfinance – One Potential Scenario: An interview with CGAP expert Kate McKee. Retrieved November 12, 2008, from Consultative Group to Assist the Poorest: http://www.cgap.org/p/site/c/template.rc/1.26.4507 MicroCapital. (2007, October). Blackstone and Carlyle Considering Microfinance. MicroCapital Monitor , 2 (10). MixMarket (2012), Technology Service Providers, www.mixmarket.org/service-providers/Technology%20Provider /all/all/all/all Oracle (2009) Target Account Selling, Siebel Applications Administration Guide. Oracle Inc.

Organizational Buyer Behavior in the Microfinance MIS Market

129

Poyo, P., & Young, R. (1999). Commercialisation of Microfinance; A Framework for Latin America. Development Alternatives Inc. Rhyne, E., & Otero, M. (2006). Microfinance throught the Next Decade: Visioning the Who, What, Where and How. Accion International. Robinson, M. (2002). The Microfinance Revolution (Vol. 2: Lessons from Indonesia). Washington, DC: World Bank and Open Society Institute. Rogers, E.M. (1995) Diffusion of innovations, Fourth Edition ed., New York, Free Press. Scott, W.R. and Christensen, S. (1995) The institutional construction of organizations: International and longitudinal studies Thousand Oaks, CA, Sage Publications. Supka, G. & Ling, J. (2009), Technology for Microfinance. International Banking Systems Journal Supplement. IBS Publishing Tornatzky, L. and Fleischer, M. (1990) The process of technology innovation, Lexington, MA, Lexington Books. Webster, F., & Wind, Y. (1972). A General Model for Understanding Organizational Buying Behavior. The Journal of Marketing.

Additional Reading CGAP. (2005). Donor Brief No. 23:Funding Microfinance Technology. Consultative Group to Assist the Poor Technology Resource Center. Cracknell, D. (2004). Electronic Banking for the Poor - Panacea, Potential and Pitfalls. Small Enterprise Development , 15 (4), 8-24. Flynn, P. (2007). Microfinance Technology: Implications for Non-Profit Financial Institutions. Exempt Magazine (Spring). Goodman, P. (2004). Microfinance Investment Funds: Objectives, Players, Potential. 2004 KfW Financial Sector Development Symposium. Berlin: Appui au Développement Autonome. Helms, B. (2007). Access for All: Building Inclusive Financial Systems. Washington, DC: World Bank. Ivatury, G. (2004). Harnessing Technology to Transform Financial Services for the Poor. Small Enterprise Development , 15 (4), 25-30. KPMG. (2003). Banking Systems Survey. Luxembourg: KPMG . Mathison, S. (2005). Electronic Banking with the Poor; Increasing the Outreach and Sustainability of Microfinance through ICT Innovation. Brisbane, Australia: The Foundation for Development Cooperation. Sundaran, S. (2007). Microfinance: Emerging Trends and Challenges. Edward Elgar Publishing. USAID. (2008a). Microfinance Core MIS Systems - The Business Case for Outsourcing. United States Agency for International Development.

130

Chapter Six

USAID. (2008b). Outsourced Microfinance MIS Systems - a Decision Guide for Microfinance Institutions. United States Agency for International Development. World Bank. (2003). Standard Bidding Documents; Supply and Installation ofInformation Systems - 2-Stage Process.

CHAPTER SEVEN THE CHALLENGES OF BEING AN MIS SERVICE PROVIDER IN MICROFINANCE: CASES FROM INDIA GAURAV SINHA

Setting the context: Role of MIS in microfinance “Technologies cannot make good institutions, but they can make good institutions better.” —Hany Assad, International Finance Corporation1 Case 1: Equity Building Society (EBS) had 40,000 manually operated/managed accounts in the year 2000. The following year witnessed the installation of a new Information Technology (IT) Management Information System (MIS) at EBS. This resulted in rapid growth of two-and-a-half times in the number of accounts. Business also grew two to three times with existing employees, indicating enhanced productivity (Frankiewicz, 2003). Case 2: Swayam Krishi Sangamam (SKS) is the largest microfinance institution (MFI) in India and was the first to get listed on the country's stock exchange.2 SKS cites its information system, designed and deployed in the organization, as one of the key catalysts for the growth of its business, with nearly 5.3 million customers (Ashta, 2010). Case 3: Another large MFI3, which has been using an MIS since its inception, has grown from having 85 branches, 175,000 customers, and 850 employees to 4,000 employees serving 750,000 customers across a network of 310 branches (Kashyap, 2011).

Box 1: The Magic of MIS in Microfinance

132

Chapter Seven

Box 1 contains three cases, among many, that show the importance of IT-based MIS in enhancing the effectiveness and efficiency of any organization (Khan and Khan, 2011). Rangarajan (2011) underlines several benefits of using technology in the banking sector in India, such as the reduction in intermediation cost (defined as the ratio of operating expense to total assets) of the scheduled commercial banks from 2.59% in 1991-92 to 1.71% in 2010-11. He cites several other examples, such as declining cost-income ratio from 55.3% in 1999-2000 to 45.21% in 201011, plus the several-fold increase in business per employee and business per branch of Indian banks (Rangarajan, 2011). Microfinance, although a recent phenomenon compared to the banking sector, has not been left untouched by the benefits of IT-based MIS (Kashyap, 2011). Research studies have proved the benefits of MIS usage in MFIs (e.g. Barua, Kriebel & Mukhopadhyay, 1991; Ledgerwood, 1999; Ivatury & Pasricha, 2005; Intellecap, 2006; Ratan, 2007; Kashyap, 2011). Examples of the benefits of using MIS in MFIs highlighted by these studies are compiled in Box 2. x x x x x x x x x x x x

Enhanced operational efficiency Risk mitigation Effective budgeting Improved cash management Reduced turnaround time, which results in lower operating costs Savings in manpower usage resulting in higher productivity Improved outreach and geographical coverage Faster, more efficient, and more accurate reporting Improved service offerings Better informed decisions Greater transparency The introduction of features such as smart cards (carried by borrowers) and hand-held devices (carried by loan officers) that enable the recording of transaction data directly into the MIS without manual intervention

Box 2: Benefits of using MIS in MFIs

The Challenges of Being an MIS Service Provider in Microfinance

133

Initiatives by regulators in several countries have made MIS a prerequisite to fulfill the MFI regulation requirements. In India for example, MFIs can disburse loans only after due verification with the Credit Bureaus. The government has given licenses to four Credit Bureaus and two of these (High Mark and Equifax) are now operational (Ashta, 2012). Although these initiatives are at a nascent stage, they are laying foundations for more robust information systems in MFIs. Moreover, with the rapid growth in the microfinance sector and increasing demands for a robust MIS, the number of MIS vendors has increased many fold (Srinivasan, 2009). In India, the information technology service-providers (or vendors) have made significant efforts to streamline MIS in their client MFIs in order to help them better manage information relating to the clients and use the reports generated appropriately. Although there is no disagreement on the enabling roles played by these service providers in the microfinance sector, there are few, if any, studies researching the supply side issues and challenges of MIS service providers in India. Conversely, the rationale for using a particular kind of MIS or the significance of using MIS by MFIs has been studied by many researchers. With this background, this research study explores the challenges faced by MIS vendors in India. It further examines the steps undertaken by them to address these challenges. Overall, this study is an attempt to bring forth the supply side perspective of the MIS vendors to the MFIs in India. This paper details these issues according to the following structure. The next section presents the existing knowledge on this topic through a review of secondary sources. The subsequent section explains the research framework and methodology used for data collection. Then the findings based on the interactions with founders or chief executives of the MIS vending companies are discussed in detail. Finally, the recommendations and conclusion section summarizes the entire study and gives brief recommendations for action.

What does the literature say? There are various factors, internal and environmental, that affect technology service providers around the world. These factors are studied extensively by various researchers (Miller & Côté, 1987, Eisenhardt & Schoonhoven, 1990; Shan, 1990; Lee, Lee & Pennings, 2001; Nambisan, 2002). These factors include, but are not limited to, organizational resources—financial and human, top management, strategy, market and

134

Chapter Seven

competition, government regulation, infrastructure, risks, product portfolio and marketing/ distribution strategy. Several studies on Indian software firms complement these findings. For instance, Bajpai & Shastri (1998) have mentioned government’s role as a facilitator for the software firms. They have stressed the importance of human and financial resources, value chain, operational issues and the market for software firms in India. Sastry (2001) has focused on operational issues of software firms in India. He finds servicing at clients’ level as one of the important challenges that a software firm faces. In another study of the Indian software industry, Chakraborty & Dutta (2002) have identified several factors impeding the software companies, which include infrastructure, regulation, risks, limited funds, and marketing and distribution. Another study on the Indian software industry endorses similar views and indicates a substantial role of government policies, infrastructure, competitive niche and resources including manpower in the growth and development of software industries (Bhatnagar, 2006). Jhamb (2011) has taken into account the importance of infrastructure, government policies, processes and usage of resources in software industry in India. A broader classification of the factors, discussed above, can further help in understanding the challenges facing a software firm in more specific terms. For instance, Nambisan (2002) has specifically classified internal and external factors affecting software firms in his extensive literature survey on the important determinants impacting software firms. On the one hand, he mentions industry characteristics, technology characteristics, economic and technological infrastructure, regulatory infrastructure, regional culture and external characteristics as external factors. On the other hand, his list of internal factors includes founding conditions of the firm, strategic factors, the firm’s resources and competencies, and internal stakeholders’ characteristics. An in-depth analysis of these factors gives an idea for further sub-grouping. These factors can be grouped as external factors—government, market and clients—and internal factors—operational issues and resources availability. Both of these factors play decisive roles in the growth of the software firms. This classification is akin to what Moore (1993) has proposed for understanding the actors of a business. He has proposed an ecosystem approach to understand the companies in the context of entire business environment. There are various aspects of a business ecosystem.4 These include actors, relationship between actors, performance, dynamics and strategies, and behaviors of actors. A business ecosystem typically includes clients, markets, products, processes, organizations, stakeholders,

The C Challenges of Being B an MIS Service Providerr in Microfinance

135

and governm ment / societyy as actors in the t ecosystem m. These actors have an impact on the growth of o companiess. Resources including caapital are analogous too energy in any a ecosystem m, and these hhelp the com mpanies to prosper (Pow wer and Jerjiaan, 2001). If we look at the miccrofinance secctor using thiss ecosystem approach, a MIS vendorrs can be vieewed as one of the actorss. As discusseed in the previous secction, the rolee of an MIS vendor v is vitall for sustainin ng growth of MFIs. Hoowever, extannt studies eitheer emphasize the role of MIS M in the growth of M MFIs or pressent the facto ors affecting tthe software industry. There is harrdly any specific research av vailable on unnderstanding the t issues and challengges associatedd with the MIS S vendors in th the microfinan nce sector in India. Thiis study is a first fi step to exp plore, in-depthh, the challenges faced by the suppply side, the MIS vendorss. For the puurpose of thiss study a framework ((Figure 111) was developeed based on thhe preceding literature review on bbusiness ecosyystem of MIS S vendors draawing knowledge from Moore’s moodel.

Governmeent policies an nd norms

Market challengees

Figure 1: Ressearch Framewoork.

Challen nges of MIS S vendo ors

A INTERNAL

EXTERNAL

Client leveel issues

Resou urce challeenges

Operational challeenges

136

Chapter Seven

Using this model, we can study the role of key actors on the MIS vendors in India. In this study, the key actors, i.e. the government, clients and market, act as external resources while organizational resources and operations as internal actors have been studied in detail according to the proposed framework. But before discussing the research framework in detail, the following section gives an overview of the methodology adopted for data collection and analysis.

How it was done: Research methodology The study was qualitative in nature and used a case study method to collect data. Both primary and secondary tools were used for data collection. First, interviews, based on open-ended guided questions, were conducted with the founders and/or the chief functionaries of the organizations studied. A guiding note as a precursor to the interviews was sent to the respondents to make the discussion more focused. The case study method helped in detailed examination of the significant issues, which the vendors faced. This method also helped to document the learning of vendors in tackling these challenges. Subsequently, secondary data available through annual reports of various years and other information available on their websites were studied extensively as a medium to supplement the findings obtained through primary data. Grounded analysis was done in order to validate the proposed framework developed for the study. For the purpose of analysis, responses of the vendors were compared and substantiated with literature review. The data collected was tabulated and then classified according to the research framework (Fig. 1). For the purpose of this study, three vendors were selected based on the length of their experience in the microfinance market and keeping in view the convenience of the researchers. A brief description of the institutions studied has been presented in the end (Appendix A).

Research Findings The findings of this research study are categorized into two sections. The first section discusses the challenges faced by the MIS vendors in detail, and the second examines the responses of these vendors as to how these challenges were dealt with.

The Challenges of Being an MIS Service Provider in Microfinance

137

1. Challenges for MIS vendors in microfinance MIS plays an important role in the growth of an MFI and so do the MIS vendors to these MFIs. There are several challenges faced by MIS vendors that affect their businesses. Since vendors play an important role in developing an effective MIS for the MFIs, it is important to understand their challenges and issues in providing these essential services. The responses, gathered during the data collection process, have been classified according to the proposed research framework (refer Figure 111) and are discussed in detail hereunder: 1.1 The pros and cons of Government interventions Government interventions, through the issuance of policies, rules, guidelines, notifications, circulars, and orders, affect all industries. Studies have indicated that government policies play a major role in technologybased organizations (Miller & Côté, 1987; Preston, 2001). Bhatnagar (2006) too has mentioned government’s role in developing and supporting IT firms in India. The available data on business growth in India pictures a dismal performance in providing a conducive environment for doing business. Globally, India stands at 132nd in the ranking of 185 economies on the ease of doing business (World Bank, 2012). There are numerous challenges facing business ventures. For example, bureaucratic procedures to get permits and legal steps necessary to enforce contracts obstruct the smooth functioning of business ventures in India. India stands at 182nd and 184th position respectively in the list of 185 countries on these indicators (World Bank, 2012). MIS vendors or technology service providers are no exception in this regard. The study finds both enabling and disabling aspects to any government intervention. For example, findings of this study indicate that policies related to standardization, improved Internet availability, reduced cost of bandwidth and promotional subsidies act as enablers. However, legislation which obstructs the growth of the microfinance sector and barriers to entry in government procurement systems for small vendors act as disablers for the MIS vendors. Policies or regulations, which drive standardization in the microfinance sector, can help an MIS vendor. For example, the Reserve Bank of India (RBI, India’s Central Bank) has issued a regulation for the microfinance sector to charge interest on a reducing balance basis. Safal views such standardizations as opportunities for developing applications that are more generic and thereby reducing cost of operations. Such policy level changes result in mounting pressures on the MFIs to go to MIS vendors in order to

138

Chapter Seven

update their software applications. The cases of Force Ten and Jayam substantiate this point, whereby clients have started querying on changes in the software as soon as the regulation was made. Another policy change related to generating clients’ history before lending and sharing data related to clients’ acquisition with the Credit Bureaus has had an impact on the business of MIS vendors and started generating work for the MIS vendors. MFIs have already started enquiring about integrating these changes in their applications. For Force Ten and Jayam, clients are coming to them to update their applications as soon as possible. Another factor, i.e., low cost of bandwidth and Internet availability across the country (a policy announcement by the government of India) has been perceived as an enabler by the MIS vendors. Safal perceives availability of Internet in remote rural locations and bringing down the cost of bandwidth as an opportunity for the MIS vendors. This acts as an incentive for the MIS vendors to develop web-based applications. All three respondents have mentioned the benefits of having web-based applications installed at many locations. For example, all of them have shared that they have been able to bring down their operations costs, because issues in the applications at the client level can be addressed and serviced online. Support in terms of cash, in-kind or other forms of subsidies from government or other donors, for establishing and running a business, helps the growth of the industry. As an example of direct support from the government, Force Ten obtained rented office space at a subsidized rate from the State government, though they had to struggle initially to get this space. Although the other two respondents were not so lucky and did not receive any direct support for their businesses, they consider such support indispensable in order to overcome resource challenges, especially for small sized MIS vendors. In addition to the enabling side of government interventions, there are disadvantages. The study identifies two major disabling factors: the first is regulation that adversely affects the microfinance sector, and the second, entry-level barriers to government procurement systems for small MIS vendors. For instance, the Government of Andhra Pradesh passed an ordinance for regulating the MFIs in the State in 2010-11. The law left MFIs with very little scope to do business. The repayment rates fell to less than 10%. Some of the MFIs closed branches and laid off staff. There were very few fresh disbursements from the banks. The loan disbursement by banks to MFIs was reduced sharply by approximately 30% during this period. The impact was felt by the MFIs across the country. Subsequently, the RBI

The Challenges of Being an MIS Service Provider in Microfinance

139

came up with a regulation capping interest rates and margins. Besides, banks were advised to consider lending to MFIs as priority sector lending, provided the MFIs follow the norms and rules provided in the RBI’s regulation (Srinivasan, 2011). The MIS vendors also felt the heat of this policy initiative. The respondents indicated that MFIs did not want to invest in MIS. Their focus shifted to managing funds and operations, and therefore, they were not focusing on their MIS. Moreover, vendors lost many of their clients. Jayam shared one instance of losing many good clients due to this legislative intervention in Andhra Pradesh. This, in turn, resulted in reductions in their annual revenue, delayed payments and, consequently, hampered the growth of the company during that period. On top of this, there was hardly any fresh demand for MIS or related IT services from the MFIs. Moreover, entry-level barriers to Government tenders, such as minimal capital requirements to bid for tenders, and requirements for small firms to form consortia with larger firms for project bids, play a major role in obstructing the growth of MIS service providers. Safal mentioned such a case, and this was further validated by studying documents from several banks. For example, for the Financial Inclusion Technology Fund (FITF), there was a minimum earnest money deposit5 requirement of more than INR two million, which made it impossible for small MIS vendors like Safal to bid. 1.2 Client level issues The second external key actor, i.e. the client, has an important place in the business ecosystem of an MIS vendor. However, clients often pose several challenges to the vendors, such as the difficulties of providing onsite support, unavailability of trained staff, and an unwillingness to adopt new technology (Sastry, 2001). Interactions with the MIS vendors during data collection have yielded four such factors at the clients’ level, namely keeping pace with the growth of MFI clients, educating clients on the usage of applications, human resources (HR) challenges in MFIs, and ethical practices followed by MFIs in doing business. Past trends show that MFIs grow at a very high rate in terms of volumes (Srinivasan, 2011). With increasing portfolios, MFIs are required to set up new systems or modify old systems, including information systems, to keep pace with the changing scenarios. It is probably for this reason that MFIs in Bangladesh have adopted MIS for operational efficiency and to manage the information related to a large number of

140

Chapter Seven

existing clients (Mia, 2005). These shifts require changes in the MIS to be made by the vendors. At the vendors’ level, it is difficult to come up with new systems very rapidly. One such case was shared by Safal, where both Safal and an existing MFI client in West Bengal opted for disassociation because the growth requirements at MFI could not be satisfied by the vendor. One major challenge identified by all the respondents related to clients’ capacity to use the application efficiently. The respondents stated that there were only a few clients who knew how to use analytical reports generated by the software to support their operations. They specifically mentioned that many clients were not clear about what they wanted and therefore the vendors developed applications based on their own understanding of the clients’ requirements. Many times vendors were held responsible for the failure of an application, despite the fact that the problem lay with the MFI. Jayam encountered two such cases, when clients called them in to fix a software problem. They claimed that the system had not been working since it had been implemented. Jayam staff travelled for more than 24 hours only to find that a power switch had not been turned on. MFIs face difficulties in getting trained human resources to develop or manage an appropriate MIS (Ahmad, 2003). During the data collection process, all the respondents mentioned lack of availability of key resource staff at the client level for managing MIS, especially in the small and medium-sized MFIs, as a major challenge. They shared a view of the casual approach adopted by many MFIs in assigning a dedicated person to take care of IT requirements and MIS applications. The respondents indicated that MFIs did not give much emphasis to the maintenance of MIS, and therefore they were reluctant to spend money on training their staff on how to manage an MIS. Sometimes, this type of casual approach affected efficient usage of applications. Another challenge, at the client level, is associated with staff turnover in MFIs. Once a trained person leaves an MFI, there is a need to train other employees, but again, MFIs sometimes do not want to pay the vendors as it increases their operational costs. Safal mentioned a case which damaged the business relationship with one of its clients. At one point of time, the process of using MIS applications suffered which the client termed as an application failure. However, this failure was a result of a trained employee leaving the MFI, which hindered the entire process of MIS usage (including regular data entry) in the organization. Sometimes MFIs tweak or tamper with the software according to their requirements. The author, while doing an institutional analysis (another

The Challenges of Being an MIS Service Provider in Microfinance

141

research project), came across one such incident. He found an MFI in Uttar Pradesh tweaking the software to calculate the portfolio at risk (PAR) in order to show a low PAR. The organization was dividing the amount received from its clients equally into interest and principal rather than first taking into account the payment of interest. Similarly, one of the vendor respondents in this study raised this issue of tweaking of applications by citing an example of calculation of interest rates being changed by one of its clients. According to him, this type of tweaking not only affects applications’ fundamentals but also raises doubts about ethics of MFIs in working with the poor.6 1.3 Market challenges Competition, as another key actor in the business ecosystem of an MIS vendor, plays a major role and affects the growth of an organization. Jayam cited one such incident of competitive pressure. In 2005-06, seeing the growth and potential in the microfinance sector, many software providers entered the market. Competition intensified in the sector because many vendors started offering products at very low prices. This affected both the MIS vendors and MFIs. On one hand, many MIS vendors lost some clients, while on the other the belief of MFIs in the small software firms was also shaken because of low quality software or services. Development in other related sectors can bring a mix of positive and negative factors to MIS vendors. On one side, there is competition within the sector, while on the other there is complementarity between related sectors. On the negative side, Safal quoted an example of bio-matrix cards, which came in vogue recently in India because of the perception that villagers could not remember their codes. Due to these initiatives, Safal’s business was adversely affected, as these initiatives took up the major share of the MIS requirements at its MFI clients. In addition, Safal mentioned that integration of the MIS with such technologies added to the operations costs of the MFIs, for which sometimes the MFIs did not want to pay. On the positive side, Force Ten saw this as an opportunity provided the organization is willing to pay for such changes to the MIS. 1.4 Operational challenges There are internal challenges as well that affect the growth of an organization. In a study of the Indian software industry, Sastry (2001) highlighted the role of operational challenges on a software firm. This study supports his findings and affirms the impact of operational

142

Chapter Seven

challenges on the MIS vendors in the microfinance sector, including servicing and customization at the client level. Many MFIs operate in remote areas, where the required infrastructure for MIS is not available. If the applications are installed in such places, attending to problems can take time, and sometimes, there can be issues in the tracking of data (Ahmad, 2003). Safal indicated that its cost of operations rose because of attending to the complaints received by clients in remote locations. It added that responding to such complaints was not only difficult for Safal, but also added to the cost of servicing, for which the clients did not want to pay. Force Ten cited a case of irregular data entry in the MIS in a remotely located branch of an MFI, which subsequently resulted in application failure. However, the issue was related to weak infrastructure at the location, for example irregular power supply, which led to irregular data entry and poor Internet connectivity for sending the data backup to Force Ten for analysis. This was an issue with the MFIs using a desktop MIS application rather than those using a web-based MIS application. Jayam’s case proves this point. Before Jayam developed web-based applications, there were issues pertaining to data backup, added costs and staff availability to address the complaints of the MFIs, especially at remote branches. Jayam had one more level of challenge: there were cases where MFIs tried to resolve the problems in the applications on their own in order to save the servicing cost. This later resulted into failure of applications, the blame of which came on the vendor. Another operational challenge exists meeting the diverse requirements of the MFIs. There are certain specific rules and policies laid down by each client, which need integration with their MIS operations. For example, some MFIs require accrual basis transactions, while others require cash basis transactions. Non-standardization of operations between MFIs poses difficulties in developing applications, and adds to the cost incurred by the vendors. For instance, one of Safal’s clients required merging the unpaid portion of the first loan cycle with the second cycle loan disbursement, which was a very specific requirement according to the policies laid down by that MFI. Force Ten considered that the inability of many MFIs to put forth their requirements in a clear and precise way led to delay in the entire customization process. Jayam endorsed this view on the lack of clarity by the MFIs on their requirements. Jayam mentioned irrational expectations as a major challenge at the client level once the process of developing applications starts. Jayam gave an example of many of his business clients to substantiate this point. The MFIs wanted the application soon after

The Challenges of Being an MIS Service Provider in Microfinance

143

submitting their operations manuals, and in some cases proper testing procedures were skipped by the MFIs. In other cases, there was a lack of supply of process illustrations and change cases by the clients, which hampered the process of application development. Jayam also mentioned that there was better relationship between the client and the vendor where vendors were strong and clear in their vision and mission, and appointed software professionals to take care of the MIS. This strong support and good relationship helped them to resolve the issues at the clients’ level easily compared to other companies. 1.5 Resource constraints at vendors’ level One major internal challenge facing any organization is that of mobilizing adequate resources for smooth functioning. Studies have highlighted that resource constraints also limit the growth of technologybased organizations. Such constraints include the difficulty of hiring the ‘right’ human resources, high attrition rate and poaching by big companies, plus infrastructure and working capital or funding limitation (Ilavarasan, 2007; Bruneel, Clarysse & Wright, 2009). This study has found evidence of these issues with the MIS vendors in the microfinance sector including talent retention, late payment by the clients and financing issues. Talent retention is a major issue in most of the industries. Microfinance and associated sectors are no exception. Safal cited talent retention as a major challenge for small and mid-sized MIS vendors because of limited working capital and the small size of the organizations. On the other hand, Jayam had been able to retain its senior management team despite stiff competition. However, it raised concerns on getting experienced technical staff with sector and domain experience to strengthen its senior management team. This needed a company with deep pockets. Adding to this, the founder of Force Ten indicated poaching of trained staff by bigger companies as a challenge for MIS vendors. He shared a case of poaching by an MFI, which grew along with Force Ten. Later on, a trained employee, who was instrumental in developing the application, joined that MFI after being offered a higher salary. This poaching not only caused talent loss to Force Ten, but also damaged the business relations between Force Ten and the MFI. Since the employee knew the MIS application well, poaching also led to loss of annual maintenance revenue for Force Ten.

Chapter Seven S

Internal Figure 2: Chaallenges of MIS S Vendors.

Further, there is a wiidespread nottion that instaalling applicattions is a one-time coost. Accordingg to Jayam, many m of its cllients, especiaally those small and m medium-sizedd, considered installing an application as a a onetime cost. They did noot consider recurring r invvestments in the MIS developmennt process. Jayyam was exp pected to makke changes an nd update the applicatiion without chharging any ad dditional fee ffor doing so.

Challenges of MIS vendors

External

144

The Challenges of Being an MIS Service Provider in Microfinance

145

In addition, late payments by clients were quoted as an important source of cash flow problems for MIS vendors. While Safal was confronted by many clients availing their services but delaying payments, Force Ten hardly faced any such problems in receiving payments from the clients to date. Safal also stated that such payment delays ultimately affected them and created problems in managing the cash flows. Jayam added that many clients were not willing to pay an annual maintenance contract for their MIS application. Lending by banks to clients with deep pockets is a known phenomenon. The converse of this is also true. Banks and other lending institutions do not trust small vendors, and therefore do not want to lend them. Safal and Jayam identified this as an important challenge, which affected the business of small or mid-sized MIS vendors in India. However, while Safal and Jayam faced difficulties in receiving support from banks, Force Ten categorically mentioned that they did not seek support from external agencies. Moreover, there was no direct support from the donors or funders in the microfinance sector to develop applications for the small MFIs. However, the respondents mentioned that sometimes the MFIs received operational and capacity-building support from lending institutions for development of the MIS. According to them, such a kind of indirect support from donors or funders played a crucial role in influencing their businesses. However, despite the challenges (summarized in Fig. 2) discussed in this section, the MIS vendors have opted to stay in the market and devised several ways to meet these challenges. The next section captures in detail the ways and means adopted by the MIS vendors in countering these challenges.

2. Addressing the challenges: Vendors’ response In order to survive in the competitive market, organizations adopt various strategies. The MIS vendors, interviewed during this study, adopted different strategies to cope with the challenges and uncertainties. Some of the strategies they adopted could be considered as counterproductive in the long run, but were necessary for survival. Other strategies can provide insights for various players in the market cutting across geographies.

146

Chapter Seven

2.1 Meeting financial requirements Meeting financial requirements is vital for the survival and smooth functioning of a company. Interestingly, Safal addresses its cash flowrelated issues, resulting from delayed payments by the clients, in a unique manner. It manages its cash flow either by taking personal loans from the directors or in the extreme cases through delayed payments of salaries to the employees. This is the last resort for Safal and its respondent did not find it to be a good idea to delay the salary payments. Two other vendors ensured regular flow of funds by retaining their profits or lowering down the cost of operations through adopting new technologies. 2.2 Managing human resources All the respondents view retention of human resources as an important strategy. Safal’s experience suggests that training non-technical people inhouse and bringing them into this sector works well in the long-term. According to Safal’s founder, these employees are more loyal to the organization because they do not have formal degrees or diplomas that can allow them to obtain a job elsewhere and therefore they stay for a longer duration. Frequently they also act as back-up employees, when required. However, Jayam has adopted a different strategy. According to Jayam, the senior management and founder members help in scaling up the business. Jayam has been able to retain its senior staff (and founder members) for 10-12 years. It has been using a three-tier approach to manage its human resources. First, the major focus of Jayam is on retaining the senior staff, defined as those with at least eight years of experience. They are given freedom to create and innovate as per the clients’ requirements, with responsibilities for managing teams in the organization. Next, as Jayam emphasizes staff retention, it considers those with three to seven years of experience to be the next line for higher order projects. As a part of this strategy, the employees in this stratum are given exposure to higher value projects and are also given additional responsibilities. Last in this series are the employees with less than three years of experience. They are taken through a developmental track, which leads them to develop their capacities. They are given exposure to MFIs and two months training on products and client handling. Jayam focuses less on this stratum owing to its experience of high attrition rate in this bracket.

The Challenges of Being an MIS Service Provider in Microfinance

147

2.3 Domain or sector diversification The respondents have addressed the issues, such as that created by government intervention in Andhra Pradesh (loss of business) or competitive environment, through domain or sector diversification. For instance, Safal has ventured into developing applications for other social sectors such as livelihoods, education, health (and other social) sectors. It has entered into commercial sectors as well and developing applications such as billing applications for cable operators. Jayam too has adopted the same strategy and ventured into providing applications for businesses in the real estate, housing valuation, vehicle finance, gold loans, and housing finance sectors plus chartered accountants. In addition, Jayam has ventured into new technologies and platforms, for example mobile and handheld devices and bio-matrix device integration, which they believe will help the MIS vendors to sustain if they experience certain external challenges. Force Ten is cautious and since its inception has been developing Enterprise Resource Planning (ERP) applications for manufacturing units, tea gardens or specific projects with multilateral donors. According to its respondent, such diversification has helped the company during difficult times. 2.4 Geographical diversification Geographical diversification, especially in other countries, gives an edge over the competitors and helps in absorbing the losses incurred in one country. The case of Force Ten validates this. Force Ten has diversified its business geographically to the Philippines and Sri Lanka. This has helped the organization to absorb the losses incurred in India, especially during the time of recession in the microfinance sector. Based on its experience, Jayam is planning to enter into other countries and is in negotiations with some companies outside India.

2.5 Relationship with the clients Continuous interactions with the clients play an important role in the success of an application and help in maintaining a long-term relationship with the clients. Safal and Jayam both have experience of mutual growth with the clients as they have shared good relationships with them. Jayam cites a case in point. The company has realized that thriving with just products (applications) in the market is difficult. After-sales service, a

148

Chapter Seven

‘must have’ in client relationships, is an integral part of business growth. According to Jayam, tough times in the microfinance sector in India have given an opportunity to strengthen relationships with the clients. Jayam did take one such step: it extended support to its clients by giving them an option to delay the payments due from them.

Recommendations and conclusion Clearly, IT-based MIS plays an important role in the life cycle of an organization. For the growth of MFIs, a good MIS is crucial. However, as we have seen that it is not easy for an MIS service provider to operate in the microfinance sector. It has to face numerous challenges. MIS vendors are affected by the environment prevailing in the microfinance sector. The challenges not only come from the external environment, but also emerge from within the business. While the external challenges come from government policies and norms, issues at the client level and market forces, the internal challenges emerge due to a lack of availability of resources and operational requirements. The ecosystem approach provides a framework for classifying these challenges in order to develop comprehensive understanding of the key actors affecting the business of the MIS vendors. The framework presented in this paper can help in detailed analysis of the impact of these key actors on the businesses of MIS vendors. Figure 2 presents a detailed classification of the challenges associated with the MIS service providers in the microfinance sector. This study has identified several challenges, which have been classified into five categories, according to the research framework mentioned above. The first category analyzes the challenges associated with government interventions. These challenges affect the MIS vendors both positively and negatively. The positive interventions in the microfinance or technology sector help in creating an enabling environment for the MIS vendors to flourish. On the other hand, however, the disabling interventions affect the businesses of these vendors adversely. Next, the study has identified challenges at the level of the clients. Lack of trained human resources at the clients’ level and their business practices increase the workload of MIS vendors. At the same time, educating the clients on the usage of applications is a huge challenge. Moreover, in the past some MFIs have grown astronomically. The major challenge facing the MIS vendors is to keep up with the pace of the growth of their clients. This sometimes results in disassociation from good clients. The third category is of challenges emerging from the market. The study indicates that the market plays a major role in the life of

The Challenges of Being an MIS Service Provider in Microfinance

149

an MIS vendor. Competition within the sector and competition from other sectors affect the business of the MIS vendors. In addition to the external challenges discussed above, there are internal challenges facing the MIS vendors. The study highlights that there are operational challenges for the vendors, which hinder their functioning. The operational challenges include servicing required at branches remotely located, popular perception of paying for software as a one-time cost and customizations required for each client. Moreover, the vendors have to struggle for resources, be they financial or human. We have also found that these challenges can be tackled or converted into opportunities by adopting various strategies. Diversifying into different domains or sectors can be an option. The vendors studied adopted this as a strategy to reduce the sector specific risks. Associated with this challenge is geographical diversification or entering new geographies. This can help in absorbing losses incurred in one region or location. Another option could be managing human resources efficiently, which ultimately leads to business growth and sustainability during critical times. MIS vendors should strengthen their relationship with their clients, especially through timely after-sales services. As the study indicates, sustained relationships with the clients act as a cushion during difficult times and help in growth and sustenance of business in the long run. There can be several other strategies to tackle the challenges facing the MIS vendors in the microfinance sector. The strategies adopted by the vendors studied here for tackling these challenges may not be the best ones, but they can provide a foundation stone for the existing and new players in the sector.

Suggestions for future research For the first time, the issues of the MIS vendors in the microfinance sector in India have been studied. There is ample scope of enhancing knowledge in this area through further research. This study can be a precursor to future research studies on similar issues, which might include continuing this research on a larger scale, ranking/rating of these factors, modifications in the research framework, or further classification of factors. Furthermore, the small sample size of three organizations, because of the qualitative nature of research, may influence the generalization of the findings. However, at the same time, it would be interesting to take forward this research to a larger sample population of vendors from different countries. Possible future research could measure quantitatively and qualitatively how constraints and opportunities faced by MIS vendors

150

Chapter Seven

vary geographically between different countries. Further research in the area could provide cues for in-depth analysis of the impact of these factors on the functioning of MIS vendors. There will be several new challenges in the future and it is hoped that vendors will come up with innovative methods to address these challenges.

Appendix A: About the organizations (Brief profile—alphabetical) Force Ten Tech Force Ten Tech (Force Ten) is an IT service-providing company that designs multiple software products for the microfinance industry. It was formally launched in early 2004 to cater to emerging IT needs of the other companies. The company started with digitization of engineering drawings and ERPs for manufacturing units. Seeing the opportunity in the microfinance sector, Force Ten ventured into the sector with its primary product BIJLI—Business Information Justified & Logically Integrated— microfinance module with unique features. Mr. Sabyasachi Chanda is the Founder and Managing Director of Force Ten Technologies. (Visit http://www.forcetentech.com/ for more information on the company) Jayam Solutions Private Limited Jayam Solutions (or Jayam) is a leading software solutions and service provider in the global market providing business solutions and high-end technology-based services to its customers with on-site, off-site and offshore development models. With a company history of more than 10 years, Jayam has designed, developed and deployed many enterprise class solutions in the areas of banking and microfinance using cutting edge technologies and re-usable frameworks. Initially, it was a partnership firm in the name of ‘Step in Soft’ founded on 10 December 1999. The ‘Step in Soft’ was later converted into private limited company in the name of M/s Jayam Solutions Private Limited on 24th April 2002. Mr. P.V.N. Pratap is the Founder and Managing Director of Jayam Solution Private Limited. (Visit http://jayamsolutions.com/ for more information on the company) Safal Solutions Private Limited Safal is an acronym for System Automation in Finance and Livelihood. Safal Solutions Private Limited (or Safal) is an IT solution company based in Secunderabad with an aim to develop customized IT solutions for nongovernmental organizations (NGOs) promoting microfinance, health, education and livelihood. It was founded in 2003 by Mr. Subodh Kumar

The Challenges of Being an MIS Service Provider in Microfinance

151

Gupta, a seasoned practitioner in the field of livelihoods and development, keeping in view the IT needs of small organizations in the non-profit or development sector. Mr. Gupta is currently heading Safal. (Visit http:// www.safalsolutions.com/ for more information on the company)

Bibliography Ahmad, A. 2003. Management Information Systems (MIS) for Microfinance. BWTP-Banking with the Poor Network. Retrieved Nov 26, 2012, from http://www.bwtp.org/pdfs/arcm/5Ahmad.pdf Ashta, A. 2010. MIS software for the microfinance market: An analysis. Social Science Research Network. Retrieved Nov 10, 2012, from http://ssrn.com/abstract=1583131 —. 2012. How MIS raises sustainable growth: The case of SV Creditline. Kuala-Lumpur. Retrieved Oct 29, 2012, from www.microfinancefocus.com/mfinews/how-mis-raises-sustainablegrowth-case-sv-creditline Bajpai, N., and Shastri, V. 1998. Software industry in India: A case study. Development Discussion Paper, 1. Harvard Institute of International Development. Barua, A., Kriebel, C. H., and Mukhopadhyay, T. 1991. An Economic Analysis of Strategic Information Technology Investments. MIS Quarterly 15(3), 313-331. Bhatnagar, S. 2006. India's software industry. In Technology, adaptation and exports: How some developing countries got it right, edited by V. Chandra. World Bank. Bruneel, J., Clarysse, B., and Wright, M. 2009. Linking Entrepreneurial Strategy and Firm Growth. Frontiers of Entrepreneurship Research, 29(13). Chakraborty, C., and Dutta, D. 2002. Indian software industry: Growth patterns, constraints and government initiatives. Working Papers, University of Sydney. Retrieved Nov 21, 2012, from http://econpapers.repec.org/paper/sydwpaper/2123_2f7655.htm Eisenhardt, K. M., and Schoonhoven, C. B. 1990. Organizational Growth: Linking Founding Team, Strategy, Environment, and Growth Among U.S. Semiconductor Ventures, 1978-1988. Administrative Science Quarterly, 35(3), 504-529. Retrieved Nov 23, 2012, from http://www.jstor.org/stable/2393315 . Frankiewicz, C. 2003. Information technology as a strategic tool for microfinance in Africa: A seminar report. AfriCap Seminar. Calmeadow.

152

Chapter Seven

Ilavarasan, P. V. 2007. Small IT firms face big growth challenges. The Economic Times. Retrieved Nov 23, 2012, from http://articles.economictimes.indiatimes.com/2007-0816/news/29486364_1_small-firms-software-industry-trade-associations Intellecap. 2006. Evolution of technology: Application in microfinance. Care-Cashe Project. Ivatury, G., and Pasricha, N. 2005. Funding microfinance technology. Helping to improve donor effectiveness in microfinance: Donor Brief No. 23. Washington, DC: CGAP. Jhamb, R. K. 2011. Contribution of software industry in the growth of Indian economy in the last decade. ZENITH International Journal of Business Economics & Management Research, 1(3), 97-11. Kashyap, A. 2011. MIS Usage at MFIs: A Pan-India Study. Banking With The Poor Network. Khan, M. E., and Khan, F. 2011. Conceptual overview of MIS and its importance in an organisation. Journal of Information and Knowledge Management, 1(2), 15-22. Ledgerwood, J. 1999. Microfinance Handbook: An institutional and financial perspective. Washington, DC: The World Bank. Lee, C., Lee, K., and Pennings, J. M. 2001. Internal capabilities, external networks, and performance: a study on technology-based ventures. Special Issue: Strategic Entrepreneurship: Entrepreneurial Strategies for Wealth Creation, Strategic Management Journal, 22(6-7), 615-640. doi:10.1002/smj.181 Mia, B. 2005. IT in microfinance: A Bangaldesh perspective. (S. Mathison, Ed.) Brisbane, Australia: Foundation for Development Cooperation. Miller, R.-E., and Côté, M. 1987. Growing the next Silicon Valley: A Guide for Successful Regional Planning. Lexington Books. Moore, J. F. 1993. Predators and Prey: A new ecology of competition. Harvard Business Review. Retrieved Nov 22, 2012, from http://blogs.law.harvard.edu/jim/files/2010/04/Predators-and-Prey.pdf Nambisan, S. 2002. Software firm evolution and innovation–orientation. Journal of Engineering and Technology Management (JET-M), 19, 141-165. Power, T., and Jerjian, G. 2001. Ecosystem: Living the 12 principles of networked business. Prentice Hall. Preston, J. T. 2001. Success factors in technology-based entrepreneurship. Transcript of a Lecture Delivered in Tokyo in 1997. Massachusetts Institute of Technology, MIT Entrepreneurship Center.

The Challenges of Being an MIS Service Provider in Microfinance

153

Rangarajan, C. 2011. Role of Technology in Development of Banking. The IDRBT Foundation Day Lecture. Institute for Development and Research in Banking Technology, Hyderabad. Retrieved Nov 19, 2012, from http://eac.gov.in/aboutus/chspe/RolTechdev_Bankg.pdf Ratan, A. L. 2007. An assessment of Pradan’s ‘Computer Munshi’ intervention to improve microfinance accounting operations. Microsoft Research India. Bangalore, India. Sastry, S. 2001. Operational and Tactical Challenges faced by Indian Software Firms. (163). IIM Bangalore Research Paper. doi:10.2139/ssrn.2161711 Shan, W. 1990. An empirical analysis of organizational strategies by entrepreneurial high-technology firms. Strategic Management Journal, 11(2), 129-131. Srinivasan, N. 2009. Microfinance India: State of the Sector Report 2009. New Delhi, India: Sage Publications Pvt Ltd. —. 2011. Microfinance India: State of the Sector Report 2011. New Delhi, India: Sage Publications India Pvt Ltd. World Bank. 2012. Doing Business 2013: Smarter regulation for small and medium sized enterprises. Washington, DC: World Bank Group.

Notes 1

AfriCap Seminar: Information Technology as a Strategic Tool for Microfinance in Africa, Nairobi, Kenya, April 26-27, 2003. 2 SKS was the first Indian MFI to go public and made US $350 million from its initial public offering in August 2010. 3 Cited reference does not indicate the name of the MFI. 4 See http://www.provenmodels.com/574 for more details. 5 A request for proposal (RFP) by a Regional Rural Bank, Andhra Pragathi Grameena Bank, Andhra Pradesh was studied to validate this statement for providing earnest money deposit (EMD) of more than INR two million for implementing a technology solution under Financial Inclusion Technology Fund (FITF). The RFP is available at http://www.apgbank.com/FInancial%20Inclusion%20Under%20End%20to%20En d%20Solution.doc. In the case of other three public sector banks, the EMD was INR one million. However other clauses, like minimum turnover of INR 50 million to 200 million, and in some cases going up to INR one billion for a company to be eligible for bidding, make it impossible for small MIS vendors. Some of the RFPs studied are (accessed on 27 Nov 2012): http://www.syndicatebank.in/downloads/FICO_RFP-Karnataka_Goa07052012.pdf http://www.canarabank.com/Upload/English/Tenders/RFP-for-implementation-ofFI-Solution080111.pdf

154

Chapter Seven

http://www.mahagramin.in/downloads/RFP_FIP.pdf Names not mentioned for the reasons of confidentiality and request by the respondent. 6

CHAPTER EIGHT THE EVOLVING INDUSTRY FOR MICROFINANCE SOFTWARE: EVALUATION AND GUIDE FOR MFIS AND MIS VENDORS ARVIND ASHTA, VITALIE BUMACOV, MIKHAIL CHERKAS AND DINOS CONSTANTINOU

Introduction Management Information Systems (MIS) are a key ingredient to achieving scale in the financial services industry—and this is equally true in the microfinance segment. This chapter deals with a number of related questions, such as: Can appropriate software help microfinance institutions scale-up and contribute to poverty reduction? What software is available in this dynamic and socially-oriented sector? How do these software vendors innovate to serve needs of their main customers—the MFIs? Microfinance is the provision of financial services to the poor, to micro entrepreneurs and to microenterprises that currently are financially excluded. This movement has grown rapidly over the last four decades since its inception in Brazil, Bolivia and Bangladesh. It has been crowned with a Nobel Prize for a Bangladeshi microfinance institution and its founder, Muhammad Yunus. Today Microcredit Summit reports over 3500 MFIs serving well over 200 million borrowers. With an average family size of five, the microfinance industry serves over 1 billion people (Maes and Reed, 2012). However, it is considered that there are over 4 billion poor people who are unbanked and therefore a large gap still exists. Many of those lacking access to financial services are self-employed entrepreneurs and small businesses that are difficult to serve as they

156

Chapter Eight

operate in the informal sector. Facilitating access to this market segment is therefore crucial not only from an income-smoothing perspective but, more significantly, from a broader perspective of economic development. A solid MIS backbone combined with adjacent technologies (such as the case of M-Pesa 1 in Kenya) is fundamental in allowing MFIs to deliver financial products and services to a large number of micro and small businesses. Innovation is a key driver in serving better and more customers in developing markets. Technological innovation can reduce transaction costs and facilitate growth, thus helping MFIs to achieve financial sustainability (Rosenberg, Gonzalez et al., 2009). Innovative technologies include mobile banking, online lending and MIS (Ashta, 2011). The chapter provides a brief literature review of innovation theory, followed by a note on the role of management information systems in microfinance. This in turn leads onto a comparative analysis of MIS software packages adapted for the microfinance market, based on our research. Finally, the chapter ends with an attempt to distil some recommendations for the benefit of MFIs and vendors, provide some general conclusions and highlight possible, future research directions.

Literature review Traditionally, innovation has been classified into incremental innovation and radical innovation (Abernathy, 1978). Initial research looked at the strategy and structures required for innovation and considered that innovation was radical if technology was new to the unit or its referent group. The risks of radical innovation could more easily be assumed by centralized hierarchies who were committed to technological change, as opposed to decentralized managers who preferred incremental innovation or new product adoption (Ettlie, Bridges et al., 1984). Within the radical innovation category, a distinction has been made between disruptive innovations that provide a challenge to incumbent players (Christensen, Bohmer et al., 2000), and sustaining innovations that allow incumbents to maintain their dominant position in the market. Disruptive innovation usually requires changing the price-product equation so that the new markets are created (Christensen, Johnson et al., 2002) and the business model itself is changed (Christensen, 2006). Although the recent development of cloud computing is a disruptive innovation in the microfinance software market (Ashta and Patel, 2013, forthcoming), the bulk of the market continues to rely on off-the-shelf software (Barnett, 2011). Furthermore, the specific MIS needs of the microfinance industry

The Evolving Industry for Microfinance Software

157

have not been adequately researched so far. This paper seeks to fill the research application gap with respect to incremental innovation. Henderson and Clark (1990) introduced the concepts of architectural and modular innovation within incremental innovation. These emphasized the difference between changing components (modular) and changing links between components (architectural). According to them, radical innovation destroyed both modular and architectural knowledge, creating very different challenges for established firms. According to Starr (1992), an important form of incremental innovation is providing new features of existing products to cater to niche markets as well as existing customers. Companies that are able to provide incremental innovation on an on-going basis are more likely to attract customers than those who provide only high but stable quality of existing products. The resulting high profits facilitate further innovation, which in turn enhances customer retention and acquisition. It should be pointed out that the two forms of innovation are not necessarily mutually exclusive, with incremental innovation often providing the basic probing, learning and motivation required for radical innovation (Lynn, Morone et al., 1996; Kanter, 2010); while radical innovation can lead to a first-mover advantage that can later be consolidated through incremental innovation (Lynn, Morone et al., 1996). On the other hand, incremental innovation is usually lower-risk and requires less initial infrastructure than radical innovation (Treacy, 2004). An assumption of this study is that firms continuously innovate in both modular and architectural ways in an effort to retain or enhance competitiveness. The customers themselves adopt software based on criteria such as compatibility, complexity, relative advantage, ‘trialability’ and ‘observability’ (Rogers, 2003). According to this theory, diffusion occurs progressively among groups of potential users through established communication channels. The continuous innovation by vendors necessitates that the product remains compatible with other infrastructure that the customer has purchased. The importance of the customer in the innovation process has been stressed in the literature (Mazzarol and Reboud, 2008), notably the form of relationship between the supplier and a principal customer.

The Evolution of Microfinance Of 1,112 MFIs volunteering return on equity (ROE) data to the Microfinance Information Exchange (MIX)2 for 20093, 66% are profitable. Nevertheless, half4 of the MFIs have less than 10,000 borrowers, which

158

Chapter Eight

would imply a loan portfolio of less than US $7 million with an average loan size reported on the MIX of US $700. The average loan size in countries in South Asia is even smaller (US $150). The transaction size of the loan is so small that transaction cost becomes a significant obstacle for the borrower and the lender. This inefficiency, in addition to the relatively low margin on borrowing, squeezes out room for profits (Ashta, 2009; Rosenberg, Gonzalez et al., 2009). As a result, managing the large client base that comes with growth poses a major challenge for MFIs. In particular, the effective management of information about large numbers of clients is a key aspect of effective underwriting, monitoring, collection and credit risk management for microfinance organizations. Most financial institutions that deal with retail and small business segments use MIS for client relationship management, credit analysis loan monitoring, collection, and so on. An important role of microfinance MIS is to support the scaling-up of MFIs. Microfinance software solutions are information systems that assist management in decision-making, reporting, control and other key aspects of the management process. In the hierarchy of the management process, information systems are near the base. They are support mechanisms that allow organizations to pursue their mission, vision and objectives— providing input for strategy formulation, management and control. In other words, the design of appropriate information systems supports decisionmaking in strategic planning and control (Edstrom, 1973). Furthermore, an important component of any MIS is the database from which it draws information. Management information systems, therefore, might be described as tools for converting raw data into usable information to facilitate the management of organizations, such as financial institutions. In view of the above, it is not surprising that most successful microfinance institutions indicate that an effective MIS is high on their list of key success factors. For example, the website 5 of one Indian MFI outlines the role of technology and information systems in facilitating growth: “SKS firmly believes that Technology is one of its biggest differentiators in the industry. … The systems designed and deployed at SKS have enabled the business to grow to nearly 5.3 million customers and are providing the technology foundation to achieve the next phase of growth. SKS has designed and deployed a web-based Business Intelligence portal using state-of-art technology and a highly flexible and scalable platform to support the business growth and operations.”

The Evolving Industry for Microfinance Software

159

When implemented and managed well, a sound information system can help MFIs improve reporting leading to lower risks 6 , lower costs, increased productivity, better decisions (Barua, Kriebel et al., 1991) and result in lower prices for customers. Informed decisions lead to better economic and social performance, while lower costs can enhance an institution’s profitability and/or competitiveness. For some, there is a trend to consider IT as a strategic instrument (Barua, Kriebel et al., 1991; Brown, Gatian et al., 1995). On the other hand, the rapidly improving performance and falling costs of IT—together with outsourcing opportunities and the availability of cheap, off-the-shelf, standardized software (Clemons, Reddi et al., 1993)—may have reduced the strategic importance of IT (Clemons and Row, 1991). There are some peculiarities of MIS for microfinance, which imply that the information needs of MFIs differ to some extent from those of mainstream commercial banks. These peculiarities arise from the nature of the microfinance industry, namely with respect to human resources, infrastructure, IT support and standardization (Iyengar, Quadri et al., 2010). While large MFIs may be in a position to draw on internal and external IT experts to customize microfinance software in accordance with their specific needs, most small MFIs cannot afford this expense and have to make do with off-the-shelf software or software that can be adapted at low cost. This paper looks at the diversity of existing microfinance software solutions from the point of view of the specificities of the microfinance industry. The research question has relevance for MFI managers seeking guidance in their MIS software selection.

Research methodology and sample description A sample set of microfinance software packages was analyzed with respect to the responsiveness of the individual solution to the specific needs of MFIs. The research methodology is based largely on CGAP 7 reviews of a number of MIS solutions widely used by the microfinance industry. The analysis, therefore, is limited by any errors or inaccuracies that may have been present in the CGAP database. MIS solutions targeting the microfinance market appear to be relatively new, with 72% of MIS packages in our sample first released after the year 2000. The oldest MIS in our list appears to be Emortelle (formerly CUMIS Plus). It was released in 1986, initially to serve cooperatives and credit unions in Trinidad and Tobago on a commercial basis. A second, pioneering MIS, the FAO-GTZ MicroBanking System, was released in

160

Chapter Eight

1988. The market for microfinance software solutions has taken off since that time, with almost a hundred providers being registered in the CGAP database (the incentive for inclusion being the opportunity to raise market awareness for their product). Twenty-five of the products reviewed by CGAP since 2009 were selected for the purposes of our study. Using this dataset, we calculated correlation coefficients and tested their statistical significance (see Table 6). Of course correlation does not prove causation, especially since the sample was small and only a few significant results emerged from the analysis. A brief description of each MIS is shown in Table 1 below. The product versions provide an idea of the frequency of upgrading required in the software industry—generally once or twice a year. The changing needs of growing organizations have major implications for management information systems, with software manufacturers in a competitive market under constant pressure to develop new versions to retain customers. This is in line with the observation that building a responsive IT infrastructure is the ‘number one’ concern of managers (Brancheau, Janz et al., 1996). Company

Product Name

Aquadev (Belgium) Avmei Cia. Ltda. (Ecuador) Cagecfi (Togo) Craft Silicon (Kenya) Crystal Clear Software Ltd (Uganda) Fao-Gtz Programme (Thailand) Fern Software (UK) Fern Software (UK) Fundamicro Foundation (San Salvador) Gradatim It Ventures (India) Grameen Foundation (USA) Infrasoft Technologies Limited (UK) Mediasoft (Togo) Mfsi (Guatemala) Cobiscorp (Ecuador) Micro Software Designs (Trinidad And Tobago) Neptune Software Plc. (UK)

Adbanking Conexus Perfect Bankers Realm Loan Performer

Year Reviewed Release First Version Date Release 2003 3.0.4 06/2008 2002 Econx 2007 2002 6.1 01/2009 2000 3.0 10/2008 1998 7.10.11 02/2009

Mbwin

1988

Abacus Oneworld Cubis Sim / Sim.Net

2004 2004 2000

Mfresolve Mifos Omnienterprise For Microfinance Microfina Siem Cobis Emortelle (Cumis Plus) Orbit

4.7.0

11/2008

1.08.10 8 3 (SIM.Net) 2007 1.1 2006 1.3 2001 3.0.01

03/2009 06/2008 01/2007 12/2008 07/2009 01/2009

2004 1990 1995 1986

2.8 7.3 3.5 8

09/2009 2008 04/2008 06/2009

2005

503_13

01/2009

The Evolving Industry for Microfinance Software Company

Product Name

Octopus Microfinance (France)

Octopus Microfinance Suite Delphix

Sathguru Management Consultants Pvt Ltd (India) Sigma Data & Computers Finance Solutions Ltd. (Uganda) Snowwood Infocom Mfasys Technologies Pvt Ltd (India) Southttech Limited Southtech Ascend Banking Sysde (Costa Rica) Sysde Saf Technical Development Kredits Solutions, L.L.C. (USA) Microfinance & Banking Software Top System S.A. (Uruguay) Topaz

161

Year Reviewed Release First Version Date Release 2007 2.5.8 04/2009

2001

5.1

01/2009

2003

8.36

01/2009

2007

3.0

04/2010

2002

5.0.0

10/2008

1999 1998

3.2 5.54

03/2009 02/2009

2001

5.3

02/2009

Table 1: Sample description.

Findings: Specific needs and market fragmentation Comparison of these products brings to the fore the diversity of the software—which can be explained by the broad needs of the microfinance industry, on the one hand, and the development stage of the microfinance software market, on the other. A discussion of these two aspects is followed by a description of the results of the quantitative correlation analysis.

1. Specificities of MIS software resulting from the diversity of the microfinance market Table 2 provides an idea of the product characteristics and helps explain the diversity of MFI needs. For the sake of simplicity, a subset of MIS was selected—namely, (CGAP reviewed) systems that had been launched in the period 2001 to 2003. The relatively long presence of these MIS in the market meant that their manufacturers had more time to adjust their software to reflect market needs (e.g. through the development of required functionalities) but also to distinguish their product from those of competitors.

Perfect

Finance Solutions AD Banking CONEXUS

Southtech Ascend TOPAZ

MIS Name: / MIS Feature: Loan portfolio management Individual loans Solidarity groups with individual loans Solidarity groups with group loans Credit scoring Savings management Savings accounts Group savings Insurance Accounting: Chart of accounts management Finance Budget management Treasure management Assets management Reporting Operational reporting Regulatory reporting Financial reporting

Chapter Eight

OMNI Enterprise Delphix

162

YES YES YES YES YES YES YES YES YES YES YES YES Part Part YES YES YES YES YES YES YES YES YES YES Add NO Sub YES Part NO YES NO YES YES Part YES

YES NO YES YES

YES YES Part YES

YES Part Sub YES

YES Part Part YES

YES Part Part YES

Add YES Part Sub Part NO Add NO Add YES Part NO Add YES Add Part YES NO

YES YES NO YES

YES YES Part YES

Part YES Part Part Part Add

YES Part YES YES YES Part YES YES YES YES Part YES Part Part YES Part YES YES YES YES YES Part YES YES

Part – functionality partially covered; Add – additional module has to be purchased; Sub – recommended supplier provides the feature.

Table 2: Coverage of MIS features. The MIS needs of MFIs include functionalities such as accounting, credit portfolio management, deposits management, insurance, payroll, and reporting. Some MIS provide most or all of these functions. Others only provide the basic functions of accounting, credit portfolio management, deposits management and simplified reporting. The remainder provide a range of functionality somewhere between the two ‘extremes’ mentioned above. Payroll is often not included, as it varies from country to country depending on legal and fiscal specificities that render parameterization difficult. While some software products only target French-speaking or Englishspeaking countries, many packages support several languages and most (widely-used) scripts 2 . Some software packages are suitable only for 2

A script is a writing system (e.g. Greek, Latin, Chinese and Arabic scripts).

The Evolving Industry for Microfinance Software

163

cooperatives and credit unions; several MIS solutions cater for the specific needs of MFIs; while some products can even satisfy the needs of mainstream commercial banks. Systems specifically developed for MFIs are usually designed with individual lending, group lending and village banking in mind. Nonetheless, some packages can only be used for individual lending—a feature inherited from mainstream banking systems that makes them unsuitable for group lending and village banking.

2. Diversity of MIS software owing to controllable factors Table 3 below provides outreach indicators of the MIS software. Newer solutions in general have fewer customers than older ones. In microfinance, it is common to divide the market into six zones: Europe and Central Asia, East Asia and the Pacific, South Asia, Latin America and Caribbean countries, Middle East and North Africa, and Sub-Saharan Africa. One MIS in the review list has clients in all the zones and two others in five of the six zones. One hypothesis that may explain the global strategies of some software manufacturers—and in turn may influence the degree of customer adoption—is whether other MFIs with similar business models are using the same software. For example, if a Grameen Bank replicator in one country is using one MIS, it may lead other Grameen Bank replicators to adopt the same software solution in the knowledge that the package in question ‘works’ at similar institutions. Thus the "Schumpeterian" profits of innovation in software technology get diffused rapidly and globally along the lines suggested by Cantwell and Santangelo (2000), not because governments encourage replication but rather because related organizations follow similar models around the world (akin to global corporate networks). Software manufacturers who are new or who have not been able to implement successfully in one MFI, may find themselves barred from other MFIs with similar models. Product

ADBanking Conexus Perfect Bankers Realm Loan Performer MBWin Abacus Oneworld

Year Smallest - Clients Geogra Technical Launc biggest MF (institut phical support hed client* ion) zones staff** 2003 0.5 – 60 24 1 8 2002 1.5 – 60 81 1 2 2002 0.5 – 35 28 1 7 2000 1 – 250 124 5 85 1998 0.3 – 100 220 6 5 1988 0.1 – 31 138 5 4 2004 0.5 – 200 52 4 2

164 Product

Cubis Sim / Sim.Net MFResolve Mifos Omnienterprise For Microfinance Microfina SIEM Cobis Emortelle (Cumis Plus) Orbit Octopus Microfinance Suite Delphix Finance Solutions Mfasys Southtech Ascend Banking Sysde Saf Kredits Microfinance & Banking Software Topaz

Chapter Eight Year Smallest - Clients Geogra Technical Launc biggest MF (institut phical support hed client* ion) zones staff** 2004 0.5 – 20 111 4 2 2000 0.1 – 18 14 1 2 2007 5 – 80 7 1 10 2006 0.5 – 265 8 4 3 2001 28 – 1,000 26 4 ? 2004 1990 1995 1986 2005 2007 2001 2003 2007 2002 1999 1998

0.5 – 14 1 - 100 2 – 3,000 1.2 – 140 5 – 2,000 0.5 – 15 1 – 1,500 1 – 10 14 – 75 2 – 8,000 5 – 305 1 – 600

23 34 33 132 27 12 3 102 6 9 42 39

1 4 1 1 1 3 1 2 1 2 3 5

1 6 26 8 10 3 30 2 4 32 11 3

2001

6 – 2,500

8

1

7

*Real data and estimates (in thousands); **Real data and estimates.

Table 3: MIS outreach. Table 4 below provides pricing information, which varies significantly from one MIS to another. The sales revenues of an MIS developer come from four essential components: license fees, maintenance fees, customization services and training fees. The initial license can be free or cost up to US $400,000 alone. The license can be perpetual or expire after a year or two, requiring the customer to pay a subscription. The license cost can also include the license fee of the database (DB) software (if the MFI does not have it already), plus the Operating Systems (OS) of the server and the workstations, if a different or updated OS is required. As the customer base of the MFI grows, the DB and the OS must keep pace. At a certain threshold of operations, the DB software will require a significant cost that needs to be considered. The same principle applies for the OS of the computers that will host or use the application. Some MIS require one OS for the servers and a different OS for the workstations. Although free or low cost OS are available on the market, MIS may not be compatible. This supplementary cost has also to be considered by the MFI.

The Evolving Industry for Microfinance Software MIS Name / Price item Legal status Location of clients (geo. zones) MIS source code License price

Perpetual license Maintenance per year (average)

Loan Performer Company 6

Copyright

MBWin

Mifos

SIEM

165

Kredits

NGO 5

NGO 4

Company 4

Company 5

Copyright

Open Source

Copyright

Copyright

1.500 USD / user (diminishes with value)

30,000 USD

900 USD / 500 – 5.250 0 user USD / branch (diminishes with volume, free till 500 clients) Renews with Yes Yes maintenance fee 20% of 2,500 USD 22,500 USD license fee first year, than 20% of license fee 4,000 USD 90,000 USD 75,000 USD

Renews with Renews with maintenance maintenance fee fee 180 USD / 3,000 USD user (diminishes with volume) 27,500 USD 50,000 USD

Installation cost (training included) Sys. admin. 900 USD 2,000 USD included in 3,000 USD included in training installation installation (without travel costs costs costs) Average cost 300 USD 550 USD 400 USD 480 USD 400 USD of programmer man-day man-day man-day man-day man-day customization OS required Windows Windows Windows Windows All for XP/2000, + XP/2000/Vist XP/Vista, + 2000, + workstations a, + OS required Windows Windows Ubuntu (8.04 Windows Windows for server 2000 server, + 2000 server, + LTS Server) / 2000 server, + 2000 server, + Windows 2003 server, + Database Visual Fox SQL Server MySQL 5.0, + SQL Server SQL Server software Pro 9.0 2000, + 2000, + 2005, + (included) / SQL Server 2000, + Database Included Available free Available free Available free Available free software (with (with (with (with license limitations) limitations) limitations) limitations) “+” means that superior versions of the product are also compatible.

Table 4: Example of pricing strategy.

166

Chapter Eight

A zero MIS license fee encourages trial by small MFIs who can then be locked-in after they commit costs for developing, learning and using the MIS (‘sunk costs’). The maintenance cost is often a percentage of the licensing cost, up to 25-30% in our sample. However, where licensing cost is zero, the maintenance cost is usually expressed in dollar terms. To these general services are added the initial training costs, which are more or less essential for the MFI to start operating. Since most MFIs have a legacy system prior to the MIS acquisition and implementation, data migration and customization will generate additional costs to the MFI and additional revenues to the provider. MFIs may try to perform these tasks using their own IT resources, but most require external consultants at additional cost. The difference in pricing strategies makes the software difficult to compare. Table 4 compares several strategies. We selected a subset of five MIS that have spread networks of customers (four geographical regions or more) and do their business exclusively with microfinance institutions. Table 4 presents the pricing strategy only for MIS solutions hosted onsite. Increasing numbers of MIS developers propose the application service provider (ASP) model or a SaaS model where pricing is based on such metrics as the number of transactions or number of clients in the MIS, or active accounts in the system during the year. Vendors provided the cost of their MIS software under three market scenarios. The first scenario is based on a “small MFI” offering both loan and savings products, serving 15 thousand clients, with up to 45 simultaneous users in 10 branches and 80% of operations in urban areas. The other selected scenario presupposes a “large MFI” with 100,000 clients, loans, savings and money transfer products, up to 220 simultaneous users in 50 branches and 50% of operations in urban areas. Product

ADBanking Conexus Perfect Bankers Realm Loan Performer MBWin Abacus Oneworld Cubis Sim / Sim.Net MFResolve

Average cost* per scenario (thousand USD) Small MFI Large MFI 53.5 200.0 25.0 104.0 77.3 178.5 20.0 155.0 16.9 42.6 112.5 244.0 350.3 912.0 28.0 Abacus 42.5 54.5 44.3 46.3

The Evolving Industry for Microfinance Software Product

Mifos Omnienterprise For Microfinance Microfina SIEM Cobis Emortelle (Cumis Plus) Orbit Octopus Microfinance Suite Delphix Finance Solutions Mfasys Southtech Ascend Banking Sysde Saf Kredits Microfinance & Banking Software Topaz

167

Average cost* per scenario (thousand USD) Small MFI Large MFI 62.5 97.5 112.0 342.0 43.0 97.0 93.7 315.3 340.0 736.0 N/A N/A 365.8 544.0 60.0 N/A 74.7 337.0 16.0 83.0 29.8 146.2 194.0 483.0 152.8 361.8 90.9 153.0 220.0 400.0

* As declared by the provider or estimated based on his pricing

Table 5: Estimation of the cost of MIS acquisition depending on the supplier and the scenario. As we can see, there is a strong difference in cost ranges, according to the CGAP reviews. Nevertheless, we can see that off-the-shelf software has pricing that ranges from US $16,000 to US $366,000 for small institutions. For large institutions, the price can reach US $900,000 for a full range of functionalities including those of mainstream banking institutions. It is obvious that cheaper solutions can lack functionalities that are mandatory for the business model of the MFI, such as the ability to handle several currencies. For an MFI that has disbursed loans in dollars, euro and local currency, a low price is irrelevant if the MIS cannot keep track of transactions in different currencies.

Analysis of relationships between factors The data in our sample of MIS was tested for correlation. Comments are restricted to the significant relationships. Although correlation coefficients do not reveal the direction of causality, they do allow hypotheses to be made. A binary categorization was employed to carry out the analysis. A forprofit MIS vendor was assigned to category “1”, while a not-for-profit

168

Chapter Eight

organization was assigned to category “0”. Correlations with this factor are presented in the column 3 of the Table 6. A negative significant correlation of this factor was observed (r = -0.48) with the share of the vendor’s business conducted with the microfinance institutions. This may be because these organizations are already vendors to banks and their sales to MFIs are niche products with small amendments to the banking software. For-profit vendors implement their solutions faster (r with implementation time -0.44), work with larger MFIs (r = 0.5), have better reporting functionalities (r = 0.56) and have analytical tools (r = 0.46). Column 4 of Table 6 shows the correlation of each factor with the estimated share of each MIS vendor’s business conducted in the microfinance market. Significant correlations are mainly related to the cost of the MIS. Microfinance-oriented vendors tend to have lower overall MIS costs (license, implementation, training and maintenance) and the cost of the license alone tends to be lower. Column 5 of the same table shows how each factor correlates with overall MIS costs. MIS cost correlates obviously with the cost of the MIS license and the cost of implementation. MIS cost correlates with the upper class of the targeted MFIs. The overall cost of the MIS correlates with the scale of the MIS vender as represented by the number of employees. Column 6 shows the correlations of each factor with the MIS license cost. The licenses cost correlates with same factors as the overall MIS cost; however, it also correlates significantly with better evaluation of the MIS by the CGAP reviewers, better functionality and better management reporting. A relatively strong positive correlation was found between the number of years of presence of the MIS on the market and the number of microfinance institution clients. This fact appears obvious as the number of clients is likely to increase with time, but it may also be due to increasing trust in vendors who are considered more likely to survive. Time in the market (date of survey minus year MIS launched) also shows a significant positive correlation (r = 0.45, sig at .028 level) with number of clients for the product.

The Evolving Industry for Microfinance Software

169

For Share MIS Licen Impl Impl Time Cli. MF HR Funct Repo Tech. Profit MF Cost se em. em. on Scale Cli ionali rting Cap 8 Biz Cost Cost Time Mrkt ty (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (14) (15) (16) Share Cor. -.48* 1 MF Sig. .020 9 Biz N 21 MIS Cor. .29 -.55* 1 Cost10 Sig. .198 .010 N 21 21 Licens Cor. .43 -.54* .88** 1 e Sig. .052 .012 .000 Cost11 N 21 21 20 Imple Cor. .09 -.32 .74** .66** 1 m. Sig. .685 .163 .000 .001 Cost12 N 21 21 20 21 Imple Cor. -.44* .30 -.25 -.23 -.18 1 m. Sig. .045 .181 .286 .323 .438 13 Time N 21 21 20 21 21 Time Cor. .02 .14 .05 .03 .13 .42 1 on Sig. .940 .536 .833 .936 .582 .058 Mrkt14 N 24 23 21 21 21 21 Cli. Cor. .50* -.39 .44* .5* .31 -.5* .15 1 Scale Sig. .012 .063 .047 .020 .172 .020 .493 15 N 24 23 21 21 21 21 24 MF Cor. .09 .38 -.25 -.29 -.15 .11 .45* -.24 1 Cli16 Sig. .659 .070 .281 .206 .516 .644 .028 .266 N 24 23 21 21 21 21 24 24 HR17 Cor. .33 -.58** .57** .53* .37 -.38 .04 .43* -.16 1 Sig. .124 .005 .009 .017 .111 .099 .866 .038 .475 N 23 22 20 20 20 20 23 23 23 Functi Cor. .32 -.16 .39 .45* .11 -.06 -.01 .5* .09 .26 1 onality Sig. .118 .468 .079 .041 .629 .792 .968 .013 .677 .230 18 N 24 23 21 21 21 21 24 24 24 23 Report Cor. .56** -.39 .4 .54* .23 .01 .36 .49* .23 .32 .73** 1 ing19 Sig. .005 .062 .074 .012 .319 .959 .081 .015 .269 .140 .000 N 24 23 21 21 21 21 24 24 24 23 24 Tech. Cor. -.15 -.15 .16 .1 -.04 -.40 -.6** .19 -.41* .2 .38 -.07 1 Cap20 Sig. .474 .502 .477 .680 .869 .072 .002 .370 .049 .361 .063 .744 N 24 23 21 21 21 21 24 24 24 23 24 24 Analit. Cor. .46* -.75** .43 .46* .18 -.25 -.18 .32 -.48* .46* .3 .53* .23 Tools Sig. .036 .000 .068 .049 .455 .292 .423 .152 .028 .042 .185 .013 .325 21 N 21 20 19 19 19 19 21 21 21 20 21 21 21 Cor. – indicates the Pearson correlation coefficient; Sig. – indicates significance level (2tailed); N – indicates the number of observations; * - correlation significant at the 0.05 level; ** - correlation significant at the 0.01 level. Variable

Table 6: Correlation matrix

170

Chapter Eight

Recommendations Based on the findings of this study and on the researchers’ knowledge from working with different MIS stakeholders, the following recommendations made for MFIs looking to purchase an appropriate MIS as well as for MIS software manufacturers to cater better to the needs of this segment.

1. Buyer (MFI) considerations in acquiring new MIS solutions 1. Strategy and IT Fit: an MFI should formulate its strategy (more precisely operational strategy) clearly, and evaluate its existing resources (more precisely IT resources) in order to choose and maintain a particular MIS that will match the MFI’s operations. This is particularly true because different MIS offer different modules and the MFI needs to know that if it changes its strategy compatible modules will be available. 2. Direct and Indirect Costs: MFI should consider all direct and indirect costs not only for buying, but also for maintaining and upgrading the MIS. Our review of the MIS software manufacturers shows pricing can be quite opaque and that implementation and maintenance costs may be important when purchase cost of the software is low. 3. Control of Data: MFI should evaluate future possibilities to migrate from the MIS they plan to purchase now onto a more advanced MIS. The particular concern is the database that the MIS uses to store transactional and other information. A “closed” or proprietary database may make the migration process very difficult or even impossible, and may bind the MFI to one MIS vendor. A database that is not fully controlled by the MFI, for example for database backup, may create problems for the whole business contingency planning of the MFI. 4. Technical Support: as the MIS is a key system for organization and management of the MFI operations, technical support, especially the capability of the vendor to react to serious MIS problems becomes very important for the MFI. Thus, the capacity and capability of the vendor‘s technical support team—and its reaction time to problems—may become a serious consideration in choosing a particular MIS solution. It is important to note that if the MFI is a niche customer for a banking vendor, it may not get the kind of support that the vendor’s large banking clients would get. 5. MIS Development: the MIS solution should be able to encapsulate all modifications caused by main trends of microfinance market, main local regulation requirements (in case they are applicable to microfinance), as well as changes required by the MFI’s business growth and expansion.

The Evolving Industry for Microfinance Software

171

Thus, it becomes very important for the MFI to understand the development history of the MIS, including new functionality added with previous MIS releases, and future plans for the MIS development presented by the vendor.

2. Vendor (software manufacturer) considerations in MIS development 1. Linking demand diversity to vendor capabilities. Vendors should consider the diversity of the MFI’s needs that it is trying to satisfy. Having many different MFIs in the portfolio may create a situation where the vendor cannot cope with the diversity of these demands. For example, MFIs that uses different operating models provide quite different palettes of products and services, use different approaches to business organization, and apply a range of technologies to provide access to their services. In fact, the different versions of the MIS the vendor supports owing to forking or branching (for customer-specific modifications), may evolve into multiple, discrete products in the portfolio that the vendor is obliged to service. The vendor may simply not have enough capacity to develop and maintain correctly the family of sibling systems. Over time, they will need to apply incremental innovations on-demand into multiple ‘fork’ versions. 2. Market perspectives evaluation. Apart from the needs of current client institutions, the vendor needs to evaluate main trends on the microfinance and financial markets, as well as new technologies entering and expanding in these markets, in order to develop the functionalities to be provided. This tracking of trends and technologies within the microfinance market leads to the incremental modular innovation that we see with each release of a new MIS version. 3. Integration capability. The vendor should consider the development of the MIS integration functionality in order to be able to connect its MIS with other third-party software. This, for example, may allow the vendor to use local accounting software solutions to facilitate localization of the MIS, for example to provide local regulatory reports. Moreover, the ability to integrate the MIS with solutions from other vendors working with MFIs, or with banking software present in a particular market, may significantly increase the attractiveness of the MIS. This is architectural innovation as it provides links between the modules of different software solutions. 4. Critical mass growth. The vendor needs to achieve a sizeable market share in order to realize the benefits of economies of scale. This may be achieved either by increasing the number of installation in one of the

172

Chapter Eight

neighbor markets (countries) or by acquiring smaller players that are also supplying software to MFIs, in order to expand the customer base.

Conclusions Microfinance is a fast growing industry where management information systems constitute a strategic imperative for the growth of microfinance institutions—by facilitating geographical expansion, product outreach and economic sustainability. The research carried out in the course of this study found a large diversity of MIS software solutions targeting the microfinance sector. This diversity arises partly from the variety of business models present in the microfinance industry, which in turn leads to different information needs. Software diversity is reflected in the different product features offered to clients, operational approaches in use (e.g. group vs. individual lending, village banking vs. retail banking), as well as other factors such as regulatory systems and differing functionalities responding to the needs of MFIs of different size. Another factor behind product diversity is the early stage of development of the microfinance software market. MIS vendors are essentially trying to address and satisfy the relatively broad scope of (often diverse) needs of different institutional clients—an aspect driven by different business models and commercial strategies (notably pricing). MFIs and other institutions operating in the microfinance arena are developing various business models and strategies to expand market outreach, increase business efficiency and reduce costs. The relative immaturity of the microfinance market compared to other client segments (such as retail and corporate) means that the microfinance sector has been changing at a faster rate than the more traditional part of the financial services industry. MIS vendors try to follow challenging and diverse requests that come from MFIs, an indicator of their corporate social responsiveness. Many vendors are working to accumulate critical mass and gain economies of scale in the microfinance market by expanding the geographical areas they cover and increasing the number of their clients. There are software vendors that have more than 100 installations of their MIS on the microfinance market. There are also several “internal” MIS vendors (that serve only a particular group of MFIs) that develop and implement software for microfinance networks, such as Grameen, Opportunity and Vision Fund. So far the large players of the software industry have not intervened. Large software vendors that provide their

The Evolving Industry for Microfinance Software

173

services to the formal financial sector have very little interest in this market at this moment. Parallel developments on both sides of the market (demand for, and supply of, microfinance software) are characteristic of the fast-changing microfinance industry. On the demand side (for microfinance software), MFIs are expanding their operations and developing capabilities to manage more complex and larger businesses. MIS vendors serving the microfinance market, on the other hand, follow a similar expansion pattern to keep up with the development of their clients (i.e. MFIs and other actors providing financial services to the microfinance segment of the market). The future of the software market for microfinance is likely to be very dynamic, not least in the light of the emergence of new technologies such as cloud computing. As the microfinance market evolves, so will the software market for microfinance. In the future, a consolidation process is likely to take place in both markets (microfinance and MIS for microfinance). Organic growth, consolidation and the entry of new and (often) larger players in the microfinance market is likely to motivate larger software vendors to take an interest in the microfinance MIS market. How players in both markets react and adapt to increasing and changing competition will be an interesting subject for further research. Another area of research may relate to innovations in handling the diverse reporting needs of the broad range of stakeholders present in the microfinance sector. While existing off-the-shelf software provides data that can be parameterized, lenders, donors and regulators tend to require reports in in their own specific format—which in turn poses a major challenge for MFIs and their MIS. On a closing note, the rapidly evolving information needs of the microfinance sector, combined with the fast pace of technological change affecting the broader financial services industry, are likely to prove rich pastures for academic research in coming years.

Bibliography Abernathy, W. J., (1978), The productivity dilemma: Roadblock to innovation in the automobile industry, Johns Hopkins University Press, Baltimore, MD. Applegate, l. M. and Elam, j. J., (1992), “New Information Systems Leaders: A Changing Role in a Changing World”, MIS Quarterly Vol.16, No. (4), p. 469-490.

174

Chapter Eight

Ariyachandra, T. R. and Frolick, M. N., (2008), “Critical Success Factors in Business Performance Management - Striving for Success”, Information Systems Management Vol.25, No. (2), p. 113-120. Ashta, A., (2009), “Microcredit Capital Flows and Interest Rates: An Alternative Explanation”, Journal of Economic Issues (M.E. Sharpe Inc.) Vol.43, No. (3), p. 661-683. Ashta, A., Ed. (2011), Advanced Technologies for Microfinance: Solutions and Challenges, Hershey, PA, IGI Global. Ashta, A. and Patel, J., (2013 forthcoming), “ Software as a Service: An opportunity for disruptive innovation in the Microfinance software market?”, Journal of Innovation Economics. Barnett, B., (2011), Understanding MFIs’ Reporting Burden: A study of the reporting burden for microfinance providers in sub-Saharan Africa. MicroBanking Bulletin, MIX. Barua, A., Kriebel, C. H. and Mukhopadhyay, T., (1991), “An Economic Analysis of Strategic Information Technology Investments”, MIS Quarterly Vol.15, No. (3), p. 313-331. Brancheau, J. C., Janz, B. D. and Wetherbe, J. C., (1996), “Key Issues in Information Systems Management: 1994-95 SIM Delphi Results”, MIS Quarterly Vol.20, No. (2), p. 225-242. Brown, R. M., Gatian, A. W. and Hicks Jr., J. O., (1995), “Strategic Information Systems and Financial Performance”, Journal of Management Information Systems Vol.11, No. (4), p. 215-248. Cantwell, J. and Santangelo, G. D., (2000), “Capitalism, profits and innovation in the new techno-economic paradigm”, Journal of Evolutionary Economics Vol.10, No. (1/2), p. 131. Carroll, A. B., (1979), “A Three-Dimensional Conceptual Model of Corporate Performance”, Academy of Management Review Vol.4, No. (4), p. 497-505. Child, J., (1987), “Information Technology, Organization, and the Response to Strategic Challenges”, California Management Review Vol.30, No. (1), p. 33-50. Christensen, C. M., (2006), “The Ongoing Process of Building a Theory of Disruption”, Journal of Product Innovation Management Vol.23, No. (1), p. 39-55. Christensen, C. M., Bohmer, R. and Kenagy, J., (2000), “Will Disruptive Innovations Cure Health Care?”, Harvard Business Review Vol.78, No. (5), p. 102-112. Christensen, C. M., Johnson, M. W. and Rigby, D. K., (2002), “Foundations for Growth”, MIT Sloan Management Review Vol.43, No. (3), p. 22-31.

The Evolving Industry for Microfinance Software

175

Clemons, E. K., Reddi, S. P. and Row, M. C., (1993), “The Impact of Information Technology on the Organization of Economic Activity: The "Move to the Middle" Hypothesis”, Journal of Management Information Systems Vol.10, No. (2), p. 9-35. Clemons, E. K. and Row, M. C., (1991), “Sustaining IT Advantage: The Role of Structural Differences”, MIS Quarterly Vol.15, No. (3), p. 275292. Edstrom, A., (1973), “CONCEPTUALIZATION OF THE DESIGN PROBLEM FOR MANAGEMENT INFORMATION SYSTEMS”, Journal of Management Studies Vol.10, No. (2), p. 118-132. Ernult, J. and Ashta, A., (2007), Développement durable, responsabilité sociétale de l'entreprise, théorie des parties prenantes: Évolution et perspectives. Cahiers du CEREN. Dijon, France, CEREN: pages 4-31. Ettlie, J. E., Bridges, W. P. and O'Keefe, R. D., (1984), “Organization Strategy and Structural Differences For Radical Versus Incremental Innovation”, Management Science Vol. 30, No. (6), p. 682-695. Freeman, R. E. and Reed, D. L., (1983), “Stockholders and Stakeholders: A New Perspective on Corporate Governance”, California Management Review Vol.25, No.(3), p. 88-106. Hamdouch, A. and Moulaert, F., (2006), “Knowledge Infrastructure, Innovation Dynamics, and Knowledge Creation/Diffusion/ Accumulation Processes”, Innovation: The European Journal of Social Sciences Vol.19, No.(1), p. 25-50. Henderson, R. M. and Clark, K. B., (1990), “Architectural Innovation: The Reconfiguration of Existing Product Technologies and the Failure of Established Firms”, Administrative Science Quarterly Vol.35, No.(1), p. 9-30. Iyengar, K. P., Quadri, N. A. and SINGH, V. K., (2010), “Information Technology and Microfinance Institutions: Challenges and Lessons Learned”, Journal of Electronic Commerce in Organization Vol.8, No.(2), p. 1-11. Kanter, R. M., (2010), “Block-by-Blockbuster Innovation”, Harvard Business Review Vol.88, No.(5), p. 38-38. Kotter, J. P., (2012), “Accelerate!”, Harvard Business Review Vol.90, No.(11), p. 43-58. Lynn, G. S., Morone, J. G. and Paulson, A. S., (1996), “Marketing and Discontinuous Innovation: The Probe and Learn Process”, California Management Review Vol.38, No.(3), p. 8-37. Maes, J. P. and Reed, L. R., (2012), State of the Microcredit Summit Campaign Report 2012. Washington, D.C., Microcredit Summit Campaign. .

176

Chapter Eight

Mazzarol, T. I. M. and Reboud, S., (2008), “THE ROLE OF COMPLEMENTARY ACTORS IN THE DEVELOPMENT OF INNOVATION IN SMALL FIRMS”, International Journal of Innovation Management Vol.12, No.(2), p. 223-253. Moulaert, F. and Hamdouch, A., (2006), “New Views of Innovation Systems”, Innovation: The European Journal of Social Sciences Vol.19, No.(1), p. 11-24. Prahalad, C. K. and Ramaswamy, V., (2004), The Future of Competition: Co-creating unique value with customers, Harvard Business School Press, Boston, MA. Rogers, E., (2003), Diffusion of Innovations, Free Press. Rosenberg, R., Gonzalez, A. and Narain, S., (2009), The New Moneylenders: Are the Poor Being Exploited by High Microcredit Interest Rates? Occasional Paper. Washington, D.C., CGAP. 15. Starr, M. K., (1992), “Accelerating innovation”, Business Horizons Vol.35, No. (4), p. 44. Tavakolian, H., (1989), “Linking the Information Technology Structure With Organizational Competitive Strategy: A Survey”, MIS Quarterly Vol.13, No.(3), p. 309-317. Treacy, M., (2004), Innovation as a Last Resort. Harvard Business Review, Harvard Business School Publication Corp. 82: 29-30. Uzunidis, D., (2010), Innovation et Proximité Entreprises, Entrepreneurs et Milieux Innovateurs. (French), Revue des Sciences de Géstion No. (241), p. 13-22.

Notes 1

A mobile-phone based money transfer service in Kenya that allowed microfinance borrowers to receive and repay loans using the mobile phone network, enabling the MFIs to reduce costs and thus offer lower interest rates. 2 Microfinance Information Exchange, Inc. (MIX) is a non-profit organization that acts as a business information provider in the microfinance sector – www.mixmarket.org. 3 Data downloaded on May 6, 2010. 4 Obviously the real percentage of smaller MFIs is much larger considering that most small and micro MFIs do not volunteer data to the MIX. 5 http://www.sksindia.com/technology.php consulted on April 16, 2009. Of course, SKS has other problems today but these are not related to their Information System. 6 MIS needs to respond to the major strategic risks: demand risk, innovation risk and inefficiency risk (Child 1987). 7 Consultative Group to Assist the Poor (CGAP) is an independent policy and research center, housed at the World Bank, dedicated to advancing financial access

The Evolving Industry for Microfinance Software

177

for the world's poor. It is supported by numerous development agencies and private foundations who share a common mission to alleviate poverty. 8 Indicates if the MIS vendor-developer is a for-profit institution. 9 Share of turnover generated by sales to microfinance institutions. 10 Cost of implementing the MIS based on the scenario “small MFI”. 11 Cost of license (only) based on the scenario “small MFI”. 12 Cost of implementation based on the scenario “small MFI”. 13 Declared average time of implementation based on the scenario “small MFI”. 14 Declared average time of implementation based on the scenario “small MFI”. 15 Superior boundary of the scale of targeted MFIs. 16 Number of current clients – MFIs. 17 Number of employees of the MIS developer – a proxy of developer’s scale. 18 Evaluation mark of the functionalities of the MIS given by CGAP experts. 19 Evaluation mark of the reporting functionality of the MIS given by CGAP experts. 20 Evaluation mark of the technical capabilities of the MIS given by CGAP experts. 21 Availability of analytical tools such as budgeting and cost control.

CHAPTER NINE THE OPEN SOURCE ATTITUDE IN MICROFINANCE: THE CASE OF AIRDIE VITALIE BUMACOV, FREDERIC LANET AND ARVIND ASHTA

Abstract We present a case study of a French microfinance institution Airdie that adopted Octopus—a computerized management information system based on open source software. This chapter provides insights about the advantages and disadvantages of embracing an open source attitude in microfinance. We state that management information systems based on open source software represent a viable alternative to proprietary systems in microfinance. Several free robust solutions are available. These have less functionality compared to most proprietary solutions, but could represent a stable starting point, especially if the microfinance institution has specific needs that require customization at the programming (core) level. We show that microfinance institutions are usually consumers (free riders) of open source software with rare cases of active participation. Airdie’s adoption of Octopus is such a rare case. The microfinance institution achieved partial success when deciding to further invest in the development of the Octopus. Octopus could satisfy most of Airdie’s management information needs at lower cost than the alternatives, but could not help the community benefit directly from new functionality. Lack of both funding and awareness prevents microfinance institutions from engaging actively in open source software initiatives. In such an environment, provision of subsidies and inexpensive technical assistance oriented at promoting collaboration amongst microfinance institutions

The Open Source Attitude in Microfinance: The Case of Airdie

179

around viable open source software initiatives could result in a snowballing positive effect. Key Words: Management information systems, open source software, microfinance, microfinance institutions.

Introduction In France 99% of the people living below the official poverty line have a bank account (CCSF, 2010); nevertheless, most of them suffer from other forms of financial exclusion. Financial exclusion penalizes lowincome populations by limiting their access to overdraft facilities and other forms of affordable credit. Microfinance institutions (MFIs) position themselves as providers of “alternative” financing, working with people and small businesses that banks and consumer credit companies refuse to finance, or whom they exploit by charging higher interest rates or imposing guarantee requirements that offset the limited liability of businesses. A similar need for “alternative” forms of serving the software market is the driving force behind the open source software (OSS) movement. The owners of proprietary software when in a monopolistic position charge prices that are out of the reach of low-income people, small businesses and institutions working with low-income populations. Shareware1 is limited in functionality or duration, restricting its utility for small MFIs or for the poor. Freeware2 remains rare, while recourse to pirated software is unsafe for the MFI or the low-income user, especially if the business depends on such software to generate income. Since both microfinance and OSS movements, developed from the need to provide alternative options to low-income people, it would seem natural that microfinance should favor OSS and vice-versa. However the use of free and open source management information systems (MIS) in microfinance remains limited. Explanations for this limited use of open source MIS in MFIs could include the immaturity of the technology, the lack of basic software skills, a limited number of choices, and a low number of active participants. The environment in which most MFIs operate does not facilitate active and significant contribution to OSS initiatives. Moreover, microfinance operations, even if usually of small value compared to commercial banks, require the same rigor as those practiced in mainstream banking. A one year micro loan of as low as US $50 generates at least 49 movements on the accounts of the MFI: one disbursal, 12 calculations of

180

Chapter Nine

the interest each month, registration of the payment of each of the 12 installments, 12 settlements of the incurred interest and 12 reimbursements of the principal. Mandatory savings, disbursal commissions, grace periods, penalties for late payments, weekly reimbursement plans and membership in joint liability groups can seriously complicate the architecture of the MIS. Banking MIS has been unable to respond to most of the needs of small MFIs that require stable solutions to allow them to increase in scale. This is because banking software comes with high license prices and unaffordable consulting, development and implementation fees. In spite of increased flexibility and parameterization options, specific to banking systems, such software could not support some specific microfinance functionalities such as credit disbursal to joint-liability groups, weekly reimbursement schemes, or the possibility of distinguishing between real (informal) and reported balance sheets and cash flow reports3. Cheaper low-end solutions have been developed especially to meet the needs of loan and saving associations, but these lack the robustness and many of the functionalities of the core banking systems—defined as advanced computerized MIS that automates the processing of (most) banking operations, including automatic updating of corresponding accounting books. Most core banking systems include a loan management module, a deposit management module, customer relationship management facilities, an accountancy module and reporting tools. All the components of the system are interconnected and synchronized. For credit institutions such as the MFIs and retail banks, the module that manages the loan portfolio is the most important. In 2004, 46% of MFIs surveyed by CGAP4 were using manual (paper) systems or spreadsheets to manage their loan portfolio. A slightly lower percentage was using custom-built software. The share of the off-the-shelf, packaged solutions for portfolio management was marginal (CGAP, 2009). A similar survey conducted four years later revealed a strong expansion of the packaged software that reached almost one third of surveyed MFIs (CGAP, 2009). Within this context two projects emerged dedicated to open source software for microfinance: Mifos and Octopus. Both MIS, built around the loan management tool, were initially released in 2006-2007 to serve the needs of specific microfinance networks and later proposed as free OSS. Airdie—a French MFI—was the first non-network affiliated institution to participate actively in an OSS initiative in the microfinance sector. This analysis of the case of Airdie’s adoption of Octopus MIS provides insights

The Open Source Attitude in Microfinance: The Case of Airdie

181

into the advantages and disadvantages of embracing an open source approach in microfinance.

Literature review on open source software Personal computers, plus related peripheral hardware and software, were created to help people perform various pre-defined tasks faster and with fewer errors. Subsequently local and wide area networks were built to enable people share information, ideas and knowledge, including how to achieve greater use of the emerging technology. These networks were extended and merged into the World Wide Web, which triggered the Internet revolution. Software makes computers perform programmed tasks. The software’s source code contains the instructions about how to deliver a certain output based on the inputs the user provides. The price of hardware, which is often manufactured using rare earths, can be high. Software, on the other hand, is intangible. Continuing to impose strong copyright restrictions on software after investment costs have been recovered and high profits reached, raises ethical questions. However, other questions emerge if society fails to protect intellectual property. Richard Stallman’s initiative of 1983 was directed to limit monopolistic risks in the emerging software industry. He created the Free5 Software Foundation and launched the GNU6 project with the objective of creating a free operating system (OS)7, developing other free software, sharing knowledge and raising awareness about the risks of monopolistic domination of the software market. He pleaded that users should have the rights, which are otherwise restricted through copyright and encryption, to modify the code of the programs to fit their needs and share this code (Stallman, 1999). The GNU project used copyright law to prevent free software from becoming proprietary. This original idea, termed “copyleft”, gives the right to run, copy, modify and distribute code modifications as long as one does not impose copyright restrictions on downstream developments. A developer is obliged to provide the supplier of the initial code a copy of his/her modifications. If the developer wants to restrict access to the enhanced code by encrypting it (and selling the enhancement), the initial suppler can freely distribute the new code, which was left by the developer, removing any competitive advantage of the encrypted software. Copyleft principles were formalized into the GNU General Public Licensing agreement (GPL) (Stallman, 1999).

182

Chapter Nine

GNU GPL discouraged software developers, who were interested in keeping some form of copyright, from participating in the open source movement. At the same time, the achievements of the Free Software Foundation were considered to be modest compared to the soaring potential created by mass discovery the Internet (Raymond, 1999). In 1997 a new branch broke off from the movement’s core. It coined the term ‘Open Source’ and defined it differently, although in the spirit of GPL, to allow flexibility when licensing source code containing a hybrid of free and proprietary components (DiBona et al., 1999). Open Source innovation requires the existence of user-based communities so that all can gain from the innovations of a few. For this, there needs to be a sufficiently large user base, plus the ability and willingness of some users to innovate and share their innovations. Open Source innovation is especially important if packaged software vendors are unable or unwilling to satisfy all users’ needs. According to Von Hippel (2001), three conditions must be met for an Open Source project to survive. First, at least some open source users must have sufficient incentive to innovate. Second, at least some users must have both an incentive to voluntarily reveal their innovations and the means to do so. And last, user-led diffusion of innovations must be able to compete with commercial production and distribution. Sharing of knowledge can take place if the costs are lower than the expected rewards of sharing. Rewards are high if one active user feels that others will also share, or if there are reputational/psychological gains. In this chapter we define the MIS based on OSS (or OSS MIS) to be any computerized management information system that doesn’t restrict the right of the user to modify the source code and to share the resulting modification with the community. There are currently two versions of the Octopus MIS: Octopus Professional and Octopus Community. The Community version is under the GNU Lesser General Public License—a concession to the GNU GPL that allows linkage with proprietary source code, respecting its copyright. Octopus Community is inferior to Octopus Professional in that it provides only single-branch architecture, contains fewer predefined reports and does not include several additional functionalities such as support of handheld devices and integration of accounting functionalities. Access to proprietary features is granted based on a paid subscription. Mifos chose the more permissive Apache Version 2.0 license for its MIS. Recognized by the Free Software Foundation as a free software license, Apache V2.0 neither imposes on the developers the obligation to

The Open Source Attitude in Microfinance: The Case of Airdie

183

share with the community their modifications, nor the requirement to maintain the same copyright for subsequent developments.

1. Advantages The direct advantages for the end user of OSS are: reduced or no cost for the software; possibility to adapt the software based on the users’ needs; constant updates; new features based on the needs of the users; and the support of the community that organizes itself around online discussion forums, blogs and Internet portals to promote and share the work of peers. The task of understanding the economics behind OSS is more complicated. What drives “egoist” individuals and companies to share the knowhow they control? Is it sustainable? Is there room for business? Assuming a critical mass of open source content is already available on the market and companies (and people) make use of it, then for-profit businesses can provide customized support to the users, who are not necessarily aware of all the developments and minor technical aspects. This business model works by analogy to the relationship between the law and the lawyers (Tiemann, 1999). In a democratic country the law is a large set of documents that are available for free to every citizen. In spite of the open source nature of the law, the services of lawyers are usually very expensive. Distributing source code in a potentially competitive market ensures a wider adoption by the community of this code. There is a possibility that the source code will become a widely used standard. If so, it is much easier to operate in a market that uses the standards you have designed. But why do people share? Initially, software was developed by researchers in universities (Perens, 1999). The role of academics is to share the knowledge they generate and their rewards come in the form of citations and references to their works by other researchers. Sharing was the rule before companies rushed in to exploit business opportunities. Fame may have motivated the student Linus Torvalds in sharing the code of Linux—the famous free and open source operating system that he designed. The current “optic fiber” era provides the opportunity for authors and inventors to enjoy fame immediately, compared to previous eras when celebrity often arrived posthumously. The quest for win-win situations can also explain the existence of the OSS movement as a cooperative model of software development and testing. If one has a specific software-related need and no optimal solution, one might share this problem with the community of users-developers.

184

Chapter Nine

People with similar needs might contribute to solving the problem by giving ideas, sharing their own incomplete solutions or providing a better approach. Some members of the community will report the bugs they have observed while using the software, thus testing any modification until the problem finds a robust solution. All active participants benefit from the process, as do free riders who use these innovations. Some companies share proprietary source code under their CSR activities to improve their corporate image. Microfinance is a costintensive and scale-dependent industry. Robust OSS developed for the needs of microfinance has strong potential in increasing the outreach and reducing operational costs per poor customer (Augsburg et al., 2011).

2. Difficulties The literature identifies five main barriers in adopting OSS: (1) the knowledge barrier, (2) legacy integration, (3) forking, (4) sunk costs and (5) technological immaturity (Nagy et al., 2010). We examine the perception and implications of each barrier in deciding the adoption of microfinancespecific MIS based on OSS. The knowledge barrier refers to the lack of awareness that OSS is available and potentially useful to satisfy the needs of the users. Besides the awareness required for innovational adoption and imitation by early adopters and laggards (Rogers, 1962; Bass, 1969), this obstacle also refers to the lack of skills required to implement, customize and correctly use the MIS. In the microfinance field the knowledge barrier is a difficult obstacle to overcome. Many MFIs do not have access to the Internet (CGAP, 2009), which restricts diffusion of OSS and the required knowledge to operate it. Moreover, only 56% of the CGAP-surveyed MFIs declared confidence in the ability of their IT department to service correctly the MFI’s hardware and software, and almost one in two MFIs was not satisfied by its level of information concerning best practices in the use of technology (CGAP, 2009). The legacy integration barrier is determined by the difficulty of connecting the OSS with existing hardware and software that the organization cannot easily replace. This obstacle is less important in microfinance since legacy systems most often are based on spreadsheets or basic database software with limited functionalities, and can be incorporated into the new system or efficiently substituted. Forking is a barrier that affects OSS. Since software is continuously improved and changed by different teams and individuals, not necessarily communicating with each other, the risk is high that versions will fork and simultaneous development will continue separately without any possibility

The Open Source Attitude in Microfinance: The Case of Airdie

185

of bringing the forked versions into a common trunk. Users would then have to choose to adopt one version or another. Sunk costs present psychological and financial barriers. Past costly investments in software force the users to continue to operate it for several years in order to absorb the investment cost. This is not a barrier specific to OSS: it is an obstacle commonly observed when subjects consider adopting new software. Technological immaturity is a major barrier to adopting OSS, especially in the case of MIS for microfinance. The not-for-profit and communal nature of OSS throws doubt upon the integrity of the product. Voluntary developers have neither the obligation to deliver accurate contributions, nor always a reputation to protect. Information asymmetry (users don’t know if contributors are trustworthy) is reduced as more people use the software and provide feedback or improvements. Technological immaturity concerns OSS that have not reached a critical mass beyond which adverse selection (in small communities, less competent developers might present themselves in an unrealistically positive light) and moral hazard (developers are not concerned by the quality of their work since their personal gain is not involved) can be offset by the contributing power of the crowd. The most difficult stage of every OSS initiative is the beginning. How to be sure that development will continue and the community will care about the project in the future? OSS initiators and early adopters have a huge responsibility in creating the nucleus around which regular and development-oriented users will gravitate.

Research methodology 1. Research question Emerging MFIs can manage their portfolio and related operations using manual (paper-based) systems and spreadsheets. Some big MFIs can mobilize large budgets to invest in core banking solutions that meet their MIS needs. Affordable MIS is desperately required for those small and medium MFIs that acknowledge their information systems prevent them from achieving their goals and identify insufficient funding as the main obstacle to improving the MIS. Existing literature, although it recognizes the importance of OSS, provides little insight into the particular aspects of the initial adoption of OSS in microfinance institutions. A notable exception is the recent case study about the implementation of Mifos at an Indian MFI (Das, 2011). Our case study focuses on specific aspects of adopting the Octopus open source

186

Chapter Nine

MIS in MFIs operating in developed countries. Such MFIs have the advantage of access to superior IT infrastructure, proximity to microfinance funders and other potentially important actors on the OSS scene. The disadvantage is the relatively small available sample of MFIs and their divergent product and service ranges. More specifically, our questions would be whether Airdie has innovated? What drove it to choose an MIS based on open source software? Did it voluntarily reveal its innovation to others through sharing of its experience on its website? Can the Octopus diffusion of innovations compete with diffusion costs of commercial production and distribution?

2. Case study methodology A case study approach is useful in specific circumstances. First, it could be for studying processes within a single setting, either through multiple cases or a single case. This could be through quantitative data or qualitative data or both. The final result may be descriptive, hypothesis validation or theory building (Eisenhardt, 1989). We are building exploratory, descriptive research by studying a new sector—open source software for microfinance. Specifically, we study Airdie—the earlyadopter of an OSS MIS developed by Octopus for the microfinance sector.

AIRDIE: Case Study in Open Innovations 1. Description of Airdie Airdie8 is a not-for-profit association created in 1994 that operates in Languedoc-Roussillon—one of the 27 regions of France. Airdie’s mission is to contribute to financing the social and solidarity-based economy of the region. Assistance to poor financially-excluded people has strong utility for the society and receives increasing mass media attention. Airdie targets financially-excluded people as well as larger entrepreneurial projects that involve socially- and financially-excluded people. In France, a formal full-time job, even if paid at the minimum wage, provides earnings above the national poverty line. Employment of poor people is systematically associated with their graduation to the non-poor status, which comes with increased purchasing power and better financial inclusion. Airdie finances the investment projects of unemployed persons to help them get a formal full-time job. Self-employment is another solution to social and financial exclusion. Airdie helps financially-excluded entrepreneurs fund their projects for

The Open Source Attitude in Microfinance: The Case of Airdie

187

income-generating activities. Creation of one’s own job is riskier than employment, but its long-term advantages over unemployment (and often unsatisfactory employment) are important for the person, his/her household, the local community and the entire society. Central and local administrations acknowledge the social utility of such projects that, besides lifting people out of poverty, reduce the pressure on social security. The regional Prefecture and General Councils of the five departments (Aude, Gard, Hérault, Lozère and PyrénéesOrientales) of the Languedoc-Roussillon region are all members of Airdie, together with other financial institutions and associations interested in promoting financial inclusion. Airdie offers around 10 financial instruments divided into two broad categories: loans and guarantees. One of the leading products is the credit for agricultural and rural initiatives—a five year interest-free loan with two years of grace period. The loan ranges between 500 and 16,000 euros. The borrower doesn’t provide collateral if he finds two co-signers. Another product targets financially-excluded individuals that cannot obtain bank loans due to bad credit records. If a person doesn’t repay a loan, the claiming financial institution files a report to Banque de France that manages a blacklist: the National Register of Household Credit Repayment Incidents. People from the list cannot borrow from financial institutions and their overdraft facilities are canceled. Issuing uncovered bank cheques would result in a similar procedure involving another blacklist that imposes restrictions on the use of financial instruments by the delinquent issuer. Airdie gives such people a second chance by providing them interest-free loans of 5,000 euros for 3 years. Two co-signers are necessary. Airdie manages the national program Nacre9 for the region of Languedoc-Roussillon. Under this program, young people and job-seekers can borrow from Government sources up to 10,000 euros at 0% interest rate for 5 years for their entrepreneurial projects. Nacre obligatorily requires an NGO (such as Airdie) to help the applicant get a loan of equal or larger amount from a financial institution and to perform continuous follow-up of the financed project. On behalf of France Active—a large association in France that supports solidarity-based enterprises—Airdie manages the financing of solidarity-based enterprises in LanguedocRoussillon. These enterprises can borrow large amounts (50,000+ euros) for long periods of time. France Active also participates in financing excluded individuals and young enterprises by guaranteeing up to 65% of their bank loans. Airdie provides the logistics and the infrastructure for

188

Chapter Nine

such guarantees to be established by charging 2% of the cost of the guarantee. Airdie monitors the projects as long as the guarantee is active. Monitoring is very important in the French microfinance landscape since projects which are monitored and accompanied have a higher success rate than others (Bumacov et al., 2012). NGOs such as Airdie constantly report the survival rate of financed projects to stakeholders, who consider this indicator as an important measure of the performance of the MFI. Such ratios create mission drift10 tensions, as low risk projects which have higher chances of success are more likely to be selected for financing. On the other hand, the “obligation” to have a high survival rate imposes better monitoring, which is generally beneficial for the borrower. As a third example, Banque Populaire du Sud subcontracts Airdie to select craftsmen and rural projects that need large loans. Airdie helps in preparing the loan application and supplements it with the guarantee of France Active. Airdie’s complex financing mechanisms necessitate a strong and flexible MIS. Collaboration with numerous stakeholders requires constant reporting of specific indicators and sharing of information, which the MIS has to facilitate. In 2011 Airdie participated in the financing of 500 projects, mobilizing 8 million euro, with an employment generation impact of about 1,200 jobs (Airdie, 2011). By the end of October 2012, Airdie was employing 26 people including 14 loan officers and 6 assistants. Airdie had no dedicated IT specialist. Therefore, we are analyzing a case that defies conventional wisdom. We would expect that NGOs with strong IT skills would experiment with OSS. Yet Airdie, without an IT specialist, decided to adopt an OSS MIS.

2. The information needs of Airdie Until 2007, Airdie was managing, as a regional representative, the microfinance activity of ADIE11—the largest French MFI. Airdie was using ADIE’s MIS for credit-related back-office operations. Separation from ADIE, decided by Airdie in 2007, obliged the MFI to acquire its own MIS. After the split, ADIE granted Airdie access to its MIS for an additional six months for entering new clients and new loans, and unrestricted access to Airdie’s existing loan accounts until their extinction. Airdie required from the new MIS the correct integration and convenience of use of its ‘multi-stakeholder’ microfinance activities. The MFI identified four objectives that the new MIS had to meet and four modules that the new MIS needed to include.

The Open Source Attitude in Microfinance: The Case of Airdie

189

The first objective required the creation of a single database to manage data on all the customers and all of Airdie’s financial instruments (loans and guarantees). The second objective involved automation of all predefined procedures such as generation of documents related to contracts with the customers (automatic generation of the contract to be signed, creation and dispatch of a reminder message if the client misses a payment, etc.) and registration of corresponding accounting transactions related to the movements of funds on customers’ accounts. The third objective focused on limiting operational errors (through restricting new contract engagements to available balances and exposure limits, and reconciliation with accounting and treasury). The last objective concerned reporting and analytical functions. The multi-stakeholder environment in which Airdie operates imposes rigorous and specific reporting. The burden on the MFI is very high as provision of services on behalf of a stakeholder, or using external financing / donations, requires specific reporting, including impact assessment. Provision of subsidized capital is justified if a certain social return on investment is observed. Each stakeholder measures impact using a different method and the MFI has to comply with the reporting requirements of them all. The four identified modules, with details presented in Table 1, were: (1) stakeholder management, (2) production (portfolio management: loans and guarantees), (3) accountancy and (4) budgeting. Initially, stakeholder management was performed on paper, production was supported by third party MIS and spreadsheets, accountancy was managed using proprietary software that Airdie planned to keep, and budgeting was managed using spreadsheets. As stated, Airdie differs from most MFIs in developing countries. It has a large number of stakeholders that require different analytical and statistical reporting. Airdie manages cash and non-cash products. The MFI is required to provide formal monitoring of financed clients during the contractual period. Clients can be individuals, companies (single or multiowner entities) and associations. Different categories would require different contractual requirements as opposed to the practice observed in developing countries where most MFIs disburse personal loans regardless of the status of the applicant (company or person). Since almost every individual and all businesses in France have bank accounts, repayment of loans is generally performed using automatic debit—a procedure executed by the bank managing the account of Airdie. Upon demand, Airdie’s bank debits the bank accounts of Airdie’s borrowers. An integrated communication between the MIS of Airdie and

190

Chapter Nine

the MIS of the bank would enable Airdie to have a computerized management of repayments and faster identification of delinquencies. Stakeholder management Clients: - Individual projects; - Enterprise projects; - Cooperatives.

Production

Request for financing and analysis: - Expertise; - Preparation of the financing plan; Governance: - Submission to - Representatives of local decision committee. and regional authorities; Contracts: - France Active; - Caisse des Dépôts 1. Cash product repayment: (bank); - Representatives of a. Automatic debit; solidarity-based economy b. Cheque; networks; c. Wire transfer; - Representatives of civil d. Cash (exceptional). society and other partners. 2. Non-cash product Employees & Volunteers: (guarantees and sub- 1 HQ (also operational contracted loans). branch); - 4 separate branches. Monitoring.

Accountancy (proprietary) Generation of accounting transactions corresponding to the operations made by the production module;

Budgeting Monitoring; Internal reporting; External reporting.

Exportation of the transactions to the proprietary accountancy software.

Table 1: Required modules for Airdie’s future MIS and their main tasks

3. Selection of the MIS Selection of the MIS for Airdie’s needs was constrained by several factors: limited time, requirements specific to Airdie, and budget. The desired complexity of the loan management module (due to the high number of products: loans and guarantees) and necessity to have a connection with its bank, persuaded Airdie to consider core banking solutions. Unfortunately high license prices and high fees for programming and parameterization coupled with the obligation to employ several IT specialists to ensure correct functioning of the powerful servers and complex databases needed for these solutions ruled out this option. Low and middle-end (off-the-shelf) packaged microfinance MIS had the advantage of lower price, but disadvantages related to limited functionalities (oriented to the needs of developing countries), rigidity in

The Open Source Attitude in Microfinance: The Case of Airdie

191

parameterization and location of vendors’ business and support centers— most microfinance MIS developers and vendors are based in developing countries where their clients are located. Octopus—the French open source MIS for microfinance—immediately captured the attention of Airdie. The option of developing a customized MIS by an IT consultancy was not considered as Octopus already released the free ‘Community’ version of the MIS that had only to be modified and parameterized to Airdie’s needs. One decisive argument in favor of Octopus, besides proximity and language, was the open source character of the MIS. Airdie, as provider of alternative financial services in the region, from the beginning adopted a pro open source attitude, which was constantly promoted to the stakeholders and clients. In the effort to cut operational costs, Airdie switched to OSS for operating systems, text editors, e-mail server engine and spreadsheets (where stakeholders did not impose a proprietary format). It could be argued that SaaS12 was a viable option for meeting the MIS needs of Airdie that should have been included. However, at that time, the offer of SaaS for microfinance was just emerging and thus not considered.

4. Avoiding OSS barriers In 2009 and 2010, CGAP performed in-depth surveys of a set of 25 packaged MIS software products for microfinance. Based on these surveys, in Figure 1 we present the number of features that we consider important in managing microfinance operations available in each product. An example of such features is the ability of the MIS to handle operations in different currencies. Another example is the ability of the MIS to identify client duplicates in the system. The results were based on developers’ self-declared information. The two MIS based on OSS (Octopus and Mifos) had less functionalities (54 and 55 out of 80) than other proprietary MIS. It is clear that the technology barrier is still shaping the OSS market in microfinance. On the other hand, these two MIS were the only ones that didn’t charge a license fee which, in the case of the other surveyed systems, could range from several thousands to several hundreds of thousands of euros. Airdie considers that the tradeoff was justified. Instead of buying what they didn’t need and spending money on difficult customization or development, they preferred to invest in the development of required functions and share the results with the community.

192

Chapter Nine

Similar microfinance institutions in developed countries are certainly confronted with the same dilemma. Assuming the risks and benefits, Airdie wanted to pioneer the use of MIS based on OSS for this particular microfinance segment. MIS25 MIS24 MIS23 MIS22 MIS21 MIS20 MIS19 MIS18 MIS17 MIS16 MIS15 MIS14 MIS13 MIS12 MIS11 MIS10 MIS09 MIS08 MIS07 MIS06 MIS05 MIS04 MIS03 MIS02 MIS01

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

Figure1: Identified functionalities from the CGAP surveyed MIS.

Being initially developed for the needs of the OXUS Development Network—a group of MFIs each located in Tajikistan, Kyrgyzstan and Afghanistan created with the support of the French NGO ACTED— Octopus was tailored for the microfinance model of these MFIs. Airdie wanted to contribute by adding the functions required by the microfinance model used in developed countries like France, Belgium and Canada. The fact that Octopus was related to OXUS MFIs reduced the risk that the MIS would be abandoned without gaining critical mass. Currently 138 MFIs use the Octopus MIS (Octopus, 2012) and the Community version registers over 50 free downloads per week (Sourceforge, 2012). However, the use of the MIS in developed countries remains limited. In the case of Airdie, the knowledge barrier played a catalytic rather than restrictive role. Aware of the ignorance within the European MFI community that OSS-based MIS was available and potentially useful for

The Open Source Attitude in Microfinance: The Case of Airdie

193

their needs, Airdie decided to tackle this barrier in order to open the way for others and create, with the existing users and potential newcomers, a community that would solve the MIS problem by joint effort. In the long term this kind of solidarity can have a positive cost-cutting impact on the microfinance sector and ultimately on the financially-excluded people that have to bear high transaction and operational costs. The lack of technical and business knowledge needed to implement, customize and use the MIS based on OSS was overcome by the proximity with the creator of Octopus, which, in spite of being a for-profit business, had CSR commitments. The legacy integration barrier was not an important impediment for Airdie. The proprietary accountancy software was the only important legacy system that had to be connected to the MIS, but this requirement would have applied to all other MIS options. Forking was not considered a barrier when it was decided to implement Octopus. The creator of Octopus was the main developer and it had a clear strategy where the MIS should go. Other potential developers would gain from collaborating with the creator. Sunk costs certainly did not represent a barrier for Airdie. Effectively this was Airdie’s first MIS investment decision.

5. Implementation of the MIS at Airdie Due to major development needs, the main developer (OCTO Technology) decided to implement Octopus in Airdie in phases. Each phase was divided in sub-phases following an incremental development and implementation approach called ‘Agile’. The company Octopus Microfinance—a social business constituted as a joint venture between OCTO Technology, a French NGO ACTED and ABC Microfinance13— that was in charge of the free distribution of the MIS Octopus Community was not formally involved. The incremental approach practiced by OCTO Technology when implementing the MIS also played a significant role in deciding in favor of Octopus. As opposed to the classic development and implementation approach of custom-built MIS, the current version of Octopus could be implemented immediately and made available to the users in its original form. Meanwhile, the required functions were gradually developed and implemented in batches and simultaneously tested by the users. For the implementation of the project, Airdie obtained subsidized financing for the development of the multi-branch architecture (presented in the Figure 2), which was mandatory for Airdie’s business model.

Chapter Nine

194

Montpellier

Perpignan Carcassonne Mende

Nîmes

Figure 2: The network architecture of the MIS of Airdie (the text under each branch represents the city where the branch is located).

Currently Airdie is testing the results of the third major phase of development. Initial stages were delayed due to the difficulty the developer experienced maintaining an accelerated program of modifications while charging below-market fees and simultaneously integrating the divergent requirements coming from Oxus MFIs and Airdie. The open source character of Octopus attracted another company to join in the development. This is Naxos Information Technology—a small Belgian IT company diversifying its experience in banking and communications—which is developing and improving the Airdie MIS. Its business model consists of providing the improved MIS (Octopus Professional) to microfinance institutions on a SaaS basis.

6. Performance of Octopus Octopus provides an intuitive interface for users, who appreciate its simplicity. Incremental developments allow a continuous improvement of the processes, delegating the testing role to the users. Such active participation of the users usually increases the chances of faster adoption and better use of the MIS. This approach imitates the development of OSS, with some limitations: there is only one active developer.

The Open Source Attitude in Microfinance: The Case of Airdie

195

For reasons explained below, Airdie could not benefit from the work of other developers that could join the effort and thus contribute to further reducing the cost and improving the quality of Octopus developments. Airdie still remains satisfied with the MIS from both technical and economical perspectives, in spite of the fact that not all the objectives were achieved on schedule.

Recommendations: Lessons for adaptation of open source Open source software represents an alternative to proprietary software. Several free robust solutions for MIS are already available for microfinance. These have fewer functions compared to proprietary solutions, but could represent a stable starting point, especially if the MFI has specific needs that require hard programming (changing the code). Only core banking solutions are ultra-configurable, but these are very expensive and affordable only by very large MFIs. Airdie chose to adapt Octopus. Unfortunately Airdie’s contribution to the open source community was not as much as was originally planned: MIS developers retained the copyright for part of the enhanced source code to ensure the sustainability of the MIS social business. Proprietary improvements are nevertheless available to MFIs that have similar needs, at a marginal cost. Octopus Microfinance (France) offers Octopus Professional through direct installation, provided the MFI pays for implementation and maintenance, while Naxos (Belgium) provides Octopus Professional through the SaaS (rental) business model. In aggregate, from a microeconomic perspective, increased supply of MIS solutions for microfinance should have had a positive impact on software prices on the market. Considering that these MIS are based on free OSS, the effect of lowering the prices on the market should have been larger. Even though many MFIs started to use Octopus, none have proposed enhancements. The community of developers-contributors remained limited and initial adopters have not benefited (yet) from the power of the crowd. Due to the immaturity of the market, the developers had to fork the Octopus MIS into two versions (Community and Professional) to maintain a competitive advantage. OSS, like any free offering, develops fast if there are early adopters and subsequent imitators. While Airdie is a case of an early adopter, if its model is not diffused, there would be few imitators. Therefore, it would be in the interest of Octopus to capitalize and spread word of mouth on such adoptions.

196

Chapter Nine

Moreover, for the model to be widely diffused, a community of forprofit implementers promoting this model to MFIs would be needed. Thus, Octopus needs to encourage software developers and application providers who would gain from an MFI’s purchase of the software. Such developers would then promote the OSS offering. Finally, Octopus needs to understand that having a free (community) version with limited features may have its costs as well as benefits. A lot of MFIs may try the software and find it too limited for their needs and drop it. Perhaps the professional version should be made free in return for development feedback as well as word of mouth promotion to other MFIs. Octopus could also gain by encouraging a number of small local MFIs in each region of France. Airdie too could earn revenue from exchanging its expertise with such institutions.

Future research The market of MIS for microfinance institutions is relatively small compared to the market of other software that people and businesses around the world use on a daily basis. In the example of operating systems, each computer around the world has to have an OS, and even smartphones and similar gadgets are increasingly part of the same market. It is not surprising that the OS market was the first to search for open source alternatives. Some MFIs may have the advantage of having, within their workforce, IT specialists with strong programming skills. Further research should concentrate on how to leverage this potential around a common cause—in a similar way to that attempted by Airdie.

Conclusion Microfinance institutions are consumers (free riders) of open source software with rare cases of active participation, as in the case presented in this chapter. On the other hand, the microfinance sector is in need of more and better OSS solutions. Expensive proprietary software penalizes the micro-borrower either through high operational costs, when the MFI is able to buy it, or by limiting them access to software (and thus to scaling opportunity and consequent economies of scale) through dissuasive pricing policies. Donor-funded institutions were involved in transforming the two MIS for microfinance (Mifos and Octopus Community) into OSS. For Octopus, this effort was not sufficient to create a critical mass around the MIS and

The Open Source Attitude in Microfinance: The Case of Airdie

197

to incite the community to engage in self-development of additional modules and functionalities. The business model of Octopus achieved partial success. More than 130 MFIs use Octopus, but deliberate forking was necessary for the developers to maintain (probably short-term) opportunities for earning. Airdie achieved partial success when deciding to invest in the development of Octopus. The MFI was able to satisfy most of its MIS needs at a lower cost, but could not help the community benefiting directly from the new functionalities as the developers retained the copyright to the source code modifications in order to secure a competitive advantage over the market. Since the new functionalities are supplied at marginal cost, this should contribute to lowering the prices of other microfinance-specific MIS on the market. Lack of funding and awareness certainly prevents the MFIs in engaging actively in OSS initiatives. In such an environment, provision of subsidies and inexpensive technical assistance oriented at promoting collaboration amongst MFIs around viable OSS initiatives could result in a snowballing effect.

Bibliography Airdie, (2012). Activités 2011. Augsburg, B., Schmidt, J. P. and Krishnaswamy, K. (2011). Free & Open Source Software for Microfinance: Increasing Efficiency and Extending Benefits to the Poor. Ashta, A. (Ed.), Advanced technologies for microfinance: Solutions and challenges. Hershey, PA: IGI Global. Bass, F. (1969). A new product growth model for consumer durables. Management Science 15 (5), 215-227 Bumacov, V., Toutain, O. and Ashta, A., (2012). L’accompagnement du micro-emprunteur - spécificité du microcrédit français : son importance pour le bénéficiaire. http://hal.archives-ouvertes.fr/hal-00752493. Comité consultatif du secteur financier (CCSF), 2010, Les conditions d’accès aux services bancaires des ménages vivant sous le seuil de pauvreté, Banque de France—Publications, Paris. Consultative Group to Assist the Poor (CGAP), 2009, Microfinance technology survey, http://www.microfinancegateway.org/p/site/m/template.rc/1.1.2236/ accessed 21 May 2009.

198

Chapter Nine

Das, P. (2011). A case study of Mifos implementation at Asomi, in Ashta, A. (Ed.), Advanced technologies for microfinance: Solutions and challenges. Hershey, PA: IGI Global. DiBona, C., Ockman, S. and Stone, M, (1999). Open Sources: Voices from the Open Source Revolution, Sebastopol, CA: O’Reilly. Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of Management Review 14 (4), 532-550. Nagy, D., Yassin, A. M. and Bhattacherjee. A., (2010). Organizational adoption of open source software: barriers and remedies. Commun. ACM 53 (3), 148-151. Octopus, (2012). http://www.octopusnetwork.org/octopuscommunity/ourclients. accessed 30 October 2012 Perens. B., (1999). The Open Source Definition, in DiBona, C., Ockman, S. and Stone, M., (Eds.) Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O’Reilly. Rogers, E. M, (1962) Diffusion of innovation. New York: Free Press. Raymond, E., (1999). A Brief History of Hackerdom, in DiBona, C., Ockman, S. and Stone, M., (Eds.) Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O’Reilly. Sourceforge, (2012). http://sourceforge.net, accessed 30 October 2012. Stallman. R., (1999). The GNU Operating System and the Free Software Movement, in DiBona, C., Ockman, S. and Stone, M., (Eds.) Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O’Reilly. Tiemann, M., (1999). Future of Cygnus Solutions: An Entrepreneur’s Account, in DiBona, C., Ockman, S. and Stone, M., (Eds.) Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O’Reilly. Torvalds, L., (1999). The Linux Edge, in DiBona, C., Ockman, S. and Stone, M., (Eds.) Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O’Reilly. Von Hippel, E., (2001). Innovation by User Communities: Learning from Open-Source Software. MIT Sloan Management Review 42 (4), 82-86.

Notes 1

Proprietary software provided for free on a trial basis. Proprietary software provided for free, usually with legal restrictions on the nature of use. 3 In some countries the authorities require that credit institutions finance businesses based on their reported indicators to incite them declare a larger share of the 2

The Open Source Attitude in Microfinance: The Case of Airdie

199

turnover and profit. An MFI would finance a business based on the declared indicators and would complement the credit with a personal loan to the entrepreneur if the real business indicators justify a bigger volume of financing. 4 Consultative Group to Assist the Poor—an independent policy and research center dedicated to advancing financial access for the world's poor, is supported by over 30 development agencies and private foundations and is housed at the World Bank. 5 Free as in freedom, not necessarily gratis. 6 A hacker tradition of using humorous recursive acronyms: “GNU” comes from G, which is for “GNU” (recursive trick), N is for “is Not” and U is for “Unix”; GNU’s not Unix. 7 Basic software that manages the resources of the computer to allow the installation and use of application software. Microsoft Windows and Unix are famous series of such operating systems. 8 In French: “Association Interdépartementale et Régionale pour le Développement de l’Insertion par l’Economique” translates as Interdepartmental and Regional Association for the Development of the Integration through Economy. 9 French acronym meaning New Accompaniment Tool for Creation and Business Takeover. 10 Shift of the MFI from financing poor people to richer borrowers. 11 ADIE—Association pour le Droit à l’Initiative Economique. 12 SaaS—Software as a Service—supposes the remote use of software which is based on the servers of the provider. The user doesn’t pay the license and implementation fees, he pays for the effective use of the software. 13 A social business that runs the peer-to-peer micro lending web portal Babyloan.org.

CHAPTER TEN SAAS: STRATEGIC INNOVATION IN MIS FOR MICROFINANCE MARKET1 ARVIND ASHTA

Introduction Sustainable development has financial, environmental and social aspects. One of the key industries addressing both social and financial aspects of sustainable development is microfinance. Microfinance has been evolving at a very rapid pace and, for sustainable growth and scalability, a Microfinance Institution (MFI) needs an excellent information system. For this reason, new providers of information systems software have been entering the market. Today there are hundreds of such software packages, many of which are very expensive. This raises a series of questions: Is it possible to reduce the cost of information processing for this sector? Is it possible to use the latest cloud computing technology to provide a new business model of software provision so that smaller MFIs can use the information technology? Would it be in the interest of incumbents to offer a new business model or would new entrants do this? “Microfinance” is often defined as financial services for poor and lowincome clients offered by financial institutions focused on that market2. This sector started in the 1970s and is today serving over 200 million customers (Maes and Reed, 2012) through, very approximately, a few hundred thousand microfinance institutions (no one so far has counted the tiny informal MFIs—not even the World Bank). Table 1 below shows that the median return on equity is about 10% in good years and even in years of crisis the median MFI typically earns over 7% return on equity. However, only three quarters of the MFIs reporting to the Microfinance Information Exchange are breaking even, owing to their high operating

SaaS: Strategic Innovation In MIS For Microfinance Market

201

costs, even though median interest rates are about 28% per year, much higher than those paid by the richer customers of commercial banks. 2006

2007

2008

2009

2010

2011

Median ROE

10.5%

10.8%

9.4%

7.1%

7.2%

8.3%

% MFIs with positive ROE

75.1%

76.0%

74.6%

70.9%

75.1%

78.0%

Average interest rate

33.7%

32.5%

35.5%

32.8%

33.0%

32.1%

Median interest rate

29.3%

28.8%

30.5%

28.0%

28.1%

26.6%

Table 1: Profitability and interest rates in microfinance. Source: Based on MIX data downloaded on Feb 3, 2013 Figure 1 below takes a snapshot view of 1,600 MFIs reporting data to the Microfinance Information Exchange (MIX). About 50 MFIs each have an outreach of more than 300,000 customers. Another 100 MFIs have an outreach of 100,000 to 300,000 customers. These 150 MFIs are usually termed Tier 1 and are attractive customers for software manufacturers. These are followed by another 250 MFIs whose outreach is to more than 20,000 clients. These could be termed as mid-size or Tier 2 enterprises. However, the vast majority of MFIs have less than 20,000 clients. In reality, this long tail may be even longer than the data suggests as the many thousands of tiny MFIs would be unlikely to volunteer data to the MIX. A median MFI reporting to the MIX has 8,000 clients with an average loan size of about $6003. Operating costs are about 14% of assets (Hudon and Ashta, 2013). The many resultant transactions of small amounts require automated information processing for transaction costs to be reduced. Some of the top 50 MFIs use in-house developed software, while a range of packaged software vendors target the rest of the Tier 1 and Tier 2 institutions. While the large MFIs have adequate Management Information Systems (MIS), the smaller institutions may find it expensive to build their own MIS software or buy off-the-shelf MIS packages, despite the presence of over a hundred vendors (see Chapter Eight by Ashta, Bumacov et al. in this book). Besides those reporting to the MIX, the long tail of smaller MFIs may include many MFIs who are so small that they have no time to report data and may not even be able to collect it.

Chapter Ten

202

How to provide MIS software for the long tail of MFIs? a. Especially for the long tail of small MFIs b. But also for larger MFIs who many need to serve as examples outreach 300,000

Large Customers : Tier 1 MFIs

Your Typical Customers : Tier 2 MFIs 100,000

How to make these customers viable? Tier 3 MFIs 20,000

50

150

rank of MFI based on MiX data 400

Figure 1: MIS for the long tail of MFIs. Source: Based on MIX data downloaded on June 12, 2012

The question is what innovation is required to provide MIS packages to the long tail of small MFIs who are yet to break even? Based on this case study in the microfinance industry, we test existing assumptions of innovation theory for this sector and hope that larger empirical verification will come from other researchers testing these in other sectors.

Literature Review / Theoretical Positioning in Strategic Innovation Innovation literature has witnessed the birth and evolution of a large number of concepts and theories. Traditionally, innovation has been classified into incremental innovation and radical innovation (Abernathy 1978). Although Sundbo (1997) seems to suggest that radical innovation is just larger in terms of output (turnover and profits) compared to incremental innovation, other authors observe that in radical innovation, there is a technological change (Christensen, Suárez and Utterback, 1998). Henderson and Clark (1990) introduced the notion of architectural

SaaS: Strategic Innovation In MIS For Microfinance Market

203

innovation and modular innovation, within the incremental innovation categories, indicating the difference between changing components (modular) or changing the links between components (architectural). According to them, radical innovation destroyed both modular and architectural knowledge, creating very different challenges for established firms. This challenges include, notably, difficulties associated with a nonstandard design which then has to be accepted by users to fit into other products. Within the radical innovation category, a distinction is maintained between disruptive innovation which disrupts the field of the incumbent players (Christensen, Bohmer and Kenagy, 2000), as opposed to sustaining innovations, which permit incumbents to keep their dominant position. Figure 2 below captures the elements of this review. Disruptive innovation usually requires changing the price-product equation so that new markets are created (Christensen, Johnson and Rigby, 2002) and the business model itself is changed (Christensen, 2006). Specific to information technologies, radical innovation is leading to business model innovation such as crowdsourcing, democratized innovation, eco-systems and recombinant offerings (Shuen, 2008). All these different types of innovations could therefore be disruptive or sustaining, depending on whether they are undertaken by newcomers or incumbents. Recent research suggests that there is disruptive innovation in the making through the introduction of Software as a Service (SaaS) in the microfinance market (Ashta and Patel, 2013). This chapter develops that research by looking at this innovation from a strategic innovation perspective. H10: SaaS innovation in the microfinance sector is provided by new operators H1A: SaaS innovation in the microfinance sector is provided by established incumbents Disruptive innovation requires institutional change (Maguire and Hardy, 2009; van den Hoed and Philip, 2004), both within the firm and outside the firm by taking complementary partners or making changes in the existing institutional partners and customers values (Asselineau, 2010; Mazzarol and Reboud, 2008). Without these changes a strategic innovation of the disruptive kind cannot succeed. Some research has been done on institutional change, a field which broadens new institutional economics by giving the actors a change agent role in mobilizing embedded actors and disruption of norms and beliefs or de-institutionalism

204

Chapter Ten

(Lawrence, Suddaby and Leca, 2009; Maguire and Hardy, 2009; van den Hoed and Philip, 2004). In this chapter, we see how existing and new entrepreneurs usher in radical innovation in the MIS market in order to better serve the long tail of Microfinance Institutions (MFIs). We will try to see if the innovation requires product technology to change.

Figure 2: Positioning Strategic innovation in a theoretical map.

H20. SaaS is aimed at the smaller MFIs who are not being served (no institutional change required) H2A: SaaS is aimed at large MFIs (a change in practices required) To attract larger and more profitable customers, the SaaS should give them some basic advantages such as overcoming inflexibility of existing MIS. H30: SaaS requires more standardization in financial products H3A: SaaS permits more flexibility in financial products Closely connected with the way institutional entrepreneurs change the rules of the game is the parallel development of the concept of strategic

SaaS: Strategic Innovation In MIS For Microfinance Market

205

innovation (Markides, 1997; Govindarajan and Trimble, 2005; Mathews, 2006; Charitou and Markides, 2003), which involves changes in ideas, processes, technologies from existing business models. In Figure 2, we have indicated that strategic innovation is more likely to come from radical innovation, although it also could come from incremental innovation. The idea is that a smaller company can catch its bigger competitors off guard. The trick is not to play the game better than the competition, but compete by developing and playing an altogether different game (Markides, 1997). According to Govindarajan & Trimble (2005), strategic innovation requires making strategic experiments before a clear market leader has emerged in poorly defined but high growth industries. There are many related concepts including strategy innovation, value innovation, non-liner innovation, discontinuous innovation, competitive innovation and new-style product development (Schlegelmilch, Diamantopoulos and Kreuz, 2003). A common definition is provided by Schlegelmilch, Diamantopoulos & Kreuz (2003, p. 118): “Strategic innovation is the fundamental reconceptualization of the business model and the reshaping of existing markets (by breaking the rules and changing the nature of competition) to achieve dramatic value improvements for customers and high growth for companies.”

Mathews (Mathews, 2006) explains that latecomers and challengers at the periphery can upset incumbent market leaders by capitalizing on the scale offered by the global economy market. This is achieved through a combination of technological innovation, strategic innovation and organizational innovation, involving hybrid structures that permit the linkage and levering of resources rather than ownership. He provides many examples of successful challenger/latecomer innovation: ISPAT (now Mittal) using DRI technology in the steel industry; CEMEX being the first to use GPS technology to deliver cement; Acer using acquisitions and local partnerships to expand by creating and managing a global cluster of semi-autonomous businesses; Li & Fung using the low cost advantages of producing in China and acquiring international players; and Lenovo purchasing the computer business of IBM, which IBM considered to be peripheral. H4O: SaaS requires different business processes compared to existing software suppliers H4A: SaaS uses the same business processes as used by existing software suppliers

206

Chapter Ten

According to Markides (1997), the different game does not require radical technological change. It requires asking basic questions such as who the customer is, what the product/business is and how to supply it. These same three questions are defined by Govindarajan and Gupta (2001) as redesigning the value chain architecture, reinventing the concept of customer value and redefining customer base. How a company creates value and delivers it to its customer is defined as its business model (Osterwalder, Pigneur & Tucci, 2005). This business model is a link between strategy, organization and systems and includes elements as pricing mechanisms, customer relationships, partnering, and revenue sharing. The basic pillars of the business model are the product, the customer interface, the infrastructure management and its financial aspects (Osterwalder, Pigneur & Tucci, 2005). The business model is not the same thing as a business process. H50: SaaS targets different customers, provides a different business model and provides a different distribution network H5A: SaaS targets the same customers, with the same business model and same distribution network A strategic innovator looks at industry positioning and searches for gaps—in customer segments, customer needs and delivery channels— which are expected to grow. S/he analyses the core competencies (strategic assets) of the enterprise and develops them or recombines them to satisfy targeted niches (Markides, 1997). Not all strategic innovation is disruptive: Disruptive strategic innovation requires changing the rules of the game and being in conflict with the existing ways of doing business (Charitou and Markides, 2003). Disruptive strategies emphasize new value propositions that focus on a low-price, low-margin niche that grows and becomes a large part of the market. To do this the strategy needs to be good enough (i.e. satisfactory) in the old attributes and superior in the new ones. Based on their case studies, Govindarajan & Gupta (2001) found that strategic innovation could involve eliminating middlemen and improve the cost structure, asset base and responsiveness; selling total solutions instead of parts, and gaining customers if the total value is superior in quality and/or lower in cost than buying individual parts; and finding hidden customer segments that grow and become the dominant model. H60: SaaS is providing a low price low margin product

SaaS: Strategic Innovation In MIS For Microfinance Market

207

H60: SaaS is not changing the overall economics However, strategic innovation usually comes from new players as opposed to entrenched incumbents, who tend to prefer to hold on to existing positions (Markides, 1998), either from inertia, fear or costs of change, leading to technological lock-in. It is difficult for a dominant player to upset its existing organizational partners by responding to a challenger (Charitou and Markides, 2003). In general, they choose to ignore the new challenge or form a separate unit to deal with the challenger. H70: Incumbents permit SaaS operators in business areas they consider peripheral H7A: Incumbents compete to block SaaS operators If the disruptive innovation is not technologically based but strategy based, the incumbent could respond by improving its existing products, ignoring the competitor, by entering into the logic of the challenger through a separate company, embracing the logic of the challenger, or by further disrupting the market of the challenger (Charitou and Markides, 2003). H80: The incumbent software providers will not introduce SaaS H8A: The incumbent providers will improve their products by providing a SaaS option Incumbents can themselves create strategic innovation by monitoring their strategic health, creating positive crises, challenging the strategic planning process, institutionalizing a questioning attitude and experimenting (Markides, 1998). A culture of experimenting with different strategies and learning rapidly thorough theory-based planning could help incumbents attain strategic advantage (Govindarajan and Trimble, 2004). This theory-based planning involves using a few critical variables and leading indicators, creating cause and effect maps, focusing on broad trends over a large period of time and frequent reviews. Strategic experiments allow combining the two models (traditional and new) in the same organization but keeping them largely separate by creating links in R&D and marketing. The idea behind responding within the same organization is to use existing resources and combine them with the new ones to produce a better response (Govindarajan and Trimble, 2005). For strategic experiments, Govindarajan & Trimble recommend that an

208

Chapter Ten

outsider heads the new unit but reports to the same person as the head of the traditional unit; the new unit must have its own organizational structure, processes, and performance measures; its performance should be measured in speed of learning based on initial assumptions and not on outputs; its culture should be of learning and experimentation rather than efficiency; incentive mechanisms between the two units must ensure that there is low conflict. Being in the same organization confers advantages: Sharing assets and resources (including intangibles like brands and knowledge) provides advantages over competitors. Schlegelmilch, Diamantopoulos & Kreuz (2003) also stress the role of culture, processes, people and resources in creating customer value and strategic positioning. H90: Incumbents start a separate unit for their SaaS offering H9A: Incumbents keep the SaaS offering in the same organizational setup.

Research Methodology and Sample Description The concept of Strategic Innovation is fairly new. Our objective is to see, using a case study method, whether each of the null hypotheses mentioned above hold or the alternative hypothesis is validated. We do this specifically for one sector: MIS for microfinance. The long tail of small Microfinance operators is attracting niche MIS service suppliers. Can this long tail provide the scale necessary for a supplier to use a different business model to reach them and disrupt the existing software suppliers? Is the global economy sufficiently open to MIS providers for them to capitalize on this scaling factor and enter what would otherwise be considered a niche low-margin, low-return market? What lessons can we learn from successes as well as failures in this industry? Evidently, since the technology is new and adoption has been slow, the sample size is small. Moreover, the study of such items as business processes and business models is better done through in-depth case studies rather than quantitative surveys. Case studies are useful to examine a contemporary phenomenon in its real-life context and provide explanations linked to many variables (Eisenhardt, 1989; Yin, 1994).

SaaS: Strategic Innovation In MIS For Microfinance Market

209

1. The incumbents of the MIS software serving Microfinance The incumbent software providers in the microfinance space are many and fragmented. CGAP (Consultancy Group to Assist the Poor) has collected data on about a hundred MIS suppliers and transferred this information to the MIX. These data show that four providers clearly stand out: Loan Performer, MBWin, Emortelle and Finance Solutions (Gaul, 2011). At the time of the study each of these had over a hundred customers, exclusively microfinance (i.e., not banks). We take these as the incumbents for our study. Table 2 provides some information on them. Software name

Loan Performer

MBWin

Emortelle

Finance Solutions

Enterprise name

Crystal Clear Software Ltd

FAO-GIZ MicroBanker Project

Micro Software Designs

Sigma Data & Computers ltd

Year started

1998

2000

1986

1988

Country

Uganda

Thailand

Scale of customers

Up to 50,000 clients

Can go beyond 50,000 clients

For profit

Yes

No

Customer Sectors

Only Microfinance

Only Microfinance

N/A

Only Microfinance

Number of customers

220

138

132

102

Regional Focus

Africa, global

Only Latin America

Sub Saharan Africa, but also South Asia

but

Asia global

but

Trinidad and Tobago Can go beyond 50,000 clients Yes

Uganda Up to clients

50,000

Yes

Table 2: Profile of the incumbents. Source: Based on data downloaded from CGAP. * The project providing support for MBWin is part of FAO and has 350 customers on December 3, 2012 (interview).

210

Chapter Ten

2. Radical technology: Cloud computing and SaaS A new technology that has become available in recent years, and that could disrupt the market, is cloud computing. This very broad term can be used to include public clouds, community clouds and private clouds, as well as hybrids of these. The basic value proposition is that by initiating a pay-for-use model, cloud computing provides fast-growing firms with access to much larger computing capacity than would currently be needed, especially for sudden peak requirements (Armbrust et al., 2009). Other advantages include cost savings through the pay-per-use model; short time to start operations; centralized management: flexibility of use; rapid availability of resources/ services, e.g. for temporary, short-term use (peak loads); reduced need for hardware and processors; energy/resource savings; better protection against external attacks; use of the latest technology; location-independence, permitting user mobility; and the ability of the institution to focus on its core-business, rather than on supporting its IT infrastructure (Bittman, 2012; Gartner, 2011; PWC, 2011a, b). Cloud computing includes not only the data centers which could be called Infrastructure as a Service (IaaS), but also Platform as a Service (PaaS) and Software as a Service (SaaS). Within the cloud computing concept, SaaS is a multi-tenant architecture product that permits many customers to share the same platform. For example, Google or Amazon could provide the physical cloud on which SaaS operators would be based. According to Gartner research (Bittman 2012), the market for cloud computing was $12.3 billion in 2011 and expected to be about $22.1 billion by 2015. The biggest market is North America, followed by the EU. Within this market, software vendors are more aggressively pursuing SaaS buyers outside traditional markets by offering local-language availability, forming alliances and constructing data centers to accommodate local requirements. A number of recent market surveys (Gartner, 2011; PWC, 2011a, b; Totango, 2012) provide details of perceptions and concerns of this industry. The cost sharing advantages of this extra capacity is higher in public clouds than in private clouds, which require the client to purchase their own servers and data centers. Nevertheless, customers perceive higher risk in public clouds than in private clouds. Their concerns include security concerns; problems with system integration; lack of flexibility in a complicated market; compliance violations; market opacity; forced reliance on outside companies owing to transfer of data sovereignty; increased administrative, staff education and training costs; network instability; longer-than-expected deployments; political instability;

SaaS: Strategic Innovation In MIS For Microfinance Market

211

possible bankruptcy of SaaS operator; and difficulties of contract enforcement. As a result, 78% of respondents to a survey said they would be building private clouds by 2014 and 47% want to be hybrid by 2015: only small enterprises wanted to go directly to publicly provided SaaS (Gartner, 2011). Another survey confirms that 12% of enterprises are using clouds (18% of the large-sized and 9% of the middle market), and that large firms are choosing private clouds (PWC, 2011a). SaaS is different from Application Service Providers (ASPs) who provide off-the-shelf software that is hosted in the cloud. The latter saves some costs such as hardware maintenance but the software company still has to maintain different versions for different ASP offerings. Thus, in effect, they will have n different installations for n different customers. Achieving scale and handling change management becomes very challenging in this approach as the number of customers increases. These costs of complexity and customization trickle down to the customer.

3. The challengers of the MIS software serving Microfinance A number of new entrants have entered the MIS for microfinance market by providing software as a service (Ashta and Patel, 2013). However, after eliminating those that are ASP providers hosted in the clouds (Fino), and those that are merely servers (IBM) for other software, we can see from Table 3 that SaaS operators include Mambu, MicroPlanet Technology, MFIFlex and Mostfit. We note that not all SaaS operators make their own software: MicroPlanet, for example, uses InfraSoftTech Software. Similarly, Mifos software could be deployed by several different SaaS operators.

4. Excluding the really large players We do not cover the very large players like IBM and SAP because they are not present in the microfinance niche where the customers are too small for them. Although Oracle’s Flexcube and Temenos, providers of banking software, are present in microfinance, their microfinance presence was lower than the incumbents we selected. According to Snehal Fulzelz, founder of MFIFlex, Temenos is primarily serving banks that are downscaling into microfinance.

Chapter Ten

212

Software name

Mambu

Enterprise name

Mambu GmbH

Country

Germany

USA, India

USA, India

Scale

Small/ Medium

Small/ Medium

Medium / Large

For profit

Yes

Yes

No

Yes

Customer Sectors

MFIs but also any financial services

Only Microfinance

Only Microfinance

Only Microfinance

Regional Focus

Global

Global

Global

India, but Global

MFIFlex

MicroPlanet Technologies

Mostfit

Intellecap India

Table 3: Profile of the challengers. Source: Based on website, personal interviews and correspondence. None provided the number of customers. Signis Aliks, who heads MBWin, mentioned that Temenos and Flexcube are sometimes selected by bigger MFIs moving into bank-like operations, but he believes that such customers probably would not consider the SaaS model.

5. Data Collection The research was based on exploring the websites and interviews— where possible, with the founders / CEOs—of the challengers (Mambu, MFIFlex, MicroPlanet) and incumbents (MBWin). The discussion with MicroPlanet was in 2009 and led to a joint paper for which written communication was exchanged with the different SaaS operators that were then exploring the market (Ashta and Patel, 2013). This paper builds on that research, with fresh interviews with the founders of Mambu and MFIFlex. These interviews were held over Skype for about an hour each in June 2012 and annotated. The notes were sent back to the respondents who then modified them where required. Similarly, a 45-minute interview with the General Manager of MBWin took place in December 2012 and notes were sent for his approval. The questions asked at the first stage to

SaaS: Strategic Innovation In MIS For Microfinance Market

213

the challengers were on their experience and perception of the market. Once it was clear that there was a broad similarity in their perceptions, a review of the literature led to the hypothesis building and the interview with MBWin was on each of the hypotheses. Secondary data was also taken from the CGAP website.

Findings The findings are presented in the form of a meaningful story. We start with the incumbents since they have been present in the market for a longer time.

1. The incumbents Our interview with Signis Aliks confirms that Loan Performer, MBWin and Finance Solutions are leaders in this market, but he had not heard of Emortelle. The business model of packaged software is based on a complex pricing mechanism including licensing costs, implementation costs, maintenance costs and training, as illustrated in the first part of Table 4. These separate categories of pricing are necessary because different customers have different needs. However, as each vendor provides a totally different pricing structure, the customer is confused on the total cost he would have to bear in each case, and is unable to make comparisons before starting negotiations. To overcome this problem, CGAP asked the vendors to indicate, in three different scenarios (100,000 clients, 40,000 clients, 15,000 clients), what the total cost to the customer would be using some simplifying common assumptions. This is shown in the shaded part of Table 4. As we can see, software costs for an MFI could be anywhere between $16,000 and $244,000. What this means is that an MFI with 15,000 customers pays more than $1 per customer just for information processing software (without counting its own personnel cost). The smaller the MFI, the higher this cost. Thus an MFI with 4,000 customers would pay about $4 per customer just for information processing. For this long tail of small customers, especially in regions where average loan size is less than $400 per year, saving on software costs could be very important. Signis Aliks clarified that, theoretically, for a single user application, the license fee is only $1,500, but implementation and support costs could add considerably to the total costs.

Chapter Ten

214

Software name

Loan Performer

MBWin

License Fee

$900 to $3000

MBWin Core with one module: $1,500 for standalone $2,000 for LAN configuration (up to 5 users, with 200 USD per additional user)

Implementation

$4000

$400-$700 per day

Maintenance fee

20% of license

$2,500 for first year, 15% to 25% thereafter

12.5% of license fee

20% of license

Training Fee

$300 to $500 per person

$250 per day

$75 per hour

$300 per day

Cost for 100,000 clients, 220 users

42600

244000

not disclosed

83000

Cost for 40,000 clients, 60 users

20400

141000

not disclosed

29000

Cost for 15,000 clients, 45 users

16920

112500

not disclosed

16000

SaaS option

No

Yes

No

No

Emortelle

$39,000 for license and implementat ion

Finance Solutions

$0 to $15000

$300 per day

Table 4: Incumbents software costs. Source: Based on CGAP surveys for 2009/2010. * The low end estimate of $112,500 assumes 45 users and many branches. This is hardly ever the case for very small MFIs and is more applicable to an average sized MFI.

SaaS: Strategic Innovation In MIS For Microfinance Market

215

Distribution is usually through commercial partners. However, if there are no commercial partners in a country and a MFI approaches the software provider, direct sales are also possible. MBWin started an experiment with Software as a Service in 2011. This is based on a server in Bangkok using the same standard software package, but each customer has a separate database. This is used for very small customers: usually cooperatives with 50 to 500 members in Vietnam, Thailand and Cambodia, with each country requiring a different language. Currently, MBWin is handling SaaS for 20 such customers. MBWin doesn’t sell the license, but they charge $240 per group per year for support. Emortelle mentioned cloud computing and its desire to experiment with SaaS in a newsletter in 2011 but there seems to be little done after that4.

2. The Challengers SaaS is similar to renting a facility with access to huge capacity in case the user requires it. It is not a lease versus financing decision because the capacity to which the institution has access is huge. This rental option radically reduces the cost of access to such capacity. Mambu is the only one of the four challengers we studied that is transparent about its pricing policy and mentions it on its website. It varies between $25 per user per month to $85 per user per month. This works out to an annual cost between $300 and $780 per user. For 45 users, the lower end cost would at $13,500 be about 15% cheaper than the incumbents. In fact, the smaller the MFI and the fewer the number of users, the more advantageous is SaaS in terms of costs. Moreover, to have a common global pricing, one needs a homogenous base. In the microfinance sector, the problem is that MFIs with higher loan sizes would have fewer customers but conversely would have more users to process and assist each of these clients. For example, average loan sizes may be $150 in South Asia but $350 in Africa and $1,000 in Latin America and over $10,000 in Europe. The higher-value customer in a more advanced market may require additional features such as indexed interest rates or accrual-based interest accounting. MFIFlex’s website indicates that it offers a basic version and a premium version of its SaaS product. The pricing policy is not indicated on the website, although premium support costs of $5,000 per year are indicated. However, based on an interview with the founder, it seems prices vary from $0.55 to $1 per client of the MFI per year (Ashta, 2012). For 20,000 clients, this would

216

Chapter Ten

work out to about $11,000 to $20,000 per annum, and this compares closely to the lower end of incumbent packaged software pricing which, as we saw above, starts at $16,000. Thus, smaller MFIs would undoubtedly gain by the MFIFlex SaaS offering, but at 20,000 clients onwards, the MFI may be tempted to remain in packaged software mode, especially if premium support costs are factored in5. Mambu departed from a client-based pricing approach, owing to its inflexibility to the various markets and needs. Clients have different values to different MFIs. A client with a $10,000 loan is hundreds of times more valuable than one with a small deposit account yet similar services need to be provided to both. The user based pricing model captures the real value to the MFI created using a system like Mambu, namely increased efficiency of the staff and far lower costs and risk of errors and fraud. From this perspective, it is not the MFI’s customer numbers that are significant, but the services offered to the customers. Our exploration of the websites of these eight cases (four incumbents and four challengers), plus interviews with some of them, indicate that SaaS may be a radical innovation and may provide a sufficiently different pricing model to permit strategic innovation, either by new entrants or by smaller existing vendors. The incumbents have been slow to respond, indicating that the strategic innovation may lead to disruption rather than sustaining the existing incumbents.

Discussion 1. Disruption in the Market Place So far, SaaS seems to be offered largely by new entrants but incumbents are also experimenting, thus tempering our expectations of hypothesis 1 and hypothesis 7. Nevertheless, a look at the four websites indicates that MicroPlanet Technologies is talking more about its ability to use Temenos T24 software than its ability to provide Infrasoft Technologies’ OMNIenterprise software as a service. This means that one of the four challengers is already finding it difficult to get the necessary adoption rates. Our discussion with MBWin also indicates that it would be happy to remain a software developer and have SaaS run by commercial partners. The founders of Mambu and of MFIFlex both think the major competition is not among SaaS providers, but from the incumbent packaged, on-premises software. One problem which confronts them additionally is the confusion in the market caused by ASP hosted

SaaaS: Strategic Innnovation In MIS M For Microfiinance Market

217

solutions. T These problem ms come not from the leeading incumb bents we studied, but from the myrriad of smallerr MIS operatoors who wish to deploy their softwaare in the clouud without maaking it multii-tenant. Therefore, we could say, foor hypothesis 8, that the leaading incumbeents would no ot respond immediatelyy, but smallerr existing sofftware providders may mod dify their technology, either partiallly or fully to respond r to thee threat of new w entrants in an alreadyy very fragmeented market. On the demand sidde, the mark ket potential is probably 100,000 microfinancce institutionss plus many y institutionss on the frringes of microfinancce. As we saw w from our stu udy of MIX ddata presented in figure 1, one convenient way off segmenting the market w would be largee, big and medium deppending on the t number of o the clientss of the MFII. This is captured in tthe rows of Fiigure 3.

Figure 3: Seggmenting the MIS for Microfin nance market

A seconnd dimensionn of customeer segmentattion is based d on the innovative ccharacteristicss of the manaagement of thhe MFI. Manaagers that have grownn up with Innternet and arre comfortab le online aree seeking innovative ssolutions hostted in the clo oud to see hoow these coulld change their businesss. The old wave of manag gers, or managgers in more trraditional

218

Chapter Ten

organizations, are still in the ownership mode. Figure 3 captures this discussion with MFI Flex and Mambu. In our proposed market segmentation we have positioned, approximately, those using each different kind of information system. Typically, the large players have their own customized systems. Some of them may be ready to share these with other smaller MFIs to recover some of their costs. Most of the medium MFIs, and some large ones would use packaged software since they would not have the financial autonomy to afford their own customized software. Some of the owners of the packaged software, especially those that are struggling to attain market share, may try other cloud-based options. To this end, they may host their existing solutions on the cloud. However, true, multi-tenant SaaS solutions are aimed at smaller MFIs where there is a distinct cost advantage to the customer, as we have seen above, thereby confirming hypothesis 2. Very small MFIs, however, continue with their existing Excel and/or manual systems because even the SaaS offerings may be too expensive for them. Thus, as we saw in hypothesis 7, the SaaS niche may be peripheral to the operations of the incumbents, limited to technologically competent owners or managers of small MFIs. Also, in line with hypothesis 5, the SaaS operators therefore need both to market themselves to such computerliterate businessmen and to educate the owners of MFIs to shift them towards an openness to SaaS. The new SaaS players are therefore looking at the lower end of the unserved market, hoping that as the trend spreads, mid-sized and even larger early innovators would come to them. The advantages of approaching the green-field players are: firstly, less legacy friction costs in getting people to change since they are not tied in to any existing software; secondly; less costs of data migration; and, thirdly, a presumed openness to novel solutions. Overall, for hypothesis 6, we can say that SaaS constitutes a lower price solution only for very small MFIs.

2. Is there a new business model? The cloud is seen to be making it possible for value to be added to smaller firms who can tap into capacity at any time. How can a cloud based service model be marketed? One instantly hypothesizes that the marketing would be online and viral. This is the model being used by MFIFlex (Ashta, 2012). However, our discussion with Mambu indicated that although most of the existing marketing is done directly online, nevertheless, a partner distribution channel is being developed. Although sales are currently 70%

SaaS: Strategic Innovation In MIS For Microfinance Market

219

direct online purchases and 30% indirect, the future could be the other way around. “The main challenge is about connecting the full value chain from the MFI in various developing countries to a successful implementation. This means having local partners who know their own market and our one global solution providing the best possible product, platform, services on a trustworthy brand,” says Eugene Danilkis, a founder of Mambu. That this needs to use standard delivery models indicates that the innovation has yet to disrupt the market, strategic though it may be. This is confirmed by the incumbent, Signis Aliks of MBWin, who indicates that, at the moment, they are responding to direct enquiries from small cooperative MFIs, but eventually, they would like to spin off SaaS to a separate SaaS operator who could be one of their existing partners. Therefore, although the pricing model is different, hypothesis 4 about changing the entire business model needs to be modulated. Really big players such as IBM are just not interested in the SaaS market as the client MFIs are too micro for them. They can support initiatives such as Mifos and provide their cloud, but they have a big challenge in trying to target their product and bank-focused business model at small and medium MFIs. Essentially, this is because they are used to million-dollar contracts and their cost structure does not permit them to downscale as broadly as needed. However, if vendors already have financial packages, such as Oracle’s Flexcube, they may eventually enter into the market to serve their existing banking clients that downscale into microfinance. Nevertheless, they are unlikely to be a threat in the Tier 2, Tier 3 and Tier 4 markets, which are essentially where SaaS currently provides the biggest benefits.

3. Blocks and challenges One technological block is managing flexibility in the system. MFIFlex found that “instant updates and innovation could spoil the individual MFI’s unique modifications. A Sandbox methodology was used to first make a local functional replica (more than a data backup) and then make the required changes on that frozen replica version. After testing changes in Sandbox, the changes are pushed to the real operating version.” (Ashta, 2012). Thus, it may turn out, as suggested by hypothesis 3, that a SaaS may require reduced flexibility. This is admitted by MBWin manager Signis Aliks who indicates that although their product is configurable to almost any financial product, the support costs of different configurations are unbearable for the small cooperatives selecting the SaaS model, and they therefore choose groups of customers with similar needs

220

Chapter Ten

for the SaaS model. Thus, markets with demanding customers or peripheral needs are left to new entrants, in line with the discussion in hypothesis 7. It may be noted that all this is in line with the concept of strategic innovation: To provide a basic version of a product in a way that it becomes the new model. The biggest institutional bottleneck is lack of customer awareness about SaaS. Researchers, practitioners, consultants and global networks need to emphasize educating MFIs on the benefits of cloud computing and what it can mean for the microfinance industry as an enabling mechanism. A second bottleneck, quite comprehensible, is that people prefer to work with local vendors in their local language. SaaS has a global viewpoint. This makes it difficult to enter the market. Third, a potential regulatory bottleneck may come in if countries, especially large countries like India and China, decide that data must be hosted in the country and cannot be hosted outside. While a hybrid system can maintain the data in the country, the main system has to lie outside in the developed world which has the kind of infrastructure that permit highquality functioning and safeguards. Fourth, diffusion rates of software can go up if free software is provided to encourage adoption and imitation. However, if the early adopters do not talk about their experience to other potential users, the imitation lag can go beyond the sustainable capacity (in terms of time) of the SaaS innovator, who, as we have seen, is usually a small new incumbent. The SaaS operators therefore need free offers to be contingent on promoting the SaaS model (through, for example, tweets or status updates).

Conclusion SaaS may be a radical innovation and may provide a sufficiently different pricing model to permit strategic innovation, either by new entrants or by smaller existing vendors. In the MIS for microfinance market, so far, incumbents have been slow to respond, indicating that the strategic innovation may lead to disruption rather than sustaining the existing incumbents. Part of the reason may be that the incumbents are operating in the small, medium and medium-large markets, of which the latter are more lucrative. However, SaaS is lucrative only for the small MFIs if they do not have many users and clients. Therefore, the SaaS offering may be

SaaS: Strategic Innovation In MIS For Microfinance Market

221

taken up only by the very small manufacturers. One possibility is that the SaaS operator is different from the developer. One limitation of this study is that we have not looked at the smaller off-the-shelf operators, some of whom may just host their existing API offering on the web while others may develop multi-tenant cloud offerings, as Temenos claims to have done. To the extent that many of these are smaller players, they are quite close to the new challengers. But to the extent that Temenos may be followed by bigger providers into this market, the incumbents may be edged out not by smaller entrants but by big banking-oriented software vendors that suddenly find that the cost of deployment in the SaaS model is permitting lower marketing expenses to reach out to smaller MFIs. Very recently, a leading MFI network with over 5 million clients in 20 countries has engaged Temenos to develop the software to incorporate social performance indicators for microfinance6. Thus, a possible lesson for the theory of strategic innovation for niche markets from our case study is also that the disruption to niche markets may not necessarily come from new small entrants, but from even larger entrants who now find entering the niche market profitable. In either case, incumbents who do not move may find themselves losing out. To overcome this, they may buy out the successful innovators.

References Abernathy, W.J. 1978. The productivity dilemma: Roadblock to innovation in the automobile industry. Baltimore, MD: Johns Hopkins University Press. Armbrust, Michael, Armando Fox, Rean Griffith, Anthony D. Joseph, Randy Katz, Andy Konwinski, Gunho Lee, David Patterson, Ariel Rabkin, Ion Stoica, and Matei Zaharia. 2009. Above the Clouds: A Berkeley View of Cloud Computing. UC Berkeley Reliable Adaptive Distributed Systems Laboratory. Ashta, Arvind. 2012. MFI Flex: A case study of a SaaS for Microfinance operator. Microfinance Focus, http://www.microfinancefocus.com/mffnews/mfi-flex-case-study-saasmicrofinance-operator. Ashta, Arvind, and Jiten Patel. 2013. "Software as a Service: An opportunity for disruptive innovation in the Microfinance software market?" Journal of Innovation Economics no. 11 (1):55-82. Asselineau, Alexandre. 2010. "Quand un ‘cas d'école’ d'innovation stratégique est un échec...: Une lecture en termes de légitimité." Revue française de gestion (203):71-84. doi: 10.3166/rfg.203.71-83.

222

Chapter Ten

Bittman, Thomas 2012. Ten Steps to Building Private Cloud Services. Gartner. Charitou, Constantinos D., and Constantinos C. Markides. 2003. "Responses to Disruptive Strategic Innovation." MIT Sloan Management Review no. 44 (2):55-63. Christensen, Clayton M. 2006. "The Ongoing Process of Building a Theory of Disruption." Journal of Product Innovation Management no. 23 (1):39-55. doi: 10.1111/j.1540-5885.2005.00180.x. Christensen, Clayton M., Richard Bohmer, and John Kenagy. 2000. "Will Disruptive Innovations Cure Health Care?" Harvard Business Review no. 78 (5):102-112. Christensen, Clayton M., Mark W. Johnson, and Darrell K. Rigby. 2002. "Foundations for Growth." MIT Sloan Management Review no. 43 (3):22-31. Christensen, Clayton M., Fernando F. Suárez, and James M. Utterback. 1998. "Strategies for Survival in Fast-Changing Industries." Management Science no. 44 (12):S207-S220. Eisenhardt, K. M. 1989. "Building Theories from Case Study Research." Academy of Management Review no. 14 (4):532-550. Gartner. 2011. Data Center Conference. Gartner. Gaul, Scott. 2011. A first look at technology provider market share In Microbanking Bulletin: Microfinance Information Exchange. Govindarajan, Vijay, and Anil K. Gupta. 2001. "Strategic Innovation: A Conceptual Road Map." Business Horizons no. 44 (4):3. Govindarajan, Vijay, and Chris Trimble. 2004. "Strategic Innovation and the Science of Learning." MIT Sloan Management Review no. 45 (2):67-75. Govindarajan, Vijay, and Chris Trimble. 2005. "Organizational DNA for Strategic Innovation." California Management Review no. 47 (3):4776. Henderson, Rebecca M., and Kim B. Clark. 1990. "Architectural Innovation: The Reconfiguration of Existing Product Technologies and the Failure of Established Firms." Administrative Science Quarterly no. 35 (1):9-30. Hudon, Marek, and Arvind Ashta. 2013. "Fairness and microcredit interest rates: From Rawlsian principles of justice to the distribution of the bargaining range." Business Ethics: A European Review. no. 22 (3):277-291 Lawrence, Thomas, Roy Suddaby, and Bernard Leca. 2009. Institutional Work: Actors and Agency in Institutional Studies of Organizations. Cambridge, U.K.: Cambridge University Press.

SaaS: Strategic Innovation In MIS For Microfinance Market

223

Maes, Jan P. , and Larry R. Reed. 2012. State of the Microcredit Summit Campaign Report 2012. Washington, D.C.: Microcredit Summit Campaign. Maguire, Steve, and Cynthia Hardy. 2009. "Discourse And Deinstitutionalization: The Decline Of Ddt." Academy of Management Executive no. 52:148-178. doi: 10.5465/amj.2009.36461993. Markides, Constantinos. 1997. "Strategic Innovation." Sloan Management Review no. 38 (3):9-23. —. 1998. "Strategic Innovation in Established Companies." Sloan Management Review no. 39 (3):31-42. Mathews, John. 2006. "Dragon multinationals: New players in 21st century globalization." Asia Pacific Journal of Management no. 23 (1):5-27. doi: 10.1007/s10490-006-6113-0. Mazzarol, T. I. M., and Sophie Reboud. 2008. "The Role Of Complementary Actors In The Development Of Innovation In Small Firms." International Journal of Innovation Management no. 12 (2):223-253. Osterwalder, Alexander, Yves Pigneur, and Chistopher L. Tucci. 2005. "Clarifying Business Models: Origins, Present, And Future Of The Concept." Communications of AIS no. 16:1-25. PWC. 2011a. Cloud Computing in the Middle Market. PWC. —. 2011b. Cloud Computing Navigating the Cloud. PWC. Schlegelmilch, Bodo B., Adamantios Diamantopoulos, and Peter Kreuz. 2003. "Strategic innovation: the construct, its drivers and its strategic outcomes." Journal of Strategic Marketing no. 11 (2):117. Shuen, Amy. 2008. Web 2.0: A Strategy Guide. Sebastopol, CA: O'Reilly Media Inc. Sundbo, Jon. 1997. "Management of Innovation in Service." Service Industries Journal no. 17 (3):432-455. Totango. 2012. SaaS Free Trial, Freemium and Pricing Benchmark. Totango. van den Hoed, Robert, and J. Vergragt Philip. 2004. "Institutional Change in the Automotive Industry." Greener Management International (47):45-61. Yin, R. 1994. Case study research: Design and methods. Beverly Hills, CA: Sage Publishing.

224

Chapter Ten

Notes 1

My thanks to Jiten Patel, Snehal Fulzele, Eugene Danilkis, Frederik Pfisterer and Signis Aliks, founders/ managers of the different enterprises, for their time in helping me understand this market. 2 http://www.microfinancegateway.org/p/site/m/template.rc/1.26.12263/#1 3 Using 2011 data downloaded from MIX on February 3, 2013. 4 http://msd-tt.com/newsletters/Vol7Issue2_newsletter.pdf 5 These comments should keep in mind some opacity in pricing information and it is not really possible to compare the offered cost of running the SaaS model with some with the values of the traditional model from the Table 4. Such comparison ignores the fact that the costs in Table 4 include also the costs of training, implementation and the data migration. These costs often account for 60-80% of the total cost of the implementation of the software. One would have to add similar costs also to the SaaS model to have a comparative pricing for the first year. Notice also that the costs provided by SaaS vendors are annual costs for every year of operation while the traditional approach gives the one time advance cost and the following annual costs are much lower, usually 10-25% of the license (only license, not the whole implementation) cost. See also http://www.journaldunet.com/solutions/expert/50780/comparer-les-couts-entrelicence-globale-et-saas---pas-si-simple.shtml 6 http://www.opportunity.org/press-releases/opportunity-international-engagestemenos-to-gauge-social-impact-of-microfinance-on-fighting-poverty/ March 6, 2013

CHAPTER ELEVEN RISKS AND MITIGATION IN CLOUD COMPUTING FOR MICROFINANCE BRYAN BARNETT, PH.D., J.D. Introduction In recent decades, microfinance has achieved remarkable growth. Beginning with a limited number of institutions narrowly focused on microcredit, it is now a mature industry with thousands of institutions advancing financial inclusion across a range of products and services. Still, the microfinance marketplace is relatively bifurcated between a few very large institutions and a very large number of small ones. That so few institutions have been able to achieve really significant scale is one of the greatest shortcomings of microfinance to date. The reasons for this are many, but an important one is the inability of smaller institutions to afford the information technology needed to manage a large transaction-intensive enterprise. Information technology is expensive to purchase and it requires support from skilled staff that are difficult for MFIs to retain. Yet, because MFIs cannot grow without the efficiency and transparency that sophisticated information systems alone can deliver, this remains one of the central challenges facing the future expansion of the microfinance sector. To many observers the most promising potential solution to this problem is the emergence of cloud computing, specifically Software-as-aService (SaaS). In this computing model the end user does not have to purchase any expensive software. They need not own or maintain any expensive servers or do more than maintain basic desktop computers and Internet connectivity. As a result, this model has the potential to not only lower costs significantly, but provide better security, greater agility, and reduced risk (Iyer, 2012). So far, the market for cloud-based services for microfinance institutions is limited, but early results are promising. Through companies like

226

Chapter Eleven

Mambu, MicroPlanet, Temenos or MFI Flex, small to medium sized institutions now have access to powerful information technology that would otherwise be completely unaffordable. But while the benefits of cloud-based services are genuinely compelling, there are risks as well. And if those risks should not be exaggerated, neither should they be ignored. After all, the cloud customer is entrusting mission critical data, perhaps their most valuable asset, to the physical custody of a third party. Loss of that data, or even temporary loss of access to that data, will have a profound impact on any microfinance institution. In this chapter we look carefully first at the essential components of a SaaS offering and then at the risks inherent in these offerings. And we will consider how those risks might be addressed by the MFI cloud customer.1 In any such discussion it is important to remember that cloud computing is quite new and in many respects the marketplace is still immature. Many elements such as regulatory provisions, audit standards and insurance are still relatively undeveloped. It is very likely that these aspects of the cloud services marketplace will evolve as the market matures. In the meantime potential cloud customers are well served to be fully aware of both the risks and benefits of emerging services.

A Cloud Vernacular One of the hallmarks of significant technological innovation is often the need for new vocabulary to describe its features. In the case of cloud computing it is important to distinguish among different types of cloudbased services and to differentiate different roles played by suppliers and consumers of these services. Accordingly we will here adhere to the widely recognized lexicon proposed by the American National Institute of Standards & Technology (Fang Liu et al., 2011). x Cloud Consumer: an individual or organization that acquires or uses cloud-based services. x Cloud (Service) Provider: a purveyor of one or more cloud-based services. x Software-as-a-Service (SaaS): the provision of a software application accessed by end users who do not own or license the application but pay for using it. x Infrastructure-as-a-Service (IaaS): the provision of foundational IT capacity, including CPU, storage, networking, etc. that cloud consumers access and use to run applications.

Risks and Mitigation in Cloud Computing for Microfinance

227

x Platform-as-a-Service (PaaS): the provision of a specialized programming environment allowing the development of custom software applications hosted on an associated infrastructure. In addition to these terms, it is also important to distinguish between public and private clouds. In the case of very large organizations that maintain their own data centers, certain elements of cloud technology may be used in providing data services to the host organization. Such cloud installations are referred to as private clouds because they are owned and operated by a single organization for its own use. By contrast, public clouds are operated by one organization for the benefit of others, to whom cloud services are sold. Private clouds have very little application in the microfinance arena, since there are few if any institutions large enough to support their own cloud infrastructure. Rather, it is public clouds that offer great value to microfinance institutions and are the primary focus of attention here.

The Anatomy of Cloud Services In one respect, the phrase “the cloud” is unfortunately misleading. For the cloud is not a singular thing but is a set of services composed of a number of different components often supplied by different entities. The reference model diagramed in Figure 1, developed by the U.S. National Institute of Standards, illustrates the different elements of a cloud-based service. (Fang Liu et al., 2011) While all the elements in Figure 1 together make up the cloud computing universe, for purposes of the present discussion we will focus on just two: the cloud service and the audit components. Every SaaS offering of microfinance software has three key elements. There is first the service provider who assembles the other elements, and owns the customer relationship, including billing and technical support. Next there is the core software application (i.e. for portfolio management or accounting) that the MFI customer will actually use. Finally, there is the underlying infrastructure of servers and networking, etc. where the software application runs. It is important to realize that these three components are independent of one another and may be supplied by the same or different companies. For example, the service provider may be the developer of the core software (as with Mambu) or may license the core application from another company (as with MicroPlanet using software from either InfraSoft Technologies or Temenos). The service provider may provide their own application hosting, or more likely will outsource this to

228

Chapter Eleven

a public cloud provider as do Mambu (hosting with Amazon), Temenos (hosting with Microsoft) and MFI Flex (hosting with SalesForce).

Figure 1: NIST Cloud Reference Model

Regardless of where it originates, the core software used in a SaaS offering will not be significantly different from the same software installed and run locally at an MFI. But the infrastructure for hosting that software is completely different in the cloud context. In a cloud scenario the physical facility housing the actual servers, software and data, known as a “data center,” is a specially built facility housing anywhere from a few dozen to tens of thousands of computer servers. In its simplest incarnation, a data center may be nothing more than a room where separate individual businesses lease space for their own servers. But increasingly, modern data centers are very large installations operated by companies like Microsoft, Amazon, Google, IBM or RackSpace where all the machines are owned and managed by the data center operator, who supplies the complete IT infrastructure (IaaS) to customers who merely run their own software in the facility but don’t own or control any of the underlying physical infrastructure. In addition some companies provide a platform for the development of custom applications. Microsoft’s Azure or SalesForce’s Force.com are prominent examples

Risks and Mitigation in Cloud Computing for Microfinance

229

In these large state-of-the-art data centers lies the technical essence of true cloud computing. Here the computers are managed in a very unique way. Using a technology called “virtualization” computing resources are “pooled” in a way that allows different resources (CPU, memory, storage, etc.) to be very rapidly allocated to particular applications, and just as rapidly released when no longer needed. The effect is that computer processing becomes very much like electricity, a resource that is constantly available at virtually any level of demand. And just as the modern user can use and pay for electricity without the need to own or operate their own generator, the user of the modern commercial data center simply uses and pays for computing power without owning any of the required hardware or infrastructure. And so it is that companies like Amazon (the early pioneer of this business) offer access to their data centers to the public, providing a complete environment in which customers can run their own applications using more or less computing resources as needed and paying only for what is used. By virtue of the great economies of scale and superior optimization of their expensive computing investment, these true cloud data centers are able to offer extraordinary value to customers. This is especially important to small and medium size service providers who could never afford infrastructure of this power and quality from their own resources. Thus the ability to compose cloud services from components provided by different entities can produce a much higher quality SaaS offering than one relying only on home-grown components.

Types of Cloud Risks With the foregoing in mind we can now turn attention to the distinct types of risks associated with SaaS offerings from the cloud. Evaluating risks in any particular case requires both an estimate of any potential loss together with an estimate of the likelihood that a specific type of event will occur. Whereas the usual approach to this is to look at historical data, in the case of very recent innovations like cloud computing the historical record is limited. Nonetheless, anecdotal evidence is indicative of the possibilities. Broadly speaking there are four basic risks inherent in any cloudbased service offering as suggested in Table 1. We here discuss each in turn.

Chapter Eleven

230 Risk Internet Service Interruption Data Center Compromised Unauthorized Access to Data

Business Failure/Change of control

How likely to occur Somewhat common in developing countries. Rare but not unknown. Least likely with large established data centers. Difficult to assess because often undetected at time of occurrence and victims are reluctant to disclose. Most likely among smaller early stage companies with limited resources.

Potential Impact Service levels compromised but no loss of data. Complete service interruption with potential loss of data. Direct financial losses and immediate loss of customer trust. Immediate loss of access to data and operational software.

Table 1: Cloud Risks. Internet service interruption. The Internet is designed to be very fault tolerant and is not subject to wide-area outages in the way that the electrical grid might be. But in areas where microfinance is most prevalent, local connections to the Internet can be very unreliable due to such things as power outages or when available bandwidth is insufficient to support growing traffic volumes. The resulting reduction in performance can significantly affect the usability of SaaS applications, and while there is no risk of lost data it can slow business processes. While the quality of local connections may improve over time, other types of risks are unlikely to disappear. For example, during the Egyptian revolt that toppled the Mubarak regime, the regime cut Internet service in an effort to thwart social media (Richtel, 2011). In February 2012, an undersea cable was cut off the coast of Kenya, causing telecoms and ISPs to scramble for alternate routes for Internet traffic (Mbugua, 2012) and prompting Safaricom to launch plans to relocate the software platform for its popular M-Pesa mobile money service from Germany to Nairobi (Mumo, 2012). While these types of events do occur and are hard to predict, the frequency should not be exaggerated. Measured against the millions of Internet web sites and services operating around the clock every day, the likelihood of any single customer experiencing this type of outage is certainly very small. But they do occur and serious business interruption is a likely consequence for the affected SaaS customer.

Risks and Mitigation in Cloud Computing for Microfinance

231

Data Center Stability and Security. The requirements for physical, environmental and energy security of a data center are well understood, if not always fulfilled. No doubt in part money is a factor, and in this regard larger established providers have the advantage of greater resources and experience. Properly designed data centers are more common in developed countries, but not unknown elsewhere. Again, such losses may not be common, but they do occur and the possibility should not be ignored (Miller, 2008, 2012). Data centers must be physically secure not only from unauthorized access, but from environmental hazards and power loss as well. The data center must be protected against severe weather, floods, earthquakes or similar events. The electrical systems and cooling systems must be completely reliable. These are the obvious elements of stability and security. Less obvious, but increasingly important, is the reliability of the complex software used to manage resources within the data center. Failure of this management software can cause an entire data center to cease operating properly. In April 2011, Amazon experienced a 3-day outage at one of its data centers as a result of a maintenance procedure gone awry. Many businesses were disrupted for that period (Amazon Inc., 2011). Because it is easy to store copies of data in multiple locations, failure in one data center need not cause permanent loss of data. But absent adequate preparations for automatic failover to alternative sites, serious business interruption is possible. Unauthorized access to data. Notwithstanding a secure data center, vulnerabilities in software can allow unauthorized users to access data resulting in loss of privacy or outright theft of data. In many respects, the worst part of such incidents is that they may go undetected for a long time. Here again, the unique way that computing resources are pooled in a modern cloud data center potentially increases the risk of unauthorized access to data or compromise of privacy (Owens, 2010). In considering this type of risk it is important to bear in mind that computing technology is complex and no technology is immune from security flaws. So the relevant question is not whether cloud-based solutions are perfectly secure, but whether they are inherently less secure than the alternatives. The security of any software solution depends first and foremost upon how well it is designed, tested and configured upon installation. Security flaws in application software will be no more or less a threat because the application is deployed in the cloud rather than on a local on-premises server. That said there are some differences unique to cloud deployments that present unique security issues.

232

Chapter Eleven

The efficiency of the true cloud infrastructure is achieved through the sharing of resources. In practice this means heavy reliance on so-called “multi-tenant” architectures in which different customers share a single operating environment or a single application instance while the operating environment or the application itself is responsible for keeping each customer’s data separate from data belonging to other customers. Whether these technologies pose a unique risk that customers’ data will be comingled or that data will be vulnerable to hacking by outsiders is largely a question of how well a particular multi-tenant environment or application is designed and implemented. But the extra complexity inherent in these technologies has prompted concerns about the assumed trustworthiness of cloud solutions (Owens, 2010). At first glance, these concerns can appear decisive. Yet it is important again to recall that not all cloud offerings involve multi-tenant applications and those that do are not necessarily a greater risk than other approaches. Local on-premises solutions operated by small understaffed organizations are often assumed to be much more secure than they are in fact, and in such instances a cloud solution may be more secure and reliable than the local one. Business failure or transfer of control. In any outsourcing situation, the bankruptcy or failure of the firm to which work has been outsourced is a potential risk. In few outsourcing scenarios, however, is the physical, and perhaps exclusive, custody of mission critical data surrendered to the outsourcing partner. What might in other contexts result in the interruption of normal business can, in the SaaS context, mean an imminent threat to the survival of the SaaS customer. Not only could there be loss of access to data, but there could potentially be no way to recover data if there is no way to regain access to the software supplied by the SaaS provider. This is a particular concern where the provider is a new or small company with a limited track record, as is often the case in a new market such as SaaS for microfinance. Particularly where business models are unproven and new entrants are thinly capitalized, customers must be concerned with business continuation should the provider cease operations. To date there have been no significant occurrences of this type in the microfinance market, though a decision by the Grameen Foundation to cease supporting the Mifos open source project did leave a number of Mifos customers in a position they did not anticipate. At this point, it appears too early to conclude that the business model for SaaS for microfinance has been proven, so customers need to be alert to the possibility that some entrants into the market may not survive or will be acquired.

Risks and Mitigation in Cloud Computing for Microfinance

233

Strategies for Risk Mitigation That we have devoted significant space to describing risks should not be taken to mean that SaaS offerings are to be avoided. Any IT strategy has risks and all of them must be weighed against the large potential benefits of SaaS for an MFI. So the remaining sections of this chapter are focused on three avenues for possible risk mitigation: 1) due diligence; 2) contract concerns; 3) insurance. We will look at these in turn. Due Diligence. Full and accurate information is vital to any risk assessment and knowing what questions to ask is critical for the potential SaaS customer. A (non-exhaustive) sample list of typical questions is shown in the side bar. As noted earlier, any SaaS provider may be supporting their own software or may merely be a licensee of someone else’s product. They may provide their own data center facilities or may rely on facilities provided by others. It is important to establish what level of responsibility the SaaS service provider has over these components and to assess the relative quality of each. Where possible, of course, anyone conducting due diligence would want to obtain firsthand information through on-site inspection. As a practical matter this is normally impossible for SaaS customers. They may lack the requisite expertise and ƒWho is supplying data center facilities or the data centers to be inapplication software? spected are invariably disƒWhere is data physically stored? tant and inaccessible. CusƒWhat are the data center facilities like? tomers are therefore priƒAre any audit reports available? marily reliant upon the ƒWhat terms of service govern any third-party reputation of the vendor services upon which the SaaS provider relies? and any reliable third-party ƒHow financially sound are any companies assessments that might be providing essential components of the SaaS available. Given the potenoffering? Are they small and thinly capitalized tial importance of such or large and well established? third party assessments, ƒWhat policies and procedures exist for data known as data center aureplication, fail-over, or disaster recovery? dits, it is striking that ƒHow will the customer’s data be managed? standards for these audits Will it be in a single isolated environment are not fully mature. dedicated to the customer, or in a multi-tenant environment shared with other customers? Though the notion of a ƒWhat security policies and procedures are in data-center audit is familiar place? To what extent are software events in the IT services industry, logged for future inspection? the current reality is that

234

Chapter Eleven

there is no widespread consensus on standards for such audits. And though there are a number of candidates, there remain significant disparities among them. Some putative standards do not require a direct inspection of data center conditions and procedures, while few articulate clear benchmarks against which such conditions and operations are to be evaluated. This situation perhaps reflects a certain immaturity of the IT audit industry, a situation that will hopefully improve over time. For the moment, asking a prospective SaaS provider to confirm the results of any audits of facilities upon which the provider relies is doubtless the best approach. Contract Concerns. Purchasing a SaaS offering is entering into a contractual relationship with the SaaS provider, generally known as the “terms of service” (TOS) or “service level agreement” (SLA”). To this may be added an acceptable use policy (AUP) as well as a privacy policy. For the SaaS customer there are several concerns. What are the terms of the agreement? Are they negotiable? Is the agreement enforceable? The core value proposition of cloud service providers is that they offer services at low cost because they operate very efficiently. Providers therefore cannot negotiate customized terms with individual customers without increasing their costs. In practice, this means that there are two different market segments for cloud services, one that serves very large customers who have sufficient leverage to negotiate custom arrangements, and another for small customers, including small and medium businesses as well as individuals, who have no power to alter the standard terms offered by providers. Needless to say, the non-negotiable terms offered by cloud service providers are generally weighted in their favor. A list of items potentially covered in the terms of service for a SaaS offering is lengthy and beyond the scope of this chapter but is available from the Cloud Standards Customer Council (Meegan, 2012). Any terms of service will describe the services to be provided, fees to be charged and similar items. Of greater concern from a risk management point of view are terms relating to the availability of the service, the management (including security) of data, procedures for testing and upgrading of service components and separation should the agreement come to an end. Research suggests that many of these terms are similar across major providers (Bradshaw et al., 2010). Most important for cloud consumers, there is a strong trend among service providers to disclaim any type of warranties, including responsibility for data losses. In the absence of a specially negotiated agreement, the most that cloud providers will offer in the way of compensation for any service failure is service credit. No SaaS provider will assume any liability for any losses suffered by the SaaS

Risks and Mitigation in Cloud Computing for Microfinance

235

customer as a result of the provider’s failure to meet the terms of their service agreement. It is doubtful that more than a very few, if any, microfinance institutions are large enough to be able to negotiate custom agreements that might soften these terms. Moreover, where microfinance SaaS providers rely on outsourced infrastructure for hosting their applications they are subject, in turn, to the terms of service provided by their supplier. Thus a developer of SaaS software for microfinance that hosts their application with a company like Amazon or Salesforce is subject to the terms of service of those companies, whose terms will not be negotiable. So it is important to be aware of all the several layers of service terms that may govern any SaaS offering. It is important to realize the absence of warranties is not equally significant in all instances. In some cases protection against losses is easy enough that placing responsibility with the SaaS customer is not objectionable. For example, major data center operators will most often have multiple data centers in dispersed locations and allowing the customer to purchase data replication across several of them provides ready protection from both service interruptions and data loss due to failures at any one location. Moreover, some providers will offer replication to a local server owned by the customer (used only for local storage of archived data), or provide routine copies of data on CD, providing assurance that there is a valid archive in the event of a service failure of the SaaS provider. But if it is easy to accept a lack of warranties where it is so easy to provide adequate protection, it is another matter where the SaaS customer is not in a position to know about or take remedial action. For example, where a SaaS provider is operating its service in a pooled multi-tenant environment over which it has sole control it seems less reasonable to disclaim any responsibility for the security of data in that environment, particularly because solid data backups may be no real protection. This is because, if data is comingled or systems are compromised by cyberattacks it may be some time before the breach is discovered and losses may not involve any loss of data that would need to be recovered from an archive. This is an area where one might expect some improvement as the market matures and competition increases among providers. But for the moment SaaS customers for microfinance cannot look to SaaS providers to make good any losses suffered as a result of any service failure. A contract dispute with a SaaS provider is only one of several circumstances in which the question of legal jurisdiction is raised. There is the issue of whether a local government in the location where data is stored might have access to it for law enforcement purposes even though the

236

Chapter Eleven

owner of the data is far away and not subject to local jurisdiction. Such concerns have been raised especially with respect to the USA Patriot Act, a statute giving broad authority to law enforcement authorities in the United States (Levi, 2012). The question of whose law governs and what courts might be available to a SaaS customer is therefore an important concern. Unfortunately for SaaS customers, providers’ terms of service normally specify that the relationship of the parties will be governed by the law of a jurisdiction convenient to the provider and any litigation must take place there (Bradshaw, 2012). Even where the customer is not obligated by contract to pursue remedies in a foreign jurisdiction, ordinary tests used to determine whether a court has jurisdiction are likely to favor the provider. In a conventional business setting the provider would be within the jurisdiction of the courts where the customer’s business is located because the provider has a nexus of connection to the place where it delivers services. But in the normal SaaS situation the service provider has no connection to the place where its services are consumed, no physical presence or other activity there. Its services are merely accessed remotely by the customer. So under traditional tests the provider is not subject to the jurisdiction of the courts most accessible to the customer, who must pursue any claims in foreign courts. So long as they cannot look to SaaS providers to make good any losses caused by a failure of their services, the customer’s best alternative is to insure against this risk. Here again, the marketplace appears still to be in its infancy. Large businesses have long been able to insure against losses due to technology failures. But the policies traditionally cover risks associated with IT systems owned and operated by the end user. Coverage specifically for risks associated with cloud-based services is new, but has recently begun to appear. At least one major insurer now offers a policy enhancement specifically for cloud risks that offsets lost income and the cost of transitioning to a replacement cloud provider in the event of certain cloud-related losses (Marsh, 2012). This type of insurance is targeted a very large organizations and, in this particular instance, is only available as an adjunct to a comprehensive cyber policy. It is therefore unlikely to be accessible to most microfinance institutions. But if familiar patterns hold, this type of insurance will eventually become more widely available and more affordable as insurers become more comfortable modeling and pricing the risks. With any insurance coverage cloud customers still need to be alert for policy provisions that would, for example, deny coverage in the event that the covered insured failed to take adequate security precautions (Gold, 2012). Uncertainty over such things as what constitutes “ade-

Risks and Mitigation in Cloud Computing for Microfinance

237

quate security measures” taken by the insured can open the door to costly disputes.

Conclusion Merely detailing the risks associated with a business decision like moving to a SaaS model for IT can be intimidating. Looking over a long list of risks, it is easy to overlook the real chance that any of them may come to pass. The risks identified here are real, but not necessarily likely to occur. The microfinance institution contemplating a move to a SaaS provider must weigh these risks against the very considerable benefits potentially available from a SaaS offering. As with any large scale shift in technology there are increased risks faced by early adopters that diminish as the new technology spreads, suppliers multiply, competition increases, and strategies for risk mitigation evolve. It is reasonable to expect that in time terms of service will become more varied, legal and regulatory frameworks will evolve, and appropriate insurance will become more widely available and affordable. Indeed, perhaps no change in the marketplace would have a more immediate and dramatic effect than the widespread adoption of insurance by either or both SaaS providers and customers. This alone would go a long way toward eliminating the fear that today prevents many potential SaaS customers from making that critical decision to give up custody of their mission critical data to others.

Bibliography Amazon Inc. 2011. “Summary of the Amazon EC2 and Amazon RDS Service Disruption in the US East Region.” Retrieved 11/18/2012 from http://aws.amazon.com/message/65648/ . Ashta, Arvind and Patel, Jiten, “Is SaaS the Appropriate Technology for Microfinance?” (May 11, 2010). Accessed at SSRN: http://ssrn.com/abstract=1604741 or http://dx.doi.org/10.2139/ssrn.1604741 Bradshaw, Simon; Christopher Millard & Ian Walden. 2010. “Contracts for Clouds: Comparison and Analysis of the Terms and Conditions of Cloud Computing Services”. Queen Mary University of London, School of Law Legal Studies Research Paper No. 63/2010. Fang Liu et al., 2011. “NIST Cloud Computing Reference Architecture”. NIST Special Publication 500-292. Accessed at http://collaborate.nist.gov/twiki-cloud-computing/pub/CloudComput ing/ReferenceArchitectureTaxonomy/NIST_SP_500-292_-_090611.pdf

238

Chapter Eleven

Fang Liu et al., 2011a. “NIST Draft Taxonomy and Definitions.” Accessed at http://collaborate.nist.gov/twiki-cloud-computing/pub/CloudComput ing/ReferenceArchitectureTaxonomy/Taxonomy_Terms_and_Definition s_version_1.pdf Gold, Joshua. 2012. “Protection in the Cloud: Risk Management and Insurance for Cloud Computing.” Journal of Internet Law, June 2012, Vol. 15, Issue 12, pp 1-28. Iyer, Bala; John C. Henderson, 2012. “Business Value from Clouds: Learning from Users.” MIS Quarterly Executive, Vol. 111, Issue 1, pp. 51-60. Levi, Stuart D. "Cloud Computing: Understanding Security and Jurisdictional Issues", Accessed at: http://www.skadden.com/insights/cloudcomputing-understanding-security-and-jurisdictional-issues Marsh Inc. 2012. “Marsh CloudProtect – A Cyber Policy Enhancement”. Accessed at http://usa.marsh.com/ProductsServices/MarshSolutions/ID/22090/Mars h-CloudProtectA-Cyber-Policy-Enhancement.aspx. Mbugua, James. 2012. “Kenya: Undersea Cable Cut Disrupts Local Internet Traffic”. Retrieved on 11/10/2012 from: http://allafrica.com/stories/201202281238.html Meegan, John et al., 2011. “Practical Guide to Cloud Service Level Agreements”. Cloud Standards Customer Council, April 2012. Accessed at http://www.cloud-council.org/PGCloudSLA040512MGreer.pdf . Miller, Rich. 2008. “Extensive Damage at The Planet’s Data Center”. Data Center Knowledge, June 2, 2008. Accessed at: http://www.datacenterknowledge.com/archives/2008/06/02/extensivedamage-at-the-planets-data-center/ Miller, Rich. 2012. “Data Center Fire Disrupts Key Services in Calgary”. Data Center Knowledge, July 16, 2012. Accessed at: http://www.datacenterknowledge.com/archives/2012/07/16/datacenter-fire-disrupts-calgary/ Mumo, Muthoki & Charles Wokabi. 2012. “M-Pesa platform coming home soon.” Kenya Daily Nation, November 14, 2012, p. 34. Owens, Dustin. 2010. “Securing Elasticity in The Cloud”. Communications of the ACM. Vol. 53, Issue 6, June 2010 pp 46-51. Richtel, Matt. 2011. “Egypt Cuts Off Most Internet and Cell Service”. New York Times, January 28, 2011. Accessed at: http://www.nytimes.com/2011/01/29/technology/internet/29cutoff.html

Risks and Mitigation in Cloud Computing for Microfinance

239

Notes 1

In order to focus discussion on those risks uniquely associated with a cloud-based service, we will assume throughout that any MFI takes the ordinary precautions required of any information technology operator such as managing user accounts and password access, etc. Instead we here focus on those risks unique to the cloud.

PART THREE MIS IMPLEMENTATION IN MFIS GAURAV SINHA

Microfinance, which was conceived as a social innovation, grew by leaps and bounds across the world owing to certain key features like accessibility, availability and affordability amongst several others. The growth issues and presence of age-old and large informal (moneylenders) as well as formal (banking and finance) sectors necessitated the microfinance institutions (MFIs) to think outside the box to cater to the unmet needs of the population demanding these services. Around the same time, the world was witnessing growth in many other sectors because of the implementation of new information and communication technologies (ICT). Data and information became inseparable constituents of any organization, with the costs associated with data storage, retrieval and processing dropping sharply as a result of implementing ICT. Studies and experience indicate that an effective and efficient information system is necessary to manage issues related to growth and competition. A trend has started emerging whereby organizations focus more on efficient management of data and information to address growth-related issues and gain competitive advantage. Emerging or new industries like microfinance, which were facing several pressures during their developmental stage looked towards information technology to accelerate their growth. Experience across the world has shown that well-managed information systems in MFIs yield several benefits like improved operational efficiency, informed decision making, transparency and competitive advantage, amongst others. As we have seen in the last few chapters, several studies have established the benefits of using management information systems (MIS) in MFIs. MFIs have consciously adopted this approach as a means of providing efficient and cost-effective services, with some taking a bolder approach of using advanced technologies, including the use of automated systems.

242

Part Three

However, incorporating MIS in the DNA of any organization is fraught with challenges. Many studies, like the one from Ali Ahmad of First MicroFinanceBank Ltd., highlight such challenges. It is an uphill struggle to implement a good MIS in any organization because of the number of problems common to any change initiative. Generally, the implementation of any change initiative, like setting up a management information system, demands changes in an organization’s culture. This “change in culture” also means change in the behavior and practices of the people within the organization (Fullan, 2007). This is why implementation of MIS is such a significant challenge. Added to this, the growth and maturity levels of microfinance in each country are distinct and varied. This part of the book highlights issues and challenges in the implementation of MIS systems in MFIs at various developmental stages, located in different geographies. From the world’s largest democracy, India, to the conflict-afflicted Democratic Republic of Congo (DRC), this part brings forth diverse perspectives on MIS implementation in the MFIs in three different countries. The first chapter in this section examines a model in the Indian context. In Chapter 12 Nyapati and Sandeep make an attempt to identify relevant success factors and their corresponding metrics for MFIs using the DeLone and McLean (D & M) model. They focus on three success factors viz., system quality, information quality and net benefits. They have used techniques of decomposition and quantification to analyze these three success factors on various metrics. The authors further propose to decompose system quality into three components, namely, availability, reliability and usability. Similarly, information quality has been decomposed into timeliness and accuracy, and net benefits into individual and organizational impact. The chapter analyzes data from a leading Indian MFI, Bharatiya Samruddhi Finance Limited (BSFL) based on these three success factors. It also draws some key lessons for the MFIs in ascertaining the impact of MIS. Couchoro, in Chapter 13, presents a case from Togo. Interestingly, the Central Bank in Togo had taken measures intended to introduce a single software for all MFIs in the country, because in many cases the information system of MFIs was not in accordance with the country’s legal framework. Couchoro highlights the experiences of, and the difficulties faced by, MFIs in Togo in their growth phase while using MIS. He captures the experience of three MFIs, which together account for over 70% of the market share of the microfinance sector in Togo, in implementing automated information systems in their organizations. The study revolves around four key points: identification of needs of the MFI, involving the Board of Directors, the dilemma in choosing the MIS, and

MIS Implementation in MFIs

243

satisfaction with the MIS. The author suggests that as MFIs grow and begin to adopt a commercial approach, use of efficient software would certainly provide a competitive advantage. He further recommends the creation of a central resource to identify information system problems faced by MFIs in Togo. While the first two chapters in this part attempt to look at organizational issues related to MIS implementation, the third chapter, Chapter 14, makes an effort to build a case for an efficient information system to prevent over-indebtedness among the clients of microfinance in the Democratic Republic of Congo (DRC). Here, a team of researchers, Isoketsu, Sinha, Annamraju, Sud and Srinivasan, bring forth a case from this poor and conflict-ridden country in sub-Saharan Africa. There are dual challenges confronting the DRC. On the one hand, it has lost its crucial data during the period of conflict. On the other hand, after the conflict the country faces the challenge of poor physical infrastructure for ICT. Both of these have resulted in an information asymmetry prevalent in the country and this affects the evolving microfinance sector too. The authors in this chapter examine the existing structure of information systems, its relevance to the MFIs, and the impact it has on the poor clientele in the DRC. They also present insightful steps to correct the information imbalance by establishing an efficient and up-to-date information system to deliver microfinance services that benefit the poor clientele, without creating debt distress. Given the diversity in the countries, these three chapters provide useful insights and considerations before implementing an MIS in MFIs in different countries. These chapters also provide an opportunity for crosscultural learning on the usage of MIS. For researchers, this part opens up the opportunity for modeling and further research. For service providers, there is ample scope for learning and understanding the factors to be considered before introducing any technology. For the MFIs, it certainly builds a case for adopting the right kind of MIS. For the donors, it provides an opportunity for assessing MIS impacts as well as understanding the problems in implementing an MIS. For all others (including the above stakeholders), these chapters provide interesting insights on how social innovations can be scaled up efficiently and effectively using the right kind of MIS.

244

Part Three

Bibliography Ahmad, A. 2003. Management Information Systems (MIS) for Microfinance. BWTP-Banking with the Poor Network. Retrieved Aug 31, 2013, from http://www.bwtp.org/pdfs/arcm/5Ahmad.pdf Fullan, M. 2007. The new meaning of education change. Teachers College Press

CHAPTER TWELVE PECULIARITIES OF THE MICROFINANCE SECTOR: SUCCESS FACTORS FOR MIS IMPLEMENTATIONS KRISHNA NYAPATI AND SANDEEP MYSORE SESHADRINATH

Introduction The problem of evaluating the success of Information System (IS) implementations has been recognized as a significant issue for many years and has formed the focus of much academic research over the past three decades (see DeLone and McLean, 1992; Seddon, 1997; Seddon et al., 2002). Several characteristics of IS solutions may be identified as the main factors leading to a situation in which understanding and evaluating the success of IS implementations has become a complex task, and the subject of academic research. Indeed, it has been recognized that the problem of defining and evaluating the success of IS implementations is a challenge (Petter et al., 2008) given that success factors are complex, abstract, interdependent and multidimensional. Evaluating IS investments within a typical Return on Investment (ROI) framework is fraught with conceptual and practical problems. While it is relatively easy to compute all the capital and operating expenses related to a specific IS implementation, the aspect of identifying, quantifying and monetizing benefits arising from the same implementation poses conceptual and methodological problems (see Petter et al., 2008). For example, it is not easy to attribute changes in company level metrics such as increased sales or reduced costs, directly to a specific IS implementation. Considering the Microfinance sector, which is the focus of the present paper, it is often the case that the implementation of new IS

246

Chapter Twelve

solutions takes place simultaneously with other organizational initiatives, which may include business process reengineering, training, skills upgrading, marketing and brand building. It may be observed that such initiatives often share common goals such as growth in the top line and the bottom line. Microfinance Institutions (MFI), as social enterprises, may also like to measure the impact aspect on other operations, in addition to internal, company–specific measures of performance. In summary, the first problem in using an ROI framework is the difficulty in identifying, quantifying and measuring specific benefits from IS implementations and other externalities. Another possible approach to evaluate the success of an IS implementation is what we term as the functional approach, which appears to be considered as complete in itself. For example, the World Bank has published a detailed handbook of MIS for MFIs, which serves as a compendium of what functionality should be supported by a good MIS for MFIs (Mainhart, 2012). Thus, if a particular product meets most of the specified functionality, and is implemented in a checklist mode, it is possible to evaluate the success of the IS implementation, within such a framework. Indeed, we believe that most organizations indicate a strong “functionality bias”, in the selection and evaluation of IS solutions. We also believe that such an approach has inherent limitations, and is not adequate in itself. The key problem with the functionality approach is that it does not address questions such as how useful was the IS solution in solving specific problems, or delivering additional benefits. For example, the management of an MFI would like to see that the MIS implementation leads to improvements in the productivity of field officers, in terms of the number of loans processed and the number of transactions processed in unit time, both of which are important operational parameters. At a higher level, the management may expect that a “good” MIS implementation should lead to reduced operating costs, which is, of course, impacted by improved personnel productivity. There is, therefore, a case for identifying specific impacts of IS implementations, on the basis of identified impact factors. This approach not only provides a quantifiable framework for assessing the success of an IS implementation, but also provides management guidance in terms of where potential benefits are possible, and to what extent. Hence, it may be observed that the functional approach, while necessary as part of an IS implementation strategy, is not sufficient in being able to provide adequate inputs to the management. Another approach focuses on “input” related factors, such as training, infrastructure, process complexity and functional adequacy, in evaluating

Peculiarities of the Microfinance Sector

247

the success of IS implementations, and a body of research has followed a similar approach (see Sabherwal et al., 2006; Sharma and Yetton, 2007, 2011). While this approach provides guidance on what an organization should do in order to ensure “successful” IS implementations, the focus is on avoidance of failure, and not on the impact which such implementations ought to deliver. Research in the area of success in IS implementations, since the 1980s, has attempted to overcome the limitations of other approaches, as described briefly above. DeLone and McLean (D & M) authored a seminal paper in 1992, proposing a conceptual model for assessing the success of IS implementations, using six factors and a proposed causal relationship between them. The D & M model has been widely applied to different contexts by researchers and practitioners, in the intervening period, and has been modified substantially as a result of the body of research which has resulted since the model was originally proposed. The authors reviewed the model in 2003, and this is referred to as the modified D & M model (DeLone and McLean, 2003). The present paper draws on the modified D & M model as a reference in discussing IS implementations in the MFI context. This chapter is structured as follows. First, we establish the relevance of the study in the Background section by outlining MIS needs in an MFI, and a brief overview of measuring MIS’s impact using the modified Delone and Mclean model. Next, we outline in detail the process of decomposing higher order success factors of MIS implementations into measurable metrics. Following this, we present a case study of BSFL, where we analyze company data to illustrate some of the identified metrics using time-series analysis. Finally, we offer recommendations, and draw conclusions, highlighting the limitations of the study.

Background The Microfinance sector has witnessed high growth rates across the globe in recent years, and is estimated to benefit nearly 500 million borrowers, out of a potential 3 billion according to Consultative Group to Assist the Poor (CGAP)’ estimates (CGAP, 2013). Given the nature of the business, which mainly deals with different aspects of data and information processing, IS play a key role in impacting the operations, profitability, growth paths and competitiveness of MFIs. When this primacy of IS is conflated with the rapid changes and innovations in the information ecosystem, it becomes essential for operating and top management to gain a critical appreciation of the level of success which

Chapter Twelve

248

their enterprise IS implementations have achieved, the potential available for improvement, in order to thereby be in a position to develop strategic roadmaps for their organizations. We emphasize that IS solutions critically impact the success of MFIs, and must be viewed as a key component of organizational strategy, rather than as a support function, which is often the perceived role of IS. As briefly discussed in the earlier section, assessing the success of IS implementation itself poses conceptual and methodological problems, mainly concerned with the identification, quantification and valuation of benefits. In an interview with the Network Magazine, Sanjay Sharma, IT Advisor to IDBI Bank, India mentioned a set of expectations and goals for the Bank’s IS implementation. These include the ability to reach customers through multiple channels, rapid deployment, relatively low level of training requirements, ability to scale up rapidly, CRM, ability to support new product introductions, generate growth and reduce costs. Indeed, one may say this is a very ambitious set of expectations, but Sharma views IS as a “business driver and not as a mere business enabler” (Sharma, 2012). While this example is cited from the banking industry, we believe the same logic is equally valid for the MFI segment. This also illustrates that the expectations of IS from a management perspective, are indeed complex, multi-dimensional and interdependent. For instance, Ali Ahmad, CIO of the First MicroFinanceBank Ltd, provides a very specific set of success factors for MIS implementations which illustrates this complexity (Ahmad, 2012). The focus is on information quality, individual impact, organizational impact and flexibility: a) b) c) d) e) f) g) h) i)

Easy access to accurate and up to date information. Detailed information to support tracking of customers Quicker, more accurate transaction processing User required formats of reports Deliver interactive, accessible and transparent services Improve efficiency and productivity of staff Support new product introductions Flexibility to structure products Flexibility to integrate with other applications

These success factors closely correspond to the factors identified in the D & M model. The six factors identified in the original D & M formulation are a) System Quality b) Information Quality c) Use d) User Satisfaction e) Individual Impact and f) Organizational Impact. The model was reviewed by the authors in 2003, based on several studies which had

Peculiarities of the Microfinance Sector

249

used the model, and the feedback received by various researchers (see DeLone and McLean, 2003). The D & M formulations were modified to include Service Quality and Net Benefit, while Individual Impact and Organizational Impact were subsumed under the Net Benefits factor. The interpretation of the Use factor was considerably broadened to include usage and various aspects of usage. A detailed study by Petter et al. (2008) reviewed some 90 studies which had used the modified D & M model, and attempted to identify the significance level of each of the factors in determining the overall success, the degree to which these factors are correlated, and to some extent the level of generalization which the model had achieved across a range of deployment scenarios.

Identification of Success Factors and Metrics for MFIs Our approach in this section is to identify a set of factors and metrics, which are relevant for the MFI sector, and which are based broadly on the IS success models discussed in earlier sections, and, in particular, the modified D & M model and the Sedera research model (Sedera et al., 2004). The D & M model is inherently complex, and is an attempt at abstracting key factors and causal relationships, on the basis of hundreds of studies, which span a wide range of domain areas and levels of sophistication. We believe that in the context of this paper, it would be far too ambitious to apply the D & M model in its entirety to the MFI context, an initiative which requires a much larger study, with statistically adequate time series and cross sectional data, and also a series of surveys, which would capture data from employees and customers. Similarly, the present study does not address the causal model which forms part of the D & M formulation. Our discussion is limited to a sub set of success factors, namely: a) b) c)

System Quality Information Quality Net Benefits

Our discussion and analysis of System Quality and Information Quality is based on a conceptual framework, which mainly relies on techniques of decomposition and quantification, in order to derive specific metrics which may be considered for monitoring the success levels related to these factors.

250

Chapter Twelve

1. System Quality System Quality is a complex and abstract entity, and Petter et al. (2008) have suggested that it may be decomposed into Ease of Use, Ease of Learning, User Requirements, System Features, System Accuracy, Flexibility, Availability, Sophistication, Integration and Customizability, as relevant components of system quality. However, from the perspective of the IS manager, even these lower level attributes are far too many to track, and they, in turn, require further decomposition, since they are not capable of being measured in any meaningful manner. A much more specific reference from Ali Ahmad cites accuracy, timeliness, flexibility, among system quality attributes. We propose that System Quality may be decomposed into Availability, Usability, Reliability and Flexibility, based on a survey of various definitions, including but not limited to the cited references. Thus, System Quality: ݂ሺ‫ܣ‬ǡ ܷǡ ܴǡ ‫ܨ‬ሻ Where, A = Availability, U = Usability, R = Reliability, F = Flexibility At the next level, our formulation for Usability is given below: Usability: ݂ሺ‫݈ܧ‬ǡ ‫ݑܧ‬ǡ ܵ‫ݑ‬ǡ ܲ‫ݎ‬ǡ ‫ݎݎܧ‬ሻ Where, El = Ease of Learning, Eu = Ease of Use, Su = Subjective Satisfaction, Pr = Productivity, Err = Error Rate Ease of Learning: Time taken to learn a defined set of tasks by a group with defined skill levels. Ease of Use: Time taken to perform defined set of tasks by a group with defined skill, training and experience levels. This is also a measure for individual productivity. Error Rate: Ψ݁‫ܽ݊݅ݏݎ݋ݎݎ‬ሾ݂݀݁݅݊݁݀‫ݐ݁ݏ݇ݏܽݐ‬ሿ‫ݕܾ݀݁݉ݎ݋݂ݎ݁݌‬ ‫݄ݐ݅ݓ݂݂ܽݐݏ‬ሾ݂݀݁݅݊݁݀‫ݏ݈݁ݒ݈݈݈݁݅݇ݏ‬ሿ It may be observed, that we have now performed the decomposition exercise to a level where we may easily identify relevant metrics. Thus, the usability factor may be measured at the time of training, in routine operations, and via employee surveys, in order to ensure that a) benchmarks are established for specified operations of the IS and b) usability metrics are used to drive improvements in employee productivity. Due to constraints of space, we have provided a set of proposed metrics in respect of several of the identified success factors, without much of a

Peculiarities of the Microfinance Sector

251

discussion. We also believe that these are fairly intuitive, and need to be adapted for use by specific organizations, so that there is an alignment with specific organizational goals. A detailed treatment of decomposition and quantification is available in Nyapati (2011). Availability: ݂ሺΨ‫݃݊݅݊݊ݑݎݏ݅݉݁ݐݏݕݏ݄݁ݐ݄݄ܿ݅ݓݎ݋݂݁݉݅ݐ݂݋‬ሻ and, ܴ݈ܾ݈݁݅ܽ݅݅‫ݕݐ‬ǣ݂ሺ‫ܨܤܶܯ‬ǡ ‫ܨܶܶܯ‬ǡ ‫ܴܶܶܯ‬ሻ Where, MTBF: Mean Time between Failures MTTF: Mean time to Failure MTTR: Mean Time to Repair We have provided the classical definition of availability, which considers that a system has only two states, namely, up or down. In practice, it is common to consider various quality of service (QoS) metrics, including performance metrics, since these will help in capturing the status of degraded systems, which however continue to be in the up state. Also, in view of the distributed nature of many IS implementations, which now include hand held devices with data connectivity, it becomes necessary to capture availability data at a disaggregated level, which includes central and remote IS deployment scenarios. The approach recommended is to inventory in detail the IS deployment footprint, identify QoS parameters, identify standardized methods for measuring QoS values, implement systems to capture QoS and failure / repair related data. This will help in tracking availability over time, as well as to set benchmarks and quantified goals at disaggregated levels.

2. Information Quality Some of the common characteristics of this factor include fitness for use, accuracy, timeliness and completeness. Here again, we propose a simple exercise of identifying suitable, measurable, components of these abstractions, as below: Timeliness: ݂ሺܶǡ ‫ݐܣ‬ǡ ܶ݅ሻ Where T = Time in between updates, At = Access time, Ti = Time to make information available after updates Accuracy: ݂ሺሾܰ‫ݏ݁݅ݎݐ݊݁݊݋݈݈݅݅݉ݎ݁݌ݏݎ݋ݎݎ݂݁݋ݎܾ݁݉ݑ‬ሿሻ

Chapter Twelve

252

The same approach may be used to identify specific metrics which represent identified success factors.

3. Net Benefits Net benefits is referred to by Petter et al. (2008) as ‘the extent to which IS are contributing to the success of individuals, groups, organizations, industries, and nations’ (p. 239). Typically an organization has access to data only within its own boundaries. Defining, capturing and analyzing organizational level data in itself is a daunting task for the management. Therefore, we limit our identification and discussion of metrics to those which can be captured and computed at the organizational level. Therefore, Net benefits = ݂ሺ‫ܫܫ‬ǡ ܱ‫ܫ‬ሻ Where II = Individual Impact, and OI = Organizational Impact Specific metrics relating to these two components are mentioned in Table 1 below. Success Factor

Components Availability Reliability

System Quality Usability

Information Quality

Timeliness Accuracy

Proposed Metrics % of time for which the system is running MTBF MTTF MTTR Ease of Learning: Time taken to learn a defined set of tasks by a group with defined skill set. Ease of Use: Time taken to perform defined set of tasks by a group with defined skill, training and experience levels. Subjective Satisfaction Error Rate: % of errors in a defined task set performed by staff with defined skill levels. Time in between updates Access time Time to make information available after updates Number of errors per million entries

Peculiarities of the Microfinance Sector

Individual Impact Net Benefits

Organizational Impact

253

Number of active customers per field agent Number of new businesses per field agent Average loan amount disbursed per field agent Average outstanding amount per field agent Employee-hours per Loan disbursement Employee-hours per transaction On-time repayment rate Performing Assets Return on Assets Upper Management to Field Agents ratio Number of days from loan application to sanction Number of days from loan sanction to disbursement Average time taken to generate reports Average time taken to generate receipts for customers

Table 1. Decomposition of Success Factors

Case Study: BSFL, an illustration In this section we attempt to use the metrics outlined in the previous section to illustrate how the formulations can be applied to assess the impact of an MIS implementation of a typical MFI.

1. Case Background The MFI chosen for the study was BSFL, a non-banking financial company, a part of the BASIX Limited group of companies based in India. BASIX is one of the key players and pioneers in the MFI space in India. Data as of 2011 indicates a customer base of nearly 570,520 customers and a gross loan portfolio of around US $57.4 million (MixMarket, 2012). They have also been the frontrunners in implementing MIS solutions to computerize and manage their MFI operations. Their current MIS implementation is Delphix, a mini-Enterprise Resource Planning (ERP) based MIS solution implemented in nearly 50 locations across India. In addition to implementing IS solutions for their own operations, BASIX Consulting, another group company, has also implemented IS solutions for over 100 clients in the developing world. Thus, the group has considerable expertise and experience in the implementation of IS for MFIs.

254

Chapter Twelve

2. Data Source Access to the research site was granted to us by the top management of BSFL. We could assess only the Individual and Organizational Impact on the MFI as data corresponding to only these two factors was available. To understand the individual and organizational level impact of the MIS in depth, we outlined Individual and Organizational Impact indicators as outlined in the previous section and used internal reports from ranging from the year 2004-05 to 2009-10 to generate the related metrics. In addition to the data set generated from company data, our analysis was also informed by our interactions with BFSL management team over a period of one month. Using the operational data set, we undertook simple time series analysis to measure the success metrics.

3. Findings 3.1 Individual Impact Evident from Figure 1 is the pattern of decline in the number of active customers per field agent, and the number of livelihoods supported per field agent over the analysis period. The Compound Annual Growth Rate (CAGR) for these metrics is -4.37% and -9.77% respectively over the period 2004-05 to 2010-11. Similarly, the number of new businesses, which includes ancillary products such as life and health insurance, and agriculture and business development services has also seen a major decline post 2008-09 after a huge surge from 2004-05 to 2007-08, with an overall CAGR of 13.03%. These metrics may potentially evoke managerial action to look into the underlying reasons of productivity concerns. It may be noted that these metrics are at an aggregated level, and more granular data will be needed in order to make any useful inferences in respect of the IS impact. Figure 2 depicts the average loan amount disbursed per field agent and the average outstanding amount per field agent. We see that the average loan amount disbursed has steadily, but not significantly increased over the reporting period, whereas the outstanding amount has decreased at a rate of -2.24% over the same period. There are two other metrics which we have included in our gamut of success factors, but for which we could not procure the data for analysis. These are: employee-hours per loan disbursement and per transaction. These are also important indicators of individual impact. With a successful

Peculiarities of the Microfinance Sector

255

MIS, one would expect a greater number of loan disbursements and transactions per employee hour. 1000 800 600 400 200 0

Number of active customers per field agent Number of new businesses per field agent

Figure. 1 Individual Impact (1)

in USD

80000 60000 40000 20000 0

Average loan amount disbursed per field agent Average outstanding amount per field agent

Figure. 2 Individual Impact (2)

3.2 Organizational Impact Figure 3 depicts three of the organizational impact indicators. We see that both the on-time repayment rate (ORR) with an average of 97.82% and the percentage of Performing Assets (PA) with an average of 98.18% over the reporting period are at the higher end of the spectrum; this is considered to be very good for an MFI. The return on assets has been in the range of 0.6% to 1.93%, steadily increasing over the reporting period with a CAGR of around 17.9%. Moving on to organizational impact indicators pertaining to human resources, we see that the staff strength of the direct staff (i.e. the upper management) has grown at a CAGR of 28.31%, whereas that of the indirect staff (constituting the field level staff) has grown at a CAGR of 50.69%. The field staff to upper-management ratio therefore has steadily

Chapter Twelve

256

increased from 1.44 in 2004-05 to 3.78 in 2009-10, growing at a CAGR of 17.45%. These metrics are depicted graphically in Figure 4. 100.00% 80.00% 60.00%

On-time repayment rate

40.00%

Performing Assets

20.00% 0.00%

Return on Assets

Figure 3. Organizational Impact (1) *the curves for on-time repayment rate and Return on Assets overlap 3000 2500 2000 1500 1000 500 0

Upper Management Staff Field Agents

Figure 4. Organizational Impact (2)

In trying to detail and measure metrics to assess individual and organizational level impact, we undertook a quantitative analysis using the company provided data set. In doing so we found that there are a number of metrics which can be tracked to assess individual and organizational impact, some of which we have computed and illustrated. Set against a benchmark, these metrics can be used to assess the direction in which the MFI is headed, and subsequently, take appropriate managerial action based on those insights.

Peculiarities of the Microfinance Sector

257

Recommendations Based on our identification of a limited set of success related metrics, and analysis of company level data, we would like to recommend that the IS departments of MFIs make a commitment to assess their own implementations, on an ongoing basis, based on a certain set of identified metrics, and associated methodology of data collection. Such an a priori commitment is very important, since the initiative will require additional efforts and resources in terms of data collection and analysis, requiring management approval. Once the organization makes such a commitment, and identifies the metrics which are considered most important, the sources of data and method of recording, collection and analysis will need to be specified. Our attempt indicates that data, as readily available, is not likely to meet the needs of tracking identified metrics. Hence, it is important to identify additional data requirements, including data to be derived from employee and client surveys. In summary, our main recommendation to MFIs is that adopting a success model for IS implementations is a key initiative, which can help on a variety of dimensions, including reducing costs, improving customer service and customer acquisition, improving the quality of loans, and reducing response times. Such a model will also help in benchmarking the organization on relevant metrics, in measuring improvements over time, and in identifying new initiatives which will yield the best outcomes. Given the inherent complexity of dealing with very many “success factors”, it is also recommended that a widely used model, such as the modified D & M model, may be considered as a reference for developing specific success models for individual organizations. The development and utilization of an enterprise specific models of IS success also requires participation from various internal and external stakeholders, and it may be emphasized that a generalized model, such as the D & M model, requires extensive customization before it can be effectively deployed, and this also will require teamwork from internal stakeholders.

Conclusions Our study indicates that using a well-known model, such as the D & M model of IS success, requires a fairly major effort from the organization. Such effort includes a) Identification of the model to be used as reference b) Identification of specific success factors and metrics and c) Identification of methods of data capture and analysis. In general, we

258

Chapter Twelve

expect that a given organization is unlikely to have all the data required for carrying out an assessment of IS success. Many of the success factors require data to be generated from employee and client surveys, and this is certainly an important part of the data which is not likely to be available. Considering the organizational impact related factors, our view is that it is conceptually difficult to attribute aggregate, company level metrics, to the level of success achieved in the IS implementation. To be sure, we acknowledge that the IS implementation will impact aggregate metrics, but we also believe that the effect of other factors impacting the same metrics, will need to be isolated or identified in some manner. This is a real challenge, and one which is common to any exercise in using a model, such as the D & M model.

Limitations and Future Work Our present study has been mainly at a conceptual level, and is an attempt to develop a success model for IS implementations in the MFI sector, drawing on the modified D & M model as a reference. The actual data used in the study is from one MFI, and is related to only a few of the D & M factors and sub factors, and this is clearly a major limitation of the study. Even within the data set identified for the study, it was possible to gain access to only a part of the data, and for a limited period. These limitations can be addressed by collecting data for a few years, and also collecting data from a sample group of MFIs, which have implemented IS solutions, thereby expanding the scope of the data available for analysis. The granularity of the available data is another significant limitation, since the choice of success factors and metrics had to be based on the data as available. While assessing the organizational impact factors, it becomes clear that relating such aggregate data to the success of IS implementations has to be done with care, isolating the effects of other factors which impact the same organizational factors. We would like to highlight this is a challenge in applying the D & M model to any IS implementation. Several of the success factors identified in our paper require surveys of staff and clients, ideally conducted over a few years, with statistically significant sample sizes, and under conditions of anonymity. This was one of the limitations of the present study, since it was not possible to obtain data in this manner. Another limitation which we faced in the study relates to the dramatic decline in the MFI business in India in the past few years, which also affected the MFI which we used for the study. Thus, BSFL witnessed a

Peculiarities of the Microfinance Sector

259

reduction of over 90% in asset size in the past two years, a development which has to be considered as exceptional, and unprecedented. Given this sharp downturn, we wanted to exclude this period entirely from the study, and sought information from a past period. However, we would like to suggest that future studies may like to examine the role of IS implementation in recovering from a crisis, and how they may contribute to the resilience of the organization as a whole.

Bibliography Ahmad, A. Management Information Systems for Microfinance http://www.bwtp.org/pdfs/arcm/5Ahmad.pdf (accessed on 16 November 2012). CGAP. Financial Institutions with a double bottom-line: Implications or the future of Microfinance http://www.cgap.org/sites/default/files/CGAP-Occasional-PaperFinancial-Institutions-with-a-Double-Bottom-Line-Implications-forthe-Future-of-Microfinance-Jul-2004.pdf (accessed 19 March 2013). DeLone, W. and McLean, E. (1992). “Information Systems Success: The Quest for the Dependent Variable”. Information Systems Research 3(1): 60-95. DeLone, W. and McLean, E. (2003) “The DeLone and McLean model of information systems success: a ten-year update.” Journal of Management Information Systems 19(4): 9–30. Mainhart, A. Management Information Systems for Microfinance: An Evaluation Framework, http://pdf.usaid.gov/pdf_docs/Pnacq075.pdf (accessed 1 December 2012). MixMarket, MFI Report: Bharatiya Samruddhi Finance Limited http://www.mixmarket.org/mfi/basix (accessed on 6 December 2012). Nyapati, K. (2011). Stakeholder Analysis of IT Applications for Microfinance. In A. Ashta (Ed.), Advanced Technologies for Microfinance: Solutions and Challenges (pp. 1-17). Hershey, PA: IGI Global. Petter, S., DeLone, W. and McLean, E. (2008). “Measuring information systems success: Models, dimensions, measures, and interrelationships”. European Journal of Information Systems 17:236–263. Sabherwal, R., Jeyaraj, J. and Chowa, C. (2006) Information System Success: Individual and Organizational Determinants. Management Science 52 (12): 1849-1864.

260

Chapter Twelve

Seddon, PB. (1997) “A respecification and extension of the DeLone and McLean model of IS success.” Information Systems Research 8(3): 240–253. Seddon, P., Graeser, V. and Wilcocks, LP. (2002). “Measuring organizational IS effectiveness: an overview and update of senior management perspectives.” The Data Base for Advances in Information Systems 33(2): 11–28. Sedera, D., Gable, G. and Chan, T. (2004). A factor and structural equation analysis of the enterprise systems success measurement model. In Proceedings of the Twenty-Fifth International Conference on Information Systems (Applegate L, Galliers R and Degross JI, Eds), pp. 449, Association for Information Systems, Washington, DC, USA. Sharma, R. and Yetton, P. (2007). “The Contingent Effects of Training, Technical Complexity and Task Interdependence on Successful Information Systems Implementation”. MIS Quarterly 131(2): 219239. Sharma, R. and Yetton, P. (2011) Top management support and IS implementation: Further support for the moderating role of task interdependence. European Journal of Information Systems 20(6): 703712. Sharma, S. IDBI Bank reinvents itself http://www.expresscomputeronline.com/20041011/casestudies04.shtml (accessed on 10 November 2012).

CHAPTER THIRTEEN MFI GROWTH PHASE: DIFFICULTIES IN THE MANAGEMENT INFORMATION SYSTEM MAWULI K. COUCHORO

Introduction Microfinance is seen as a solution to social problems (social and financial exclusion, poverty) and a tool to improve the quality of life for lower income families. Social innovation is meaningful only if it is adopted at the individual, local, regional and national level (Assogba, 2010). But innovation cannot spread without the help of institutional arrangements (laws, regulations, institutions, international agreements, division of powers in various scales, programs and other measures) that will overcome resistance to change (Lévesque, 2005). Social innovation refers to fundamental change, in not only institutions and structures, but also at the level of individual and collective behavior (Assogba, 2010).The diffusion of microfinance can be observed through the strong increase in demand and competition that pushes the microfinance sector to real growth. As a result, we are seeing changes in microfinance institutions especially in terms of human resources (staff training and/or recruitment of more qualified staff) and, from a technical point of view, the use of more efficient information systems that are more responsive to the objectives of microfinance. The commercialization of microfinance has increased competitive pressures on all actors in the microfinance sector. This has led MFIs to experience a growth phase that is critical to their performance and their survival. In general, all enterprises, whatever the sector, microfinance, engineering services, information technology, construction, and pharmaceutical have one thing in common: they need to be managed, planned, staơed, organized, monitored, controlled, and evaluated

262

Chapter Thirteen

(Marjolein et al., 2012). To this end, managers need to make fast decisions, allocate scarce resources efficiently, and have a clear focus. The problems faced by enterprises can be related to resource conflicts and throughput times (Marjolein et al., 2012; Maylor et al., 2006). Inadequate balancing of scarce resources often results in additional pressure on the organization, which leads to poor quality of information and longer lead times of projects (Elonen and Artto, 2003). Managers may become overwhelmed by the amount of information that is available for decision making, losing sight of relevant information or being unaware of inaccuracies. In general, poor information quality leads to poor decisionmaking (Engwall and Jerbrant, 2003). The use of Management Information Systems (MIS) is considered advantageous to managers because of the contribution regarding more timely decision-making. Thus, for the MFIs, the information system becomes a need, but at the same time, it is a high challenge. The need to improve treatment and understanding of MFI data is strong as stakeholders are becoming increasingly aware of the need for MFIs to manage large volumes of data. MFIs have to manage a large amount of data essential to their operations, from basic information on customers to detailed analysis of transactions and portfolio performance. This data must be stored, processed and, most importantly, used by management so they can make well-informed decisions. A good information system should act as a conduit through which raw data is transformed into useful information. It is a necessary tool for the proper management of a MFI. An MIS is very important to the success of a microfinance institution. Unfortunately, many MFIs, even if they have the financial means, have great difficulty with installation of MIS during their growth phase. The problem was so crucial in Togo that the Central Bank had taken measures intended to introduce a single software for all MFIs in the country. The reason was that, in many cases, the information system of MFIs was not in accordance with the legal framework of the country. However, this approach can only succeed if the difficulties faced by MFIs in terms of MIS are well known. What are the difficulties that MFIs will face in terms of MIS during their growth phase? The objective of this chapter is to highlight the difficulties and experiences of MIS usage in the growth phase of MFIs. To do this, we will initially present the need for and relevance of an effective MIS in the growth phase of MFIs. We then highlight the difficulties to implement a MIS for MFIs in the growth phase. The case of MFIs in Togo will be presented as well as some proposed solutions. Finally, we conclude.

MFI Growth Phase: Difficulties in the Management Information System 263

Analytic framework 1. Opportunity of MIS for MFIs Certainly, many MFIs consider a basic information system, such as a manual accounting system producing their reports with a delay of three months, sufficient for them. They then consider that it is pointless to invest enormous effort and financial resources to improve their information system. In fact, these tools are easier to develop and to update, but their usefulness is limited for institutions with a large volume loan portfolio. A manual system is prone to error, and inefficient for storing and retrieving data. It produces reports relatively late with considerable staff time and effort. This type of system does not easily conduct an analysis of trends. We are also aware that a system that correctly handles a low volume of activity may collapse under the effect of growing information needs as an MFI develops. Eventually, manual systems will lead to the accumulation of huge amounts of raw data and the MFI will need to move to spreadsheets. A portfolio management system that uses spreadsheets can in turn become unmanageable when the size of data increases. An institution that is ill prepared for rapid growth will jeopardize its financial health and quality. Indeed, gradually as MFIs grow more towards achieving their business goals, senior management realizes that they start to lose their ability to maintain direct contact with field activities. They find it difficult to manage their portfolio and financial transactions without better information. Manual systems and spreadsheets undoubtedly become ineffective when the institution becomes complex. Managers operating in competitive contemporary environments need comprehensive information in order to manage the important parts of the organization’s operations and thus achieve different strategic goals (Naranjo-Gil, 2009; Kaplan & Norton, 1996). MIS can provide managers with a variety of information. Choe (1996) identied the usefulness of information on four dimensions: scope, aggregation, integration and timeliness (Chenhall & Morris, 1986). Scope refers to the type and extension of MIS information in time and space. Narrow-scope information is derived from nancial information internal to the organization and with a historic orientation. Alternatively, broad-scope information includes external, non-nancial and future oriented information (Choe, 1996). Aggregation refers to the way data is aggregated over time, for example by departments or functions. Integration refers to the interaction and coordination of information among

264

Chapter Thirteen

different functions in the organization. Finally, timeliness refers to the frequency and speed of reporting (e.g. short or long run). Several authors have extended the four information characteristics to describe accounting systems in terms of MIS sophistication (Choe, 1996, 2004). MIS sophistication refers to a design that provides information that is (1) broad-scoped, (2) highly coordinated, (3) with appropriate reporting frequency, and (4) integrated across different organizational functions (Choe, 1996, 2004). Thus, we can see an increasing number of MFI leaders, having a broader scope of their activities, become increasingly aware of the need to improve their information systems. For many institutions today, methodological issues, staff training and even resource mobilization are fundamental to growth. However, it seems crucial to have an information system that enables them to determine in a timely manner, the accurate status of their portfolio. The reliability of this system can make the difference between success and failure of credit transactions, and therefore of an institution as well. An institution that implements a system capable of producing timely, accurate and complete information about its operations, especially in its credit portfolio, will be in a better position to manage its financial performance, as well as adapt to serving the needs of its customers. An information system with automated management then becomes the best solution.

2. Growth phase and MIS challenges The process of selecting a software is not part of the routine operations of a microfinance institution. Few institutions have the expertise needed to guide such a project, which makes the choice of MIS relatively difficult (NEXUS, 2000). It may also happen that employees have the skills, but do not have enough time to devote to developing the information system. Advice from an external consultant can help avoiding the major pitfalls of the process. This advantage can save the MFI more than it pays the consultant. Drawing lessons from the experience of other MFIs with the consultant, an MFI can save time and money. In terms of implementation of the automation process, MFIs face mainly the following difficulties (Waterfield and Ramsing, 1998). Firstly, for accurately identifying the information needs, executives, field staff, members of the board of directors and managers of information systems are rarely aware of all the needs of their institution. Even though they know it is necessary to monitor some key indicators these are not always defined precisely enough. Also, they can ignore the existence of other

MFI Growth Phase: Difficulties in the Management Information System 265

indicators that should also be monitored. Furthermore, systems are often constructed in an arbitrary and piecemeal manner without the needs having been fully assessed. Secondly, good communication between senior managers and specialized MIS staff must be ensured. In general, senior managers and staff of information systems within a financial institution do not speak the same language. The existing workload of employees and the tendency to compartmentalize operations exacerbate this problem. Even if they show the best intentions in the world, employees may misinterpret the wishes of the leadership. Thirdly, there must be a realistic view of what information technology can offer. In a world dominated by computers, users are often surprised not to have the desired information in a timely manner.

Togo case study methodology To the question: what are the difficulties faced by MFIs during their growth phase in term of MIS in Togo, the answer is structured around a methodology based on four key points which held our attention in the questionnaire that guided us in the interview. The four key points are: identification of needs of the MFI; involving the Board of Directors; the dilemma in choosing the MIS; and the satisfaction with MIS.

1. Identification of needs of the MFI Needs analysis is a fundamental step for the success of the MIS. It is therefore necessary for MFIs adopting a MIS to understand comprehensively the needs of the institution. It is only after this that the institution can make an initial assessment of the various options, including the improvement and linking of existing systems and the purchase of "turn-key" systems. For this reason, it is recommended that the following steps be taken: establishment of a working group; list the description of needs; determine what is feasible; evaluate the possibilities; and present the conclusions of the working group. The question would then be whether the MFI has difficulties in identifying its needs.

2. The involvement of all actors A team must ensure that all actors are represented and that project objectives are communicated to staff. The choice of members of the work team should include operations staff, head office staff and representation from the Board of Directors.

266

Chapter Thirteen

3. The choice of MIS: dilemma between an internal system and software purchase It is one thing to have an automated information system; another thing to make a good choice at the time of adoption. The choice of an automated information system for MFIs to meet their business growth is not an easy task. This choice is sometimes fraught with great difficulties, and a wrong solution can lead to worse situations than a manual system. Stories of failure and frustration abound, and in many cases involve computerized systems that never work quite properly or tend to fail just when the institution has become accustomed to rely upon it. The first difficulty arises from the dilemma of developing a customized system or the purchase of existing software (Laydeker, 2002; NEXUS, 2000). MFIs also have to choose between putting in place a sophisticated MIS that already anticipates the needs of growth, or set up the MIS that responds to growth needs over time.

4. MIS satisfaction degree The MIS cannot solve all business or operational problems of an MFI. Some of these problems may be related to a lack of staff training and internal controls. User satisfaction is undoubtedly a significant factor in the success of an MIS (Rigaud, 1984). MFI management, knowing why it wants to change its MIS, must set specific and measurable objectives for the new system and thus can estimate the degree of satisfaction that comes from the MIS.

Presentation of MFI experience To understand the difficulties faced by MFIs, we chose three MFIs which together account for over 70% of the market share of the microfinance sector in Togo. In this section, we briefly present the microfinance sector in Togo, the MFIs selected for our study, and, finally, the results of the interviews.

1. Microfinance environment in Togo and selected MFIs The microfinance sector in Togo by 20121 had more than 1.2 million clients, with customer deposits of around €177 million. This represented 15% of total deposits in the country. The total microcredit portfolio was nearly €155 million in 2012. Already by 2006, microcredit represented

MFI Growth Phase: Diifficulties in thee Management IInformation Sy ystem 267

16% of tottal credit in the country. Figure 1 sshows the grrowth of microfinancce, particularlyy since 2005. Gross loan portfolio (million of euros)

Deposits ( m million of euross)

2012

2010

0 2005

0

2000

100 1995 2000 2005 2010 2012

100

1995

200

200

Figure 1: Groowth of microfinnance in Togo. Source: CASI--MEC-Togo

The 19900s mark the beginning b of th he march of a large numberr of MFIs towards com mmercializatiion (Couchorro, 2011). Thhe latter beg gins with FUCEC (Faaîtière des Uniités de Coopérratives d’Eparrgne et de Créédit), after USAID, itss main partnner, stopped its financiall assistance in 1994. FUCEC’s fiinancial self-suufficiency waas then less thaan 40%. The operating balance requuired of the innstitution an innovation i staarting with a complete restructuratiion characterizzed by the tak king of budgett measures. Th he results of these meeasures was a reduction of the numberr of FUCEC’’s service points, from m 152 in 19999 to 72 in 200 08. FUCEC, oon the other hand, h also saw the voluume of servicees reach a larg ge proportion of the populaation with a significantt improvemennt of its outstan nding debts. F From 1994 to 2012, the number of cclients increassed 750.5%, from f 46,521 tto 400,000. Itts savings mobilized iss €8.18 millionn in 1995 and d €91.60 millioon in 2011, an n increase of more thaan 1000%. As for the volu ume of creditt, it has multtiplied by 29.52 times over the same period, increeasing from €€2.43 million to t €71.75 million. Rissk portfolio iss from 13.1% % in 2001, reaaching 3.79% in 2007, thus meetingg the standardd (5%) recom mmended in thhe WAMU zone. Since 2007, the opperating resullts of FUCEC C exceeded €11.49 million, a sign of financial perrformance. Women and Associiations for Gain G both E Economic an nd Social (WAGES), ccreated in 19994 with the su upport of CAR RE Internation nal, is one of the first aafter FUCEC to acceleratee its institutionnalization thaat became effective in 1999. The nuumber of mem mbers has varried significan ntly, from 7445 in 1999 to 160,000 in i 2012. Savin ngs mobilizedd and credit grranted for 2012 were respectively more m than €7.60 million and €22.90 millio on. As for the risk porttfolio, it was reduced r by 13 3 points comppared to 2001, reaching

268

Chapter Thirteen

1.2% in 2004. In 2012, WAGES has achieved an operating profit of €0.38 million, against €46.747 in 2006. In terms of credit granted and savings mobilized, FUCEC (mostly) and WAGES (sometimes) are beyond some banks in Togo. FUCEC is not really that far away from some banks considered as the largest in the country. CECA (La Coopérative d'épargne et de crédit des artisans), originally created with an endowment fund of the GTZ (German cooperation), specializes in Togo in the promotion of financial services for artisans in particular, and the public informal sector in general. It now has more than 10,000 members, with a savings mobilization and a credit granted estimated respectively at €3.4 million and €3.6 million. The CECA is open to other professions as traders and micro-entrepreneurs. The goal remains the same: to support the most vulnerable populations. CECA, which is only present in Lomé, is now progressively expanding in the other regions of the country. The data show that FUCEC and WAGES began a development and achieved a maturity that gives them an opening in the market that was previously held by banks. This development required the establishment of an MIS that met the needs of the MFIs growth. For all these three MFIs, the first choice of MIS has not met their expectations, which led to a second MIS.

2. FUCEC Togo Regarding the choice between internal development of an automated system and the purchase of software, FUCEC say they faced great difficulty. For FUCEC, the first choice was the development of an internal system. This choice was made in response to the growth of the MFI and not prior to a growth strategy. Nevertheless, the experience was marked by failure, although the conventional procedures, including involvement at the Board level, had been followed. Regarding the identification of needs, FUCEC estimated difficulties at four on a scale of ten. This first failure led FUCEC to have recourse to a new MIS named SYSDESAF, as a means to manage the challenges of a growing institution. SYSDESAF is the choice of the six major networks in West Africa, and it is considered as a more efficient solution in terms of existing ones on the market at the time of implementation in the Network FUCECTOGO in November 2003. According to FUCEC managers, SYSDESAF offers a lot of flexibility, security and supports all transactions of MFIs. Reports allow proactive management of the entire loan portfolio in

MFI Growth Phase: Difficulties in the Management Information System 269

accordance with international standards. It has a database capable of managing a wide range of financial products from microfinance institutions. It takes into account the needs of MFIs, including best practices for administration of the loan portfolio, the financial accounting, management accounts and customer information. It facilitates a balance between business demands and the active responsibility of social balance. FUCEC is at the version 4 of the software. The SYSDESAF is suitable for extensive branch networks but is also suitable for small sites, with a single agency. The main benefits offered by the SYSDESAF include, firstly, increased productivity by automating all daily activities, reducing manual processes and the risk of human error. Secondly, there is an increased ability to control business through getting portfolio, productivity, late payments, and hundreds of other reports. Thirdly, there is cost reduction and better use of staff time, thanks to the use of a Windows interface integrating MS Office tools. Nevertheless, the SAFSYSDE has weaknesses that are real difficulties for FUCEC. The MIS cannot edit financial statements in accordance with regulatory requirements. FUCEC sees itself obliged to have recourse to another system called STRATEGO decision, which leads to additional costs. Up to version 4.3, SAFSYSDE does not permit overtime accounts, which requires accountants to spend the night of December 31 and January 1 to work, for the year-end closing. This is because the system does not allocate operating accounts once the year is closed. In addition, the SAFSYSDE had bugs whose corrections led to several versions (the 2, 3, 4, 4.2, and 4.3). However, other bugs remain. In addition, the system is heavy and costly for smaller SFD, because at the beginning it costs US $1,200 per computer, with US $400 annual support fees. Apart from the cost of the software, one also has to purchase the Microsoft SQL Server. Moreover, the system is composed of several integrated modules: customer management, accounts and credits, which sometimes leads to unexplained accounting discrepancies. In addition, the designer of the system is in Costa Rica and only speaks Spanish and English. The users are in francophone Africa and speak only French. This raises a problem of language and distance. The DID (Développement International Desjardins), who is a technical partner from Canada, serves as an intermediary between the French speaking users and the systems designers. Given that the circuit is long, with multiple instances of decisions, corrections take time. Despite all this, FUCEC estimated more than 70% satisfaction rate with current MIS.

270

Chapter Thirteen

3. CECA Regarding the choice between internal development of an automatic system and the purchase of software, CECA, like FUCEC, reports great difficulties. The management says the choice was made in response to the growth of the MFI and not prior to a growth strategy. Nevertheless, for its MIS, CECA is on its third software, although the conventional procedures have been properly followed. The institution managers said that they had not encountered too many problems during the phase of the identification needs. Problems in this regard are estimated at three on a scale of ten. It used INFOCOOPEC, purchased software, which could not identify all aspects of the institution. Indeed, INFOCOOPEC managed membership, savings, repayments and, partially, credit. In terms of credit, this software recorded the decision to give credit, to make the loan contracts, and consult individual credits in place. However, this system does not allow for the monitoring of credit by having the age wise balance analysis. In addition, the accounting was only partly taken into account by the software. In order to use software that can fully integrate accounting, CECA developed an internal system: CECABASE. This software enabled the institution to manage all operations of the institution, to have all the information in any field of activity when needed. Its flaw was its instability: for the information requested, the base could give two different versions if it was drawn on two different computers or on the same computer at different times. This instability led to the abandonment of CECABASE for the purchase of a new SIG: “PERFECT”. However, after the trial of this MIS, the designer and CECA did not agree on the cost of using the new MIS. CECA then turned to the MIS MICROFINA, which is an improved CECABASE, taking into account all the concerns of the users. MICROFINA provided the necessary information in real time. When a user wants data that was not originally intended by the terms of reference of the designer, it has the ability to insert new fields into the database. Some imperfections are encountered in use, and these require the intervention of a designer to work out. At this step, CECA estimated more than 80% satisfaction rate with the current MIS.

4. WAGES For WAGES, due to the non-availability of expertise to develop an internal system, the choice of buying software was automatic. WAGES also says that the choice of information system was made in response to

MFI Growth Phase: Difficulties in the Management Information System 271

the growth of the MFI and not prior to a growth strategy. The institution began to automate its management information system with version 4 of the software, PERFECT, which showed its limitations at an early stage although the recommended procedures were followed. Regarding the identification of needs, WAGES claims to have had very little difficulty— two on a scale of ten. However, this version of PERFECT has many errors and instabilities that are not acceptable for WAGES. In addition, the supplier does not have a systematic procedure for treatment of anomalies and the development of WAGES products. This version does not meet the overall needs of WAGES for credit products. Perfect software was then abandoned. That led to the acquisition of new software, ADBANKING. WAGES noticed some weaknesses of this software at the time of use. It then gave a report on the problems to the designer who, unfortunately, did not follow up. Wages remarked later that the same observations had also been made by previous users, to whom the designer had never responded. Thus the institution decided to return to the new version of PERFECT, version 6, for which WAGES’ satisfaction varies between 70 and 80%. Indeed, PERFECT has improved over time. The new version covers almost more than 75% of the functional needs of MFIs, according to MIS managers of WAGES. PERFECT has integrated technological innovation such as interconnection and biometrics.

Conclusion The need for MFIs in the growth phase to develop an automated system for information management is no longer in doubt. Indeed, as MFIs grow and begin to adopt a commercial approach, MFI managers note that they lose their ability to maintain direct contact with field activities. They realize that it is difficult to manage their portfolio and financial transactions without better information. We chose for this study, the case of three institutions that have over 70% of the market share of the microfinance sector in Togo: FUCEC, CECA and WAGES. The interviews conducted with the managers of these MFIs, and the head of MIS, were used to highlight the difficulties they encounter although they followed the traditional process recommended for the installation of an MIS. These three MFIs are now working with at least their third version of MIS software, a sign of real problems of automated processing of the information. This shows that the conventional processes recommended, although important, do not guarantee a system that meets the expectations.

Chapter Thirteen

272

We notice that each MFI has its own problems with the information system. This suggests the need to create a central resource to identify information system problems faced by MFIs, to share these with the designers, and also note the software solutions. This would help MFIs in their future choices. The fact that our example MFIs say they are at least 70% satisfied with the MIS they now have, indicates that several problems have been fixed, but further improvement is possible. Institutions

FUCEC

CECA

WAGES

Questions Choice made in response to the growth of the MFI or to a growth strategy Identification of needs was made All actors were involved The institution had dilemma between an internal system and software purchase First Information system choice Estimation of difficulties faced Number of automatic MIS used to date MIS satisfaction degree

Response to the growth of the MFI

Response to the growth of the MFI

Response to the growth of the MFI

Yes

Yes

Yes

Yes

Yes

Yes

Yes

No

The purchase of software

The purchase of software 2 on a scale of 10

3

3 on a scale of 10 3

More than 70%

More than 80%

More than 75%

Yes

The development of an internal system 4 on a scale of 10

3

MFI Growth Phase: Difficulties in the Management Information System 273

Description of some difficulties

* The MIS cannot edit financial statements in accordance with regulatory requirements * MIS does not permit overtime accounts * Regular bugs * Language problems between designer and user

* Could not identify all aspects of the institution

* The system has many errors and instabilities that are not acceptable

* The system does not allow for the monitoring of credit by having the age wise balance analysis

* The supplier does not have a systematic procedure for treatment of anomalies and the development of the institution products

* The accounting was only partly taken into account by the software. * Instability: for the information requested, the base could give two different versions if it was drawn on two different computers or on the same computer at different times

Table 1: Brief comparison of the three MFIs’ MIS.

Bibliography Assogba, Y. (2007). Innovation sociale et communauté, une relecture à partir des sociologues classique, Université du Quebec en Outaouais, ARUC-ISDC, série recherche, N° 5. Assogba, Y. (2010). Théorie systémique de l’action sociale et l’innovation sociale, Université du Quebec en Outaouais, ARUC-ISDC, série recherche, N° 31.

274

Chapter Thirteen

Chambon, J-L., David, A. and Devevey J-M. (1982). Les innovations sociales, Paris, Presses Universitaires de France, 128 p, Collection « Que sais-je ». Cloutier, J. (2003). Qu’est ce que l’innovation sociale ?, Cahier du CRISES Collection Études théoriques – no ET0314 Couchoro, M. (2011). “Transformation des relations banques/institutions de microcrédit (IMC) au Togo : Une lecture à partir des cadres théoriques de l’innovation” , Revue Economie et Solidarités, N° 1-2 Vol 41. Choe, J. M. (1996). “The relationships among performance of accounting information systems, influence factors and evolution level of information systems”. Journal of Management Information Systems, 215–239. —. (2004). “The consideration of cultural differences in the design of information systems”. Information & Management, 41, 669–684. Elonen, S., and Artto, K.A. (2003). “Problems in managing internal development projects in multi-project environments”. International Journal of Project Management 21 (6), 395–402. Engwall, M. and Jerbrant, A. (2003). “The resource allocation syndrome: the prime challenge of multi-project management?” International Journal of Project Management 21 (6), 403–409. Klein, J-L. and Harrisson, D. (2007). Innovation sociale : émergence et effets sur la transformation des sociétés, Québec, Presse de l’Université de Québec, 465 p. Kaplan, R. and Norton, D. (1996). The balanced scorecard. Harvard Business Press. Kaplan, R. & Norton, D. (2001). The strategy focused organization, Harvard Business Press. Laydeker, B. (2002). Le système d’information pour une institution de microfinance , GRET. Levesque B..(2005).Innovations et transformations sociales dans le développement économique et le développement social : approches théoriques et politiques publiques, Cahiers du CRISES - Collection Études théoriques – no ET0507. Caniëls, M.C.J. and Bakens, R.J.J.M. (2012). “The effects of Project Management Information Systems on decision making in a multi project environment”, International Journal of Project Management, 30 (2012) 162–175 Maylor, H., Brady, T., Cooke-Davies, T., and Hodgson, D., (2006). “From projectification to programmification”. International Journal of Project Management 24 (8), 663–674.

MFI Growth Phase: Difficulties in the Management Information System 275

Naranjo-Gil D. (2009). “Management information systems and strategic performances: The role of top team composition”, International Journal of Information Management, 29, 104–110 NEXUS, (2000). Management Information Systems : between salvation and frustration, SEEP network publication Waterfield C. & N. Ramsing (1998), SIG pour les institutions de microfinance – Guide pratique, CGAP, Washington

Notes 1

Data from CAS-IMEC (Cellule d’Appui et de Soutien aux Institutions et Mutuelles d’Epargne et de Crédit).

CHAPTER FOURTEEN STATUS OF INFORMATION SYSTEMS IN MICROFINANCE INSTITUTIONS AND OVER-INDEBTEDNESS OF CLIENTS IN DEMOCRATIC REPUBLIC OF CONGO1 JACQUES BONGOLOMBA ISOKETSU, GAURAV SINHA, SUNDER ANNAMRAJU, RAKESH SUD AND AISHWARYA SRINIVASAN

Background and Introduction Microfinance institutions (MFIs) in the Democratic Republic of Congo (DRC) came up as a post-conflict measure to alleviate the perils and consequences of over a decade of civil war. Today, microfinance in the DRC has emerged as an important tool to address the unmet needs of a large population engaged in the informal sector (data relating to this population is often not captured by GDP statistics). In view of this, the microfinance sector has received continued and steady support from the Congolese government. This has assisted in the development of a large number of small and medium enterprises which remain the source of livelihoods of many families in DRC. Understanding the importance of this contribution, the government has established a ministry to promote microfinance as part of its plans for supporting and speeding up the postconflict reconstruction process. Moreover, the Central Bank of DRC has drafted a new regulation on microfinance, a new accounting framework and a consumer protection regulation (CGAP & MIX, 2011). These interventions have positively influenced the microfinance sector in the country. Yet, the sector is facing several challenges because of the postconflict socioeconomic situation in the country (refer to Box 1 and Table 1).

Status of Information Systems in Microfinance Institutions

277

DRC is a poor performer on many socioeconomic indicators. x It is the fourth most populous nation in Africa and the 19th most populous nation in the world. x It has a widely dispersed population of nearly 70 million.2 x It is home to around 2.3 million displaced persons and refugees living in the country. More than 0.3 million citizens of DRC are living in refugee camps outside the country (UN estimates). x Per capita income and human development indicators in DRC remain among the lowest in Africa. It ranks the last among 187 countries in the human development index (HDR, 2011). x DRC is also the poorest country in the world and significantly poorer than the other five countries where people earn less than a dollar per day (see Table 1 below).

Box 1: DRC at a glance 2011

Country Name

For the year GNI per capita, Atlas method

Per day

GNI per capita, PPP

Year end

GNI per capita, Atlas method

GNI per capita, PPP

Population (millions)

DRC

190

350

0.52

0.96

67.8

Liberia

240

520

0.66

1.42

4.13

Burundi

250

610

0.68

1.67

8.58

Sierra Leone

340

850

0.93

2.33

6.0

Malawi

340

870

0.93

2.38

15.4

Niger 6 reporting less than $1 a day 23 reporting less than $2 a day Sub-Saharan Africa (all income levels) World

360

720

0.99

1.97

16.1

246

519

0.68

1.42

117.9

435

1,018

1.19

2.79

454.1

1,266

2,251

3.47

6.17

875.56

9,491

11,574

26

31.71

6,973.7

35,986

35,094

98.59

96.15

1,245.2

OECD members

Table 1: Poorest nations in Africa and world. Source: Based on World Bank Quick query data downloaded on Sept 21, 2012 (unless otherwise mentioned, all figures are in current international $)

278

Chapter Fourteen

Non-availability of information is one of the major challenges that the country is currently facing. This is because of the loss of data and records destroyed during the conflict (IMF, 2007). In addition, lack of standardized reporting by financial institutions to the Central Bank of Congo (BCC) and the absence of information sharing between the financial institutions has made the situation worse (IMF, 2007). This is further exacerbated by the lack of adequate information and communication technology (ICT) infrastructure, especially broadband internet required for putting a good information system in place in DRC. Table 2 below provides a comparison of the telecommunication infrastructure index and its components comparing DRC with the top performing country in Africa (Seychelles in East Africa) and a European country (Liechtenstein). The data presented in Table 2 indicates that hardly any physical infrastructure is available in the country for recommending any type of online software solution or cloudbased computing. Infrastructure Component Estimated Internet users per 100 inhabitants Fixed phone lines per 100 inhabitants Mobile subscribers per 100 inhabitants Fixed Internet subscription per 100 inhabitants Fixed Broadband per 100 inhabitants Index Value

DRC 0.72

Seychelles 41.00

Liechtenstein 80.00

0.06

25.48

54.40

17.21

135.91

98.52

0.11

6.60

47.35

0.01

7.26

63.83

0.0183

0.4037

1.0000

Table 2: Infrastructure components in DRC, Seychelles and Liechtenstein, 2012: Source: United Nations E-Government Survey, 2012 Thus, the lack of both information and ICT infrastructure results in a widespread information asymmetry across the country. Even MFIs in DRC are not unaffected by this information asymmetry. The major problem for MFIs in DRC is poor access to customer information (Gilman et al., 2013) combined with the low quality of such information available (IMF, 2007; Schwarz, 2011). DRC scores zero on a scale of zero (low) to six (high) on measures of scope, access, and quality of credit information availability (IMF, 2007).

Status of Information Systems in Microfinance Institutions

279

On the one hand, the MFIs in DRC rely on constant contact between customers and loan officers in charge of the customer’s portfolio to take a decision on offerings in future (Schwarz, 2011). Furthermore, not all MFIs in DRC have implemented technology-enabled information systems. Some resort to manual operations. Among the MFIs who have implemented technological solutions, some organizations in DRC maintain secrecy of their information systems under the guise of “bank secrecy”. However, some information systems are subject to specific regulations, allowing users to access them through the monetary authority. The net result is that the reliability of data with MFIs—in terms of capture, storage and retrieval—is in question. On the other hand, the issue facing the customers of MFIs in DRC is their ignorance of the credibility and legitimacy of the MFIs and their operating processes (Schwarz, 2011a). The target population suffers from the twin handicaps of a low economic base and the lack of adequate knowledge of microfinance operations, particularly those related to the charging of interest on microcredit. This increases the risk of economic vulnerability, over-indebtedness and results in the marginalization of clients. This limited access to information is often the result of a lack of public policy and systems. Clearly, the discussion above strongly indicates the existence of information asymmetry. But experience across the world suggests that establishing an efficient and well managed information system in MFIs plays a crucial role in addressing several emerging challenges, some of which are mentioned here (Waterfield and Ramsing, 1998; CGAP, 2005; IFMR Research; IFAD, 2006; Nanyonga, 2008; African Union, 2009; Lascelles and Mendelson, 2012). a. Over-indebtedness among microfinance customers b. Transparency in operations c. Standardized reporting system for donors, financiers, government, regulators and other stakeholders d. Reducing costs & risks, scaling up and integration with financial ecosystem e. Dealing with competitive pressures Given this background, we suggest that an efficient information system in the microfinance sector might be a solution to the prevailing problems of information asymmetry that cause clients of MFIs in DRC to become over-indebted. However, for building an efficient information system, it is important to develop an understanding of the existing system vis-à-vis its

280

Chapter Fourteen

significance to the MFIs as well as their clients in DRC. Therefore, this research is an attempt to examine existing information systems, their relevance to the MFIs and their impact on the poor clientele. It also suggests a way forward to prevent over-indebtedness among microfinance customers in DRC resulting from lack of an efficient and up-to-date information system. The chapter is divided into two main parts. We first present three subsystems that make up information systems and examine their relevance to MFIs in Kinshasa, DRC, looking at information flows and constructs as a means of understanding the information needs of MFIs, and look at the technology in use to collect and process data. We then present how the lack of information flow, coupled with a mismatch between the needs of the borrowers and the products offered by MFIs in DRC contribute to poor customers in Kinshasa borrowing merely for sustaining themselves rather than economic activity and other associated activities focused on alleviation of their debt situation. The perspectives presented in our paper contribute to an understanding of the information sought by multiple stakeholders on microfinance activities. The paper also details and presents the “one-way” nature of the information flow where the customer is largely unaware of the legitimacy and the practices of MFIs, while the MFIs are continually correcting information gathered from, and about the customer. The “one-way” flow, in a different context, keeps MFIs on the blind side since they do not get to know the full profile of the borrowers and the current state of their indebtedness. Having identified the information needs and the flows required, we present little steps that can be taken today, using systems exploiting currently available technologies to pave the way for a greater integration of data and enabling the flow of information to address the situation of over-indebtedness among the poor in Kinshasa.

Methodology of data collection and analysis 1. Data Sources The data for this study was collected from three sources. The data collection process provides information on technology as well as variables such regulation in microfinance. The first dataset was based on a survey hosted by CGAP and UE/ACP on microfinance in 2008. The survey covered 193 countries across the world with an aim to collate information on usage of technologies by MFIs (CGAP, 2005/2008). This dataset helped us in statistical evaluation and effectiveness of new technologies.

Status of Information Systems in Microfinance Institutions

281

The second dataset was collected from a survey of 78 MFIs in Kinshasa in 2008. The third set of data was collected through research conducted in two different settings over a period of approximately 9 months, one with the MFIs and the other with diverse sets of existing clients, former clients, small and medium enterprises etc. Face-to-face, in-depth interviews were administered for this purpose. We included questions on the standard of information systems and technologies used by CGAP-2004 and 2008. The survey covered members, personnel and staff of MFIs operating in poor neighborhoods of Kinshasa. The questions concerned computers, technologies, website, internet, intranet and maintenance. These indicators covered the profile of MFIs, numbers of computers and management technologies, system of payment, investment in new technology and model of published information to stakeholders. Most of these characteristics helped us to measure the status of information systems between cooperative agencies, microfinance banks and informal financial organizations. In addition, we studied the process of access to credit for all 78 MFIs. The aim of this was to understand the level of the financial knowledge of clients. This helped us understand if availability of information can protect them against risks of credit such as the risks of defaults, financial exclusion and over-indebtedness.

2. Limitations The study limits itself to the study of information systems within MFIs operating in areas of urban activity in Kinshasa. The analysis does not include the governance of information systems among MFIs in DRC. This would involve further research on the prioritization of information and studying the inter-organizational relationships in this sector.

3. Description of MFIs in DRC In the DRC, MFIs are legal entities whose usual occupation is money trading (Central Bank of DRC, 2005). These organizations derive part of their revenues from clients and the balance is funded by public and private institutional investors. In this context, MFIs have common economic and social features. Their main mission is to provide microcredit, savings and micro-insurance to client populations excluded from the traditional financial system. They serve a dual purpose: social and economic, and need to remain economically sustainable while addressing the issue of impoverishment of the masses living in slums.

282

Chapter Fo ourteen

Our fieldd study combiined with a litterature review w shows that the MFIs are not evennly distributedd throughout the Congolesse territory (seee Fig.1). In 2010, thee Central Bankk of Congo gaave legal statuus to 108 orgaanizations as MFIs.

ntral Bank Figure 1: Disstribution of MFIs in provincees of DRC, 20110. Source: Cen of Kinshasa, 22010, and our studies s

Kinshasaa is home too 37 of thesee organizationns, followed by postconflict proovinces (Norrth and Soutth Kivu). Evven in Kinsh hasa, the distribution of MFIs is uneven with a highher concentrration in municipalitiies with a "ressidential” statu us. For exampple, around 44 4% of the MFIs operaate in the munnicipality of Gombe follow wed by Ngaliema and N’Djili (9% %) and Limette (8%). There are only 55% of MFIs in other provinces. T Their presencce in the neig ghborhoods off Kinshasa an nd in the provinces iss strategic in the sense thaat the aim of these MFIs is i also to develop reliigious vocatioons with the setting s up of churches, asssociations and non-ggovernmental organization ns. Their sstrategic mission is evangelizatiion and the reeduction of po overty in deprrived neighborrhoods of Congo, whiich helps to achieve a a greater number of practitioneers of the Christian reeligion amongg the slum dwellers. The social, econo omic and politico-religgious strateggy, along witth missions to develop Christian churches in the region, has h been work king due to ann insufficient supply of basic amenitties to the pooor.

Status of Information Systems in Microfinance Institutions

283

MFIs have specific characteristics in terms of organization and structure with a typical division of labor and activities. Like any job, the profession of microfinance requires special skills. Loan officers and officials of various departments become operational, more productive and effective in the field once they learn and internalize the routines. The rules, procedures and practices implemented by MFIs since their establishment in Kinshasa, help them better manage their operations and risk.

Information systems and flows We present a framework of sub-systems that underpin the information systems followed by information flows and constructs, in order to review and evaluate the operations and performance of MFIs and identify problems related to information for MFIs in DRC.

1. Different types of information systems in MFIs This section discusses three different types of information systems in the MFIs. 1.1 Type: “Logistics” The logistics sub-system helps MFIs ensure the proper functioning of micro-finance activities. These information systems provide a formal process for capturing, storing, processing and communicating the information collected about potential customers in slums based on technological tools and processes that provide support for transactional processes and organizational decisions (Hajer and Kalika, 1998). This information may be stored on computer media using software and allow MFIs in Congo to achieve better performance in the recording, storage, analysis and transmission of information. However, this technological process is inaccessible to many MFIs in Congo because of their high costs, lack of qualified staff and poor maintenance of computer media. It is therefore difficult for most of them to communicate to a wider audience in the poor neighborhoods. 1.2 Type: “Management Support” This sub-system helps in the management of operations, decisionmaking and coordination of operations (Lhuilier, 2005). The systems provide the tools to make lending decisions and support the administration

Chapter Fourteen

284

and management functions of the MFIs. Management support systems are often based on a periodic evaluation of indicators of social and financial performance. They enable MFIs in Congo to make decisions based on the needs of customers and the objectives of the organization and report on the efficiency of microfinance operations. 1.3 Type: “Critical” or “Strategic” Strategic sub-systems are tools for policymaking. The strategy of an MFI should be to anticipate the market, to respond better and faster to hazards and to develop the most effective medium to achieve these goals. To do this, political leaders should initiate a process to identify new propoor programs, set priorities and collectively define the objectives of MFIs in DRC. This will lead to better information systems that enable strategic planning and policy implementation in MFIs in Congo. They would help managers formulate and address strategic issues better. Critical systems are built at various hierarchical levels with institutional investors involved in private and public reflections, guidance and analysis of strategic decisions. These systems are strategic as they affect the entire business of microfinance as well as the activities of individual MFIs. These systems also help to assess strategic challenges posed by decisions and in setting credit policies.

2. Information flows within MFIs Available information needs to be viewed from the point of view of the information architecture. A synoptic chart of information flows would detail the movement of information. Information flows are constructed based on the social and financial realities on the ground. The structure of information flow is not self-evident given the multiplicity of MFIs in Congo and the prevalent organizational culture. The internal business processes of MFIs are critical to enable each organization to share information with multiple stakeholders and determine the flow of information. These information flows are structured for ease of analysis into social information and financial information. MFIs however, must know how to capture and build social and financial information flows. 2.1 Information constructs Social information poses problems for MFIs to collect, store and process relevant data. Impact studies and data collection are costly. The

Status of Information Systems in Microfinance Institutions

285

MFIs are not focused on collecting all information on client population in poor neighborhoods. They look at the collection of information as a secondary issue not realizing its importance. This is usually due to a lack of training provided to staff in collecting these data. This lack of knowledge results in the implementation of diverse structures for information capture and its use. Our results show that MFIs are obliged to find staff with skills and abilities to adapt to the new business of microfinance as they are faced with difficulties in meeting the needs of client populations in poor neighborhoods of Kinshasa. They need loan officers who can better understand the specificities, challenges and goals, and send feedback that can help influence policy decisions. It is indeed critical and we are far from the target where the use of social media data and informational resources are optimal. Ultimately, the MFI’s effectiveness needs to be evaluated in relation to the requirements of business and customer satisfaction in the slums. (a) Social information flux Social information flows can be derived from stocks of information collected and stored at various points of time. Collecting data on customers or staff provides raw data and is the first step. This is captured by encoders, loan officers and other personnel assigned for this purpose. The objective is to build a readable database of the social services organization. Information to capture includes the age, origin, qualification, career developments and place of residence. Externally, customers’ surveys provide information on the identity, occupation, marital status, income and household size of the clients in poor neighborhoods. These are social stream data available to statisticians and analysts to measure the social performance of MFIs. Thus, the social performance of the information system is measured by the operating costs of reported processes, processing times and return on investment. (b) Accounting and financial flows The accounting and financial information flow is constructed from the financial transactions of clients. They, in turn, provide a measure of the effectiveness and efficiency of MFIs. Recording of regular financial inflows and outflows has forced MFIs to acquire the tools of management. Computerized systems enable MFIs to build different specific patterns of flow of credit to clients and allow loan officers to manage their client portfolio in a much better manner.

286

Chapter Fourteen

The extent of information available and the flows vary according to the specific objectives of each MFI. From another point of view, the accounting and financial flows are stored and processed by the accounting software to produce summary tables in the “CGAP” accounting model (CGAP, 2003). The accounting model is in the form of a balance sheet and several accounting statements like funds flow, cash flow, and receipts and payments. These information flows enable finance professionals to calculate indicators of financial performance through the system of accounting and financial information. In this particular scheme, the tools are indispensable for taking decisions in the compressed time available. They allow adjustment of the credit policy and to produce financial mathematical models adapted to the profession of microfinance. 2.2 Characteristics of information flow Mapping information flow involves horizontal and vertical exchanges of information between various departments within an organization and between MFIs. This becomes difficult to implement because of the hybrid statutes, the complexity of the organizational structure of different MFIs, differences in technological processing and storage of information. This information from various sources requires sorting, collating, auditing for errors and distribution to internal and external users. These flows are diagrammed and codified. They can follow the life cycle of credit and the quality of the client population by place of residence in Kinshasa.

3. Problems related to information in DRC We have identified three types of problems relating to information. 3.1 Abundant and diverse information MFIs store various types of information. Emanating from various sources, this information is stored, re-structured and reported. They contain information on the credit market and the quality of the client population, the tax system, laws and regulations on MFIs. With various patterns and channels of information, these flows are published in several forms: charts, summary tables and annual reports. These input-output flows are essential for their schematization. However, faced with an increased flow of information, some MFIs are facing difficulties in accessing new technologies. It is difficult for them to capture all the available information. Consequently, these organizations employ manual

Status of Information Systems in Microfinance Institutions

287

means of schematization that are prone to errors, omissions and statistical manipulation. Ultimately, these limitations in collecting information are not known to the public. Once transparency becomes the norm we expect a genuine debate about the sincerity and credibility of published information, with doubts raised and queries answered. 3.2 Limited access to information sources Limited access to information is one of the recurring problems of MFIs. MFIs engage in a fierce competition to provide financial services to the poor. The clients form the testing ground for the theory of microfinance. There is opacity in the methods used by the MFIs, the approaches made and qualities of products offered to client populations. The information is kept confidential, with access granted only to members, partners and donors. Information exchange between organizations is almost nonexistent. A consequence of such practices is the rise in indebtedness of customers due to insufficient information on the practices of MFIs. The flow of information within the microfinance credit sector does not track customers’ issues. The customers can take separate loans from various organizations and associations of micro-credit without their biography and dealings with others being reflected in any organized credit reference system. In DRC, the post-conflict phase has attracted the interest of many private and public investors in MFIs. The motivations of these multiple donors are unknown to the public and client populations in poor neighborhoods. The absence of flow of information on savings and loans to customers are the best-kept secrets. The few publications found in the press mask the realities of the microfinance industry in DRC. For MFIs in DRC, this is a matter of survival, legitimacy and credibility with customers and donors. 3.3 Loss of data and insufficient controls The post-war DRC has serious consequences for businesses and organizations of microfinance. Most MFIs have lost their databases after armed conflicts and ethnic tensions. Many clients are missing from their usual areas of activity as well. There has been loss of several client files, computer tools and storage media. The consequences are such that it is difficult for credit unions to rebuild the missing databases, restore the flow of information and follow-up on debts. Corporations, unions and Congolese credit organizations have been adversely impacted and a large

288

Chapter Fourteen

proportion of them have fallen into bankruptcy. In this context, resuming the flow of information is a very difficult process to implement within the former credit cooperatives, now restructured as MFIs. Most of software systems have become technologically obsolete. Acquiring the tools for maintenance is costly for MFIs, especially for organizations which are not economically self-sustainable. The displacement of people during wartime has other consequences as well. Most off-the-shelf software products are not geared to rapidly shifting clients from one territory to another and providing such “shifts” in a meaningful manner in aggregate reporting.

Indebtedness and information systems MFIs provide financial and non-financial assistance to the customers living in poor neighborhoods in Kinshasa. There is a critical need for services to be made available to the poorest inhabitants. However, with the creation of many MFIs, a financial bubble has developed. Practices similar to those of commercial banks have contributed multiple loans to customers (from various MFIs) and the consequent over-indebtedness among clients. The burst has led many customers defaulting on debts.

1. Determinants of debts in slums of Kinshasa Underdeveloped areas have an insufficient supply of financial services and basic social services. The lack of social infrastructure and means of communication accentuates the precariousness of people. In order to meet their needs, people resort to debt. The sources of survival are cannibalized. In this analysis, we find that these sources are essentially debts to an MFI. For research purposes, we limit ourselves to the description of the determinants of debt of clients in poor neighborhoods. The survey covering 700 clients of MFIs in Kinshasa slums (EXACIMF, 2008/2009) shows the percentage of population over-indebted in the agricultural sector at 58.2%, agro-nutritional at 27.5%, small business at 5.7% and services at 8.6%. The causes are four-fold. Calendar system of debt settlement: The calendar system is a schedule of debt settlement based on the types of loans granted to customers. This system allows customers to repay weekly without an accumulation of debt. However, the results of surveys show that this does not help customers since the calendar system is not adapted to cycles of customers' products.

Status of Information Systems in Microfinance Institutions

289

Daily repayments are certainly a way to adapt their capabilities in relation to financial returns. Methods of recovery: The financial engineering of MFIs is identical to that of banks. The practices and methods of recovery of loan officers act as a determinant to the debts in Kinshasa’s slums. Loan officers manage the client portfolio. They are in constant contact with customers in the field and provide financial services to customers and micro-enterprises. They use methods ranging from the confiscation of savings to regular visits with families to recover money owed to the MFIs. Added to these factors is a pressure on clients to incur further debt. The confiscation of savings can affect the financial health of poor clients. This practice challenges the social vocation of MFIs and makes them akin to moneylenders and others who prey upon the poor for their own interests. Interest rates: In DRC, the interest rates on microloans are supposed to be lower than those of conventional banks. These rates allow MFIs to benefit from their investments in an industry where the risks are significant. MFIs have established different rates according to credit cycles. Mechanisms of interest application can take various forms: either they apply to balances due or they are deducted from, or added to the principal amount of the loan. The summation of this compound interest rate equalizes the interbank DRC Rates. In this context, poor clients are caught in a debt-trap. They cannot immediately repay the entire debt. As a result, they become insolvent and cannot claim amounts available in the new credit cycles. Education & training: Expensive training to organizations and customers is yet another determinant of debt. In DRC, many players offer MFIs training modules for microfinance. At the request of donors and banks, these courses are provided as modules. The course content is of a high scientific level. They contribute significantly to the training of managers and loan officers in social finance. However, these training modules are not adapted to local realities. Their content is not within the understanding of poor clients and neither does it contribute to an understanding of the needs of poor customers. One of the conditions required for MFIs to benefit from external credit lines is to include these training modules in their spending budgets. The same goes for a customer to obtain credit. The remuneration of the experts represents a fortune while the poor must submit to their offer of education to qualify for small loans.

Chapter Fourteen

290

Without prejudice to their action, these courses should be reviewed and incorporated into the local education programs.

2. Indebtedness and information system in Kinshasa Retrospective survey data on 700 credit beneficiaries in Kinshasa in 2008 are instructive. The survey had the objective of getting the maximum information on information systems and the quality of the client population. In poor neighborhoods, the client population consists of women engaged in petty trade and crafts, agricultural farmers, carpenters and shoemakers. Table 4 shows that out of 700 clients of 78 MFIs, who have outstanding loans, more than 50% are women. This is mainly because of the strategic focus of the MFIs targeting women as their clients. Other customer groups consist of "vulnerable beings". They contain refugees, people with HIV/AIDS, orphans and people with disabilities. Further information has been collected concerning the information system set up in neighborhoods and the means used to reach potential customers. It is found that poverty is multifaceted in neighborhoods. In high-income neighborhoods, residents are better trained. They master the tools better compared to those in poor neighborhoods. They can grasp the mechanisms of applicable interest rates on microloans better than those in other neighborhoods can. Characteristic of client debt Married women Widows Single women Divorced women Married men Single men Others (no debt) Total

Effective 225 12 135 29 56 68 175

Percentage 32.14 1.67 19.29 4.14 8.00 9.86 25.00

700

100

Table 3: Characteristic of indebtedness by sex in Kinshasa’s slums. Source: Survey, 2008/2009 Kinshasa Most of the MFIs operating in Kinshasa’s slums are cooperative agencies, which constitute about 47.44% of total sample (see Table 4). Many of these organizations changed their status during reconstruction

Status of Information Systems in Microfinance Institutions

291

phase. However, there are financial and management issues which are detrimental to this processes. Statute of MFI Association MF Banks NGOs Cooperatives agencies Total

Effective 21 4 16 37 78

Percentage 26.92 5.13 20.51 47.44 100

Table 4: Characteristics of 78 MFIs operating in Kinshasa’s slums. Source: Survey, 2008/2009 Table 5 indicates type of technology usage in MFI. This is used for exchange of information with clients and other MFIs in Kinshasa’s slums. In Table 5, mobile has a higher percentage (65.38%) and is an easy means of communicating information related to clients’ transactions. The MFIs use CD ROMs, mobiles and phones to exchange information amongst different branch offices in Kinshasa. Other means of information exchange include letters sent to the clients and branch offices. However, the underdeveloped state of postal services in the area poses challenges in communicating such transactions. Indicators Mobile Internet/intranet Fax /Interphones CD Rom Others (letters) Total

Effective 51 12 7 5 3 78

Percentage 65.38 15.38 8.97 6.41 3.85 100

Table 5: Management technologies used by MFIs to exchange information. Source: Survey on information systems, 2008 /2009 Banks establishing microfinance in these areas use modern, computerbased information media to capture client data. In addition, most of the MFIs use manual information systems, but our survey reveals that usage of manual information systems is higher among credit cooperatives. They use excel sheets and paper-based documents. We find usage of advanced technologies like software, intranet and excel sheets only in NGOs and

Chapter Fourteen

292

microfinance banks. This is because NGOs and banks can afford to recruit qualified professionals to manage these advanced technologies to capture client information. Indicators Manual System Worksheet Management software No particular systems Total

Effective 49 17 10 2 78

Percentage 62.83 21.79 12.82 2.56 100

Table 6: Technologies used in data management in MFIs in Kinshasa slums. Source: Survey on information, 2008/2009, and our analysis Based on the above discussion, we can summarize the status of information systems and availability of quality data for MFIs in DRC as follows: a. There is a distinct diversity of information systems in use, with microfinance banks and NGOs using relatively advanced systems and technologies, while the rest coping with a mix of paper-based and simple office automation software. b. Aggregation of data across MFIs is prone to error due to manual processing operations at some MFIs and the sheer volume of data that needs to be so processed. c. There is no established centralized system for conducting credit checks. d. Information systems do not adequately capture social data flows and displacement of people and present these in an appropriate aggregated form. e. Acquiring quality data for MFI operations comes at a cost. Data are not readily available because of losses during the period of conflict and the obsolescence of systems. f. There is an asymmetry in data flows due to intense competition among MFIs to provide financial services to the poor. Information is closely guarded and marked by an absence of flow between MFIs. g. Poor clients lack financial knowledge and, being mostly illiterate, cannot benefit from the expensive educational programs currently offered by MFIs. This poses a risk of default for the MFIs because

Status of Information Systems in Microfinance Institutions

293

their customers tend to make short-term decisions (such as taking another loan) based on their condition. A proper information system should cater not only to the stakeholders upstream (i.e., the donors and authorities) but also serve the needs of the clients downstream through appropriate and relevant financial literacy programs. These factors suggest a link between the current state of systems at MFIs and infrastructure in DRC and the actions of the clients of the MFIs in misusing the microcredit facilities for consumption rather than generating economic activity, thus getting further into debt.

Conclusion and Discussions The increasing indebtedness of client populations in poor neighborhoods is related to several factors: short-term credit, high interest rates, inadequate credit amounts and expensive training. Non-repayment of installments due may result in the confiscation of property pledged as collateral. In addition, as we have seen above, there is a positive correlation between information system and indebtedness among the clients. In addition, repayment issues and default in payments may affect access to new credit cycles and thus diminish the possibilities of risk management available to poor clients. Added to these factors is the limited access to information on the character and credibility of lenders. These factors contribute to a continuing indebtedness among poor clients of MFIs. The lack of information on the financial mechanisms increases the risk of debt distress. The extreme poverty of DRC coupled with financial illiteracy causes customers to make short-term decisions. They are forced to borrow more to sustain themselves. This poses risks for both MFIs as well as their clients. Therefore, information is essential for both organizations and microfinance clients in poor districts to reduce the risk of “over-indebtedness”. MFIs face a higher amount of financial risk due to the nature of the undertaking. Their attempt to make themselves self-sustaining by building in transaction costs (such as expensive training) and the costs of risk into the pricing model, reflected in high interest rates, detracts from their social objective of reaching out to the poorest sections of society and including them in the financial eco-system. Performance on these social objectives, beyond mere financing, needs to be highlighted, but globally available software is not designed to address these requirements. Moreover, information asymmetry can be addressed in part through policy measures,

294

Chapter Fourteen

especially in areas of education and training. The MFIs can facilitate the process by designing their information systems to work within the constraints imposed by the available infrastructure and by matching their product offering to suit the specific requirements of the client population they serve. We see the future for information systems in MFIs in the DRC at multiple levels. Firstly, the institutions need to implement computerized systems using appropriate statistical and social survey analysis software for processing qualitative and quantitative data. The Association for Survey Computing publishes a list of software relevant to data capture from social surveys and their processing (Software Register, 2006). The lending organizations need to capitalize on the current technologies available such as mobile telephony systems for data acquisition. Ultimately, the creation of a centralized credit database to capture defaults in repayment of loans by customers will allow MFIs to better calibrate their offerings and prevent customers becoming over indebted. To do this, however, a greater degree of integration within the country’s financial sector is required, bringing MFIs and the business banking systems together.

Bibliography African Union. (2009). Advancing the African microfinance sector. Addis Ababa, Ethiopia: Extraordinary Conference of African Ministers of Economy and Finance (CAMEF). Association for Survey Computing: Software Register (2006). Retrieved Dec 18, 2012, from http://www.asc.org.uk/wordpress/wp-content/ uploads/2009/11/REG-06-12a.pdf Bouquet, E., Wampfler, B., Ralison, E., & Roesch, C. (2007). Trajectories of credit and vulnerability of rural households: The case of CECAM in Madagascar. Paris. Central Bank of DR Congo. (2005). Instruction No. 1 of 12 September 2003 (Amendment No. 1 of 18 December 2005) of the Central Bank Of Congo, on the activity and control of MFIs, as amended December 18, 2005. CGAP & MIX. (2011). MIX Microfinance World: Sub-Saharan Africa Microfinance Analysis and Benchmarking Report 2010. Microfinance Information Exchange (MIX) and Consultative Group to Assist the Poor (CGAP). Retrieved Dec 12, 2012, from http://www.themix.org/sites/default/files/096_MIX_Africa%20Report_ 05-05-2011.pdf

Status of Information Systems in Microfinance Institutions

295

CGAP. (2005, Apr 23). Helping to Improve Donor Effectiveness in Micronance, Funding micronance technology. Donor Brief Foster, V., & Benitez, D. A. (2011). The Democratic Republic of Congo’s Infrastructure A Continental Perspective. Sustainable Development Department, Africa Region. The World Bank. Retrieved Dec 11, 2012, from http://www-wds.worldbank.org/servlet/WDSContentServer/WDSP/IB/ 2011/03/17/000158349_20110317160020/Rendered/PDF/WPS5602.pdf Georgel, F. (2009). IT gouvernance: management stratégiques d’un SI. (Paris, Dunod) : 286. Gilman, L., Genova, A. and Kaffenberger, M. (2013). Mobile Money in the Democratic Republic of Congo: Market insights on consumer needs and opportunities in payments and financial services. Retrieved on August 31, 2013, from http://www.gsma.com/mobilefordevelopment/wpcontent/uploads/2013/07/Mobile-Money-in-the-DRC_July-2013.pdf Hajer, K., & Kalika, M. (1998). Evaluation of the SI: A perspective of organization. 23(1), 18. Paris: Economica. HDR. (2011). Sustainability and Equity: A Better Future for All. (H. D. 2011, Producer) retrieved Dec 11, 2012, from Human Development Report 2011: http://hdrstats.undp.org/images/explanations/COD.pdf IFAD. (2006). Handbook for the analysis of the governance of microfinance institutions. International Fund for Agricultural Development (IFAD). Retrieved Dec 10, 2012, from http://www.ifad.org/ruralfinance/pub/handbook.pdf IFMR Research. (n.d.). New challenges in microfinance--Raised expectations. Presentation: Centre for Microfinance, IFMR Research. IMF. (2007). Democratic Republic of the Congo: Selected issues and statistical appendix, Country Report No. 07/329, International Monetary Fund, Retrieved on August, 31, 2013, from http://www.imf.org/external/pubs/ft/scr/2007/cr07329.pdf —. (2012). World Economic and Financial Surveys, Regional economic outlook, Sub-Saharan Africa, Sustaining Growth amid Global Uncertainty. International Monetary Fund. Joliot, D. (2003). Performance of IS: Checks, comparisons, test and measurement services for the management of the company. Paris: Economica. Lascelles, D., & Mendelson, S. (2012). Microfinance Banana Skins 2012: The CFSI survey of microfinance risks: Staying relevant. UK: Centre for the study of financial innovations.

296

Chapter Fourteen

Laudon, K., & Laudon, J. (2006). Management of IS. Paris: Person Education. Lhuilier, J. N. (2005). The systems of information management: Data and knowledge and skills. Paris: Lavoisier. Nanyonga, A. (2008, Apr 16). Microfinance Grows Up: Success Brings New Challenges for Investors, Practitioners, in Emerging Economies. Knowledge@Wharton. Retrieved Dec 10, 2012, from http://knowledge.wharton.upenn.edu/article.cfm?articleid=1940 ProCredit Bank Congo. (2012). ProCredit Holding. Retrieved dec 11, 2012, from http://www.procredit-holding.com/front_content.php?idcat=31 Reix, R. (1995). IS and organizational management. Paris: Vuibert. Schwarz, S. (2011). Financial institutions’ challenges to provide credit in the Democratic Republic of Congo, KfW Bankengruppe, Germany, Retrieved Aug 31, 2013, from https://www.kfw-entwicklungsbank.de/Download-Center/PDFDokumente-Sektoren-Berichte/2011_06_Congo-Kredit_E.pdf Schwarz, S. (2011a). Entrepreneurs’ challenges to access credit in the Democratic Republic of Congo, KfW Bankengruppe, Germany, Retrieved Aug 31, 2013, from https://www.kfw-entwicklungsbank.de/Download-Center/PDFDokumente-Sektoren-Berichte/2011_01_Congo-Unternehmer_E.pdf The World Bank. (2012). Democratic Republic of Congo Overview. Retrieved Dec 12, 2012, from The World Bank: http://www.worldbank.org/en/country/drc/overview UN. (2012). United Nations E-Government Survey. Retrieved Dec 14, 2012, from http://unpan1.un.org/intradoc/groups/public/documents/un/unpan04806 5.pdf Watereld, C., & Ramsing, R. (1998). Management Information Systems for Micronance Institutions, A Handbook. Pact Publications.

Notes 1

The authors would like to acknowledge the contribution of Mr. Devy Denadi from the Democratic Republic of Congo in providing some key data for this chapter. 2 Estimates from the National Statistical Institute (INS)

PART FOUR INFORMATION SYSTEMS CATERING TO THE MICROFINANCE INDUSTRY KRISHNA NYAPATI

The problem of credit reporting at the base of the pyramid is one of the major challenges facing MFIs worldwide. As per the CGAP Report on Credit Reporting (CGAP, 2011), while the industry has shown a healthy growth rate in terms of beneficiaries, the portfolio risk has steadily increased over the past few years, with deteriorating quality of assets. While this is due to a number of factors, it is believed that credit reporting is certainly one of the major factors which can be used to improve the quality of loans, reduce over indebtedness and improve the sustainability of MFI operations. The development of credit bureaux, addressing the needs of the MFI sector, is relatively new in India, and may be seen as an evolving component of the MFI eco system. This is a development which can have a significant, positive impact on the development and growth of the MFI sector, which attempts to address the problem of financial exclusion, using the model of social innovation. One major impact of credit bureaux is that they can provide a competitive advantage to the MFIs which make use of the information provided by them, in the market place for micro credit. In competing with banks, MFIs which base their decisions on the borrower’s credit history will be in a much better position to approve a larger proportion of loan applications, as compared to traditional banks. Considering another source of competition, namely the traditional money lender, MFIs can offer the advantages of lower interest rates, once again based on credit history. Finally, MFIs which participate as members of a credit bureau would always have an advantage over non- participating MFIs, since they would be in a much better position to take improved decisions relating to credit. Such data would enable MFIs to significantly enhance their ability to take informed decisions, reduce risks of default and improve credit

298

Part Four

discipline by borrowers, leading to significantly reduced operating expenses. Such improvements would also serve to remove major bottlenecks on both supply and demand sides of the MFI’s operations. Thus, we expect that the growth of credit bureaux will lead to the improvement of internal, operational efficiencies combined with improved competitiveness in the market place. The importance of credit ratings of potential borrowers of MFIs, and the need for a centralized database of borrowers, including payment history related data, became highlighted recently, as a consequence of the crisis faced by the MFI industry in Andhra Pradesh, India, in 2010. Some of the major contributory factors to this crisis included multiple loans availed by borrowers, inability to meet debt repayment obligations and coercive practices employed by MFIs, leading to tragic outcomes including suicides. One of the responses to this crisis was the appointment of the Malegam Committee by the Reserve Bank of India, which was mandated to recommend appropriate regulatory measures for the MFI industry. A key recommendation of the Malegam Committee was the need to establish suitable Credit Rating Bureaux, in order to address some of the identified problems, namely over indebtedness following from multiple loans. The two papers in this section address different aspects of the issue of credit bureaux. Chapter 15, contributed by Krishnan and Pal, looks at the design and architectural aspects of a system which can be deployed for collecting and sharing data required for a credit rating agency, and also looks at the problem of unique identity for each member of the database. The authors highlight the twin problems of adverse selection and moral hazard, which are to be addressed by the proposed system design. The proposed design is described as a hybrid model which includes semantic analysis using a composite key for better identification of borrowers, combined with a distributed, peer-to-peer architecture for data collection. The authors compare the data collection architecture currently used by CIBIL (a credit rating agency in operation) with their proposal, and highlight several benefits of the proposed architecture. In addition to the specific recommendations related to design and architecture, the paper also serves to highlight the many dimensions of information and data quality which need to be addressed, in particular accuracy, standardization and timeliness, in order that the industry may derive all the benefits possible from a credit rating eco system. The challenge posed by multiple institutional structures existing in the MFI sector, and the need for integrating all of them into one operational framework, are also discussed in detail.

Information Systems Catering to the Microfinance Industry

299

The second paper, Chapter 16 contributed by Prasad and Arya, is basically a case study of the Micro Finance Institutional Network (MFIN), and the role it played in the development of the credit rating institutions for MFIs. As already noted earlier, the Malegam Committee (2011) identified the need for credit rating, and MFIN played a pioneering role in taking this important decision to implementation, through a variety of initiatives, including education, engagement with all the micro lending institutions in order to get them on board, and making an equity commitment on behalf of the industry, in High Mark, an internationally reputed credit agency. The paper also identifies a variety of problems and challenges, including those related to widening and deepening the credit rating operation, and issues related to data quality.

Bibliography CGAP (2011) Credit Reporting at the Base of the Pyramid-Key Issues and Success Factors. Malegam, Y.H. 2011. Report of the Sub-Committee of the Central Board of Directors of Reserve Bank of India to Study Issues and Concerns in the MFI Sector. Reserve Bank of India.

CHAPTER FIFTEEN DESIGNING A DISTRIBUTED MICROFINANCE CREDIT BUREAU SYSTEM SUDEEP K. KRISHNAN AND DEBDATTA PAL

Introduction Unlike general transactions in spot markets, credit transactions involve significant time gaps between lending and repayment. A successful credit contract depends on a lender’s ability to gather and process information about the prospective borrower in order to minimize adverse selection and monitor the loan account to reduce moral hazards. This often leads to the suboptimal allocation of financial resources or to supply-side credit rationing (Stiglitz & Weiss, 1981). Prospective borrowers are asked to convey their creditworthiness either through the support of marketable collateral or through third-party guarantees, which they may fail to offer. In this case, a resource-poor applicant may be denied the credit facility. However, as part of the negotiation process between the lender and the prospective debtor, the latter may have invested considerable time and experienced high transaction costs, involving traveling expenses and wage losses. By anticipating further rejection, poor people may thus become reluctant to approach the lender in question. This can result in demandside credit rationing (Guirkinger & Boucher, 2008; Sarap, 1987). To overcome the issue of asymmetric information, lenders often focus on group-based microfinance to reach poor clients who are short of marketable collateral. Mutual trust, peer pressure, and joint liability can act as surrogates for marketable collateral, ensuring timely repayment (Bhatt & Tang, 1998). The very low levels of delinquency involved in microfinance models have raised interest among development agencies, which employ microfinance as a potential tool for development, and among a number of profit-oriented institutions, who use it to fill in gaps between demand and supply in credit markets in less developed countries (LDCs). For example, Banco Solidario of Bolivia and Banco Compartamos

Designing a Distributed Microfinance Credit Bureau System

301

of Mexico moved from a not-for-profit microfinance provider model into full-fledged profit-oriented financial institutions (Sriram, 2010a). One of the primary concerns involved in the commercialization of microfinance is the changing focus of microfinance institutions (MFIs) on larger loan clients rather than on small-ticket loans, as well as on repeat lending rather than adding new impoverished clients to portfolios (Navajas et al., 2003). The reason for these changing approaches is the goal of reaching efficiency and financial sustainability at a much faster pace, even at the cost of outreach (Hermes et al., 2011). The Indian microfinance market has also witnessed the entry of several for-profit MFIs, which, rather than developing new clientele (i.e., focusing on outreach), are increasingly competing against each other to lend to the same borrowers (Ghate, 2007). The justification for this behavior is that it saves the time and costs involved in group development, as mature groups are expected to have already reached a stage of solidarity. Hence, older groups are preferred to newer groups for the immediate scaling up of microfinance portfolios. The immediate effect of this scaling up through multiple lending is the inadequate assessment of the credit absorption capacity of the group or individuals, which results in their over-indebtedness and subsequent failures to honor repayment commitments (Sriram, 2010b). In response to the poor conditions of repayment, several MFIs are reported to have coercive recovery mechanisms, which include picketing in front of the houses of defaulted borrowers and using abusive language. It is alleged that such incidences have even led to suicides among defaulted borrowers (Ghate, 2007). Given the severity of this situation, the Reserve Bank of India (RBI), which is the country’s central bank, appointed a committee under the chair of Y. H. Malegam, board member of RBI, to look into issues related to microfinance. In its report, the committee observed that intensive competition among MFIs in some parts of India has resulted in both excessive borrowing and multiple lending. The committee’s view was that such occurrences could be controlled if amounts of outstanding loans and numbers of lenders could be easily ascertained. As a remedy, it suggested that MFIs not exceed the prescribed aggregate borrowing cap of any client and that the maximum number of MFIs lending to any single client be restricted to two (RBI, 2011a). By accepting the recommendations of the Malegam Committee, the RBI issued guidelines for micro-lending activities. These guidelines proposed that appropriate grading methods be employed for adequate pre-lending assessments of group members with regard to their readiness to absorb credit doses and that no more than two MFIs lend to a borrower at any given point of time (RBI, 2011b). As an institutional

302

Chapter Fifteen

mechanism to ensure that both outstanding loans and numbers of lenders remain within the prescribed limits, the Malegam Committee also recommended the establishment of one or more credit information bureaus for the microfinance sector, of which all MFIs should be members (RBI, 2011a). These will exist separately from the already existing Credit Information Bureau (India) Ltd. (CIBIL), which has been operating since 2002.

Roles of Credit Bureaus Credit bureaus are information brokers that gather, store, and share information, popularly termed credit history, among financial institutions, usually based on the principle of reciprocity (Pagano & Jappelli, 1993). Credit history primarily reflects the repayment behavior of a client, including incidences of late payment, default, and solvency (Chen, 1999). Credit history ranges from limited information on past defaults or arrears (“negative” data) to in-depth reports that, in addition to reflecting clients’ repayment history, also include information on their employment, assets and liabilities, guarantees, debt maturity structure, and family backgrounds (“positive” data) (Jappelli & Pagano, 2005). The submission of information to a credit bureau is generally voluntary, but credit bureaus operate on the principle of reciprocity, and so if a financial institution refuses to submit data on its clientele, it will also be unable to take advantage of the services that the credit bureau offers. Exchanging information with peer groups through credit bureaus leads to a number of outcomes. First, credit bureaus enable financial institutions to make more informed decisions on whether to consider applications or not, and in considering them, to determine what are appropriate limits and tenure. This reduces the chance of adverse selection and minimizes the number of “lemons” or bad borrowers that an institution must deal with (Akerlof, 1970) in its client set. Adverse selection takes place when some portion of prospective borrowers’ information remains unrevealed to the lender. Credit bureaus attempt to institute a screening mechanism through information sharing, which isolates bad borrowers from their good counterparts and ensures that loans go into the hands of clients with good repayment histories. This also improves volumes of credit outlay (Pagano & Jappelli, 1993). Second, credit bureaus act as moral police for borrowers, as noncommitment to contractual obligations with one lender may result in the denial of access to financial services from another lender. This enforces

Designing a Distributed Microfinance Credit Bureau System

303

credit discipline among borrowers (Jappelli & Pagano, 2000; Padilla & Pagano, 2000). Third, given the availability of loans from multiple lenders, borrowers who engage in bad financial planning can often borrow from multiple sources, which may allow them to exceed their borrowing capacities and become over-indebted. Conversely, a lender that is clueless as to an applicant’s overall indebtedness may sanction further credit facilities and thus be exposed to higher risk. By disclosing information to lenders, credit bureaus not only minimize borrowers’ incentives to borrow from multiple sources, but help the lenders limit their exposure to over-indebted individuals (Pagano & Jappelli, 1993). Fourth, a credit bureau also brings benefit to the borrower by reducing ‘informational rents’ (Jappelli & Pagano, 2005). When there is a dearth of appropriate information-sharing mechanisms, a lender with knowledge that is superior to that of its competitors can price loans slightly lower, and earn an economic rent1 from its broader informational access. Sharing information with peer institutions reduces the rent that arises from informational advantages, and leads to competitive pricing of loan products. Fifth, credit bureaus also increase access to credit at lesser costs for good borrowers (Jappelli & Pagano, 2005). More advanced credit bureaus that use sophisticated statistical techniques can produce credit scores based on individuals’ credit histories and other characteristics. Credit scores help financial institutions assess credit applications, and link the pricing of loan products and limits of loan facilities to credit scores. An applicant with a higher credit score can get a loan of a greater amount at a lower price than an applicant with a lower credit score can. Hence, credit bureaus ensure that good borrower behavior is rewarded with quantity and price advantages that are inaccessible to delinquent borrowers. Cross-country empirical investigations by Jappelli and Pagano (2002) have found that the presence of credit bureaus has a positive effect on expanding credit markets and minimizing credit risk. Empirical evidence (see Luoto et al., 2007) shows that information sharing among MFIs through Crediref, a centralized microfinance credit bureau in Guatemala, resulted in a decline in loan arrears of between 2 and 3.5 percentage points. Furthermore, Gonzalez (2007) has observed that MFIs from nations with credit bureaus have operating expense ratios (OERs)2 that are on average 5% lower than the MFIs of nations without them. This makes it clear that the presence of credit bureaus improves microfinance operation environments, leading to lower transaction costs, in terms of the minimization of search cost, negotiation costs, and enforcement cost in

304

Chapter Fifteen

any microfinance contract. Thus, credit bureaus enhance transparency in credit operations and foster more stable and vigorous financial systems. Unlike in developed nations, where obtaining information from credit bureaus prior to lending is standard practice (Yotopoulos & Floro, 1992) the evolution of credit bureaus in developing nations like India is still in its nascent stages. This may be attributed to the dominance of the large number of small-ticket accounts, predominantly in rural areas, poor infrastructure, non-adherence to standard formats of data submission, and a lack of official identification among low-income client groups, to name only a few factors. Currently, MFIs are scaling down their loan sizes (Sriram, 2010b) and engaging in localized exchanges of information on defaulters. However, these are not sustainable solutions for MFIs, and levels of exchanges are normally limited. This study proposes an information system framework that addresses the challenges presently faced by credit bureaus in the microfinance sector. It develops design specifications for a distributed system for microfinance credit bureaus, and builds on existing distributed systems literature, ad-hoc networks, and credit scoring studies. Distributed systems are collections of independent computers that appear to users as a single coherent system (Tanenbaum & Steen, 2002). Ad-hoc networks are decentralized networks that do not rely on pre-existing infrastructures. The designed system will be able to distinguish between end customers without specific identification variables, and calls for centralized data collection features among different credit bureaus in a distributed format.

Research Objective and Methodology The objective of this study is to develop design guidelines for a management information system (MIS) with distributed architecture for microfinance credit bureaus specific to the context of India as well as other LDCs. Advanced credit bureaus assign on-demand credit scores to each potential borrower. Credit scoring is a numerical expression assigned to a borrower or prospective borrower to estimate how risky she or he is. Hence, it is essential to identify every individual specifically. Our proposed system design considers the fact that in developing countries like India, it is hard to develop a unique identifier for individuals, and with this in mind, uses a distributed architecture for data collection by MFIs to improve efficiency and reduce latency in loan approval. The study involved the use of multiple research methodologies to develop design elements for the proposed system. The first stage included

Designing a Distributed Microfinance Credit Bureau System

305

qualitative semi-structured interviews. Qualitative enquiry provides tools for rich exploration, since this form of research reveals the “what” and “how” of phenomena (Lee, 1999). In the first stage of this study, interviews were conducted with microfinance experts, in order to analyze issues in lending processes and deficiencies in currently existing scoring systems. A theme-based (either related to issues with existing systems or proposed system features) content analysis of these interviews, using open coding methodology, was performed to identify existing issues with MFIbased data collection and multiple borrowing, along with the requirements analysis for the proposed system. Major themes were identified based on the contributions of respondents, corresponding to each of the issues and features that the proposed system should have. The themes are discussed in the following section of this chapter. In the next stage, the existing software development architectural paradigms of distributed and ad-hoc networks were adopted in a microfinance context, in order to develop the system architecture for an MIS specific to microfinance. The study proposes a high-level system design, along with data schema and data collection processes. A scenariobased analysis was also performed, describing the operational aspects of the system as a means of highlighting the efficiency of the proposed distributed data collection method. In the final stage, interviews were conducted, in order to validate the proposed system. The findings of this research study are categorized into two sections. The first section discusses the issues faced by the MIS identified during the detailed interviews, while the second explains the high-level MIS design involved in the proposed system.

Insights from Interviews – Issues in Microfinance Lending Based on a thematic analysis of the interviews conducted with microfinance experts, this study identified a number of issues in microfinance lending, and developed a requirement analysis for software design based on these issues. First, there are a large number of scattered low-income borrowers in the microfinance sector. Many of those borrowers lack proper identification documents, because documents such as voter identity cards and tax/permanent account number3 cards are normally limited to urban dwellers in countries like India. The identification of individual borrowers is one of the most important issues for all credit bureaus.4 In the absence of national identification numbers, credit bureaus may identify individuals

306

Chapter Fifteen

in generating their credit reports by combining information on their names, dates of birth, and addresses. Second, many borrowers receive loans from multiple MFIs and can become clients of government-run poverty alleviation programs and government-run self-help groups (SHG) or Swarna Jayanti Shahari Rozgar Yojana (SJSRY), a government-sponsored employment generation scheme operating through group-based thrift and credit mechanism. Information on these details is not normally disclosed to the MFIs. Third, even though MFIs work independently, an MFI may be interested in lending to an existing client of another MFI, as the client’s repayment history with the other MFI indicates his or her good credit standing. This saves the cost of developing new customers and networks and of familiarizing them with micro-borrowing, fund management, credit rating checks, and so forth. The MFIs perceive that with their previous loans, clients may have already created productive assets, and thus carry less risk than first-time borrowers. Fourth, as with any normal financial agency, the urge to grow has pushed many MFIs to engage in excessive lending. Additional loans are also often extended with inadequate grace periods on principal. The available cash surplus from the production activities of borrowers may fall short of their repayment obligations, making the clients borrow further to remain in the “good books” of their lenders. This leads to overindebtedness, perpetual indebtedness, and the potential breakdown of the system. An example of such a situation can be found in the state of Andhra Pradesh, where usurious interest rates5 and inhumane recovery mechanisms6 have met with strong responses from civil society as well as the government (Ghate, 2007). Fifth, the availability of data for credit scoring is a clear issue for MFIs. Apart from identifiers for borrowers, the non-standard information capture formats used by MFIs make data analysis operations more difficult than in the case of standard formats followed by banks. Even when appropriate data sets are available, the prediction models used may not match results for the same borrower from multiple MFIs. A major reason for the non-availability of standardized data is the general lack of expertise in data entry, as the majority of the MFIs operate on low-cost models, using local staff with limited education. Sixth, a hindering factor in countries like India was the cost of generating credit reports. The cost of a single report from the oldest and the largest credit bureau, CIBIL, ranges from US$0.45-0.557. This is a relatively high expense for an MFI to bear, given that most aim to keep costs for checking credit to a maximum of US$0.20 per client. Initially,

Designing a Distributed Microfinance Credit Bureau System

307

CIBIL was the sole service provider in the Indian financial market, and as a monopoly, advised MFIs to provide specific identifiers to locate their clients instead of developing improved operational systems. As the government issued licenses to more credit bureaus, there are three credit bureaus presently operating in the Indian microfinance sector: Highmark, Equifax, and Experian. The cost of generating a credit report for an individual loan applicant through these newer organizations is below US$0.20. Seventh, there are no incentives for Indian MFIs to share data. Unless a situation of competition arises and the MFIs foresee chances of multiple lending, credit history would not be reported. The lack of regulation and the costs of checking also exacerbate this issue. The nonexistence of features such as real-time updates, deadlines for transfer of data, and penalties for non-submission are additional identified issues. In LDCs like India, these issues are not limited to MFIs, and even credit bureaus suffer.

Proposed System The proposed design for the MIS follows the guidelines provided by the United States Agency for International Development (USAID, 2006). It involves a hybrid system that uses both statistical (decisions empirically derived from past data) and judgmental (decisions structured by individual and institutional judgment) scorecards. The high-level design of the system is depicted in Figure 1.

Figure 1: Proposed System Design

308

Chapter Fifteen

The system design in Figure 1 may be implemented as a hybrid system, by using the features identified during interviews and making use of a decision-making engine that is similar to existing credit scoring models. The USAID guidelines for developing a credit scoring model involve three main attributes—non-financial, facility, and financial. Nonfinancial attributes include demographics, repayment history, and reference factors. Facility attributes are related to the collateral coverage for loans. Financial attributes refer to projected revenues and business growth. These attributes can be directly checked to develop a scoring model for borrowers in the MFI context, provided the data are structured and the statistical models can be directly applied. The model proposed by this study differs from traditional systems in its development of a structure for collected data, and in the improved availability of data across a scattered population. The design changes affect the prediction models for the new borrowers and semantic rules depicted in Figure 1. The improvements of this design result from two main factors: 1) the data schema design and semantic rules that are applied to identifiers, and 2) the distributed data collection architecture. The semantic rules are to be applied to historical data as well as that which is newly captured. These proposed semantic rules and the corresponding database schema design are described below. The data collection methodology for new populations across MFIs will employ a distributed architecture, reducing the overall costs and efforts required of MFIs and central agencies. The proposed data collection mechanism is also particularly feasible in developing nations like India.

1. Data Schema Design A schema of a database system is its structure, described in a formal language supported by the database management system (DBMS). It refers to the organization of data to create a database, which is a collection of different data tables (Elmasri & Navathe, 2008). A database schema consists of integrity constraints imposed on tables for its structuring. An essential aspect of every database design is the unique identifier for each of the entries in the table. These identifiers are called the primary or unique keys. A primary key can be an entry from one single table that uniquely identifies each of the data entities involved. For example, in the case of microfinance, an entry that uniquely identifies an individual or firm will be the primary key. This primary key can also be a unique combination of more than one table entry. A composite key, in the context of relational

Designing a Distributed Microfinance Credit Bureau System

309

databases, is a combination of two or more columns in a table that can be used to uniquely identify each data entry in the database. Uniqueness is normally guaranteed only when the columns are combined, and when taken individually, the columns may not guarantee uniqueness (Elmasri & Navathe, 2008). The composite index for borrower identification should include borrowers’ details including first name, last name, approximate age, and location (zip code). If the borrower has a unique identifier number in the form of a voter card or ration card, this identifier can be used as a duplicate key. Once the key is developed, prior to any form of analysis, semantic rules need to be applied to the name field. This involves the decision-making engine and inputs from analysts. An example of a context in which semantic rules may be applied is as follows. The name of an applicant may be DEBDATTA. However, this name may have been captured in different formats by different collection agents. Phonetic aspects that identify the client should be applied, using semantic rules. Probable transformations for the name would include, for example, DEBDATTA, DEVDATTA, DEVDUTTA, and DEBDUTTA (DEBDATTA -> DEBDATTA / DEVDATTA / DEVDUTTA / DEBDUTTA) The biggest challenge for the system would be the different formats for the name that could be used, and it would be hard to come up with a universal set of name transformations. Human interaction, even at a minimal level, would help to tackle this issue. The necessary inputs would not need to be provided by information technology (IT) executives or microfinance experts. Operators with knowledge of local nomenclature could provide inputs to the system, and machine learning from historical data would help in the identification of clients’ payment histories. The nomenclature system could be coded with any data analysis tool, such as the Statistical Analysis System (SAS) or R, which could consider possible versions corresponding to each name as well as the developed primary key that has been mentioned above.8

2. Distributed Data Collection Model Distributed Systems or Distributed Computing Systems (DCSs) are collections of independent computers that appear to users as a single coherent system. These systems are normally collections of heterogeneous systems from various vendors running on different operating systems. The advantages of DCSs are that they offer easy-to-expand infrastructure frameworks and that their underlying communication aspects are hidden

310

Chapter Fifteen

from users. The most important characteristic of a DCS is that it is permanently available, even if parts of the systems are inactive. The data collection architecture shown in Figure 2, below, involves centralized control between the central agency and the MFIs. The proposed architecture uses decentralized (peer-to-peer) control between the collection agencies or MFIs (Lua et al., 2004). The distinctions between centralized and decentralized distributed systems involve three primary factors, namely resource discovery, availability, and communication. In a centralized system, the server or group of servers is expected to be available for communication between the end systems in the architecture, for the discovery of resources, services, and peers, and for the transfer of information. In a decentralized version, the peer systems communicate with other available peers and share information directly.

Figure 2: Distributed Data Collection Architecture

In the proposed system design, the prediction model will run in the centralized server or group of servers. A hash table consists of an array of data accessed using indexes or key values. In hash tables, mapping between all elements and positions in the array is carried out by using a hash function (Elmasri & Navathe, 2008). A hash table that is developed corresponding to the clients in a region or area of operation will be transferred and regularly updated with the end systems at collection agencies. As shown in Figure 2, the peer-to-peer (P2P) systems at the collection agencies will run on a distributed system based on ad-hoc connections. Ad-hoc connections do not require pre-existing infrastructure for network creation. However, peer recognition will occur through the central server whenever required, based on overlapping areas of operation

Designing a Distributed Microfinance Credit Bureau System

311

or between MFIs who share information. For data collection at client locations, we suggest that handheld devices be used to capture client information, which will then be regularly updated to the MFI, followed by the central bureau. Similarly, the central bureau should be able to update population details as requested by MFIs from other MFIs. This design will eliminate high-end infrastructure requirements and will work on the power backups provided by devices. It will also only require individuals performing data collection to engage in minimal training. The extension in the proposed architecture includes decision-making capabilities through exported hash tables at the member institute systems, and the possibility of communication between the member institutes.

3. Sample Case of Implementation – CIBIL CIBIL is a credit information company that collects and maintains records of transaction details involving the loans and credit cards of individuals. Its information is provided to lenders on application. The architecture for data collection used by CIBIL falls into the distributed category. The CIBIL system employs a completely centralized distributed system architecture, and there is no direct sharing of information between member institutions. CIBIL has a central server to which individual MFIs update their information, and it passes the required information to each MFI directly. This architecture ensures that data analytics and decisionmaking happen at the central agency, which leads to latency in decisionmaking and the unavailability of information to MFIs who do not request it. Moreover, direct communication between MFIs is limited. Changes that the proposed system architecture would make to the existing mode of operation are discussed in the following scenario analysis section. Two scenarios, corresponding to the normal application process and simultaneous borrowing, are discussed.

4. Scenario Analysis for the Proposed Model Latency in loan processing will be reduced if the central agency uses the proposed model to transfer the details of a required population to individual MFIs or to transfer the hash table corresponding to the population in a region in which each of the MFIs operate. The stepwise process is as shown in Figure 3 below and can be described as follows: Step 1: The hash table corresponding to the historical data for an area is transferred to the system (with the MFI) at regular intervals.

312

Chapter Fifteen

Step 2: The client applies for a loan with the MFI. Step 3: The credit history and predictive model for credit scoring is run at the MFI end and the decision on the loan approval is conveyed to the client. Step 4: The central system is updated with regard to any new loans made to the client at a regular time interval. The semantic rules analysis on the borrower name and identification and prediction models can be run at the end system level, or can be transferred to the central agency as required. The proposed model also creates facilities for communication between member institutes.

Figure 3: Process Flow – Normal Applications

The scenario shown in Figure 4 is the one in which a client is simultaneously considered for a loan by two different MFIs. In the traditional systems, the individual could not be traced if previous transactions were absent and so might have loans approved by both agencies. In the proposed system, with an appropriate frequency of data update process in the central system, this issue can be handled. The tradeoff between loan approval time and efficiency of credit scoring will have to be considered in implementing extra requirements. If the data update process occurs at a relatively high frequency, the efficiency factor can be improved. No client application will be processed without the intervention of the predictive system and historical data. Multiple applications, semantic issues with names, and location-specific problems need to be addressed by the system. However, if the objective is to improve the timeefficiency of the process, updates should occur less frequently. The distributed agencies may also connect with peer systems for real-time

Designing a Distributed Microfinance Credit Bureau System

313

decision-making, as the number of entities involved will be lower at the end system level. Loans from other member institutes can be discovered through the intervention of the central agency, or an ad-hoc network can be created to transfer information.

Figure 4: Simultaneous Borrowing Scenario

5. Proposed Deployment In the current market, MFIs join credit bureaus and share their data with the corresponding bureaus. For example, in India, MFIs join either Highmark or Equifax, or both. The amount of data that is shared may vary, and since there are no incentives for sharing data, most of the data remain within MFIs. Moreover, MFIs need to pay to check the credit histories of individuals. As such, they typically only check for credit history when competition and chances of multiple lending are high. To date, there is a lack of a regulatory interface from RBI to verify that the system of crosschecking is actually being followed. For our proposed model to function efficiently, we propose that the MIS infrastructure be provided by credit bureaus to MFIs. This serves multiple purposes. First, data will be directly shared with credit bureaus. Second, the shared data will be in a standard format. Third, MFIs will pay for consolidated services. Fourth, MFIs will not need to work with multiple credit bureaus. However, this model will require infrastructure investment on the part of credit bureaus, and the migration of MIS and data currently used by individual MFIs. The credit bureaus can also

314

Chapter Fifteen

collaborate and deliver handheld devices to MFIs that can be used to collect data from clients, which will help in standardizing the process.

6. Validation of the Model In the final stage of this study, the proposed model was presented to two microfinance experts to validate the design and to receive feedback. A major advantage of the proposed system that was pointed out by the experts was the possibility of notifications being sent to different MFIs to address a common situation currently happening. When individuals approach multiple MFIs, each MFI is normally not notified by the others, and a normal credit check would not provide this information. However, our proposed design is capable of providing direct notification at any point when the same individual applies for a loan, even before processing. The semantic analysis of unique identifiers would also ensure better results in this process. The experts perceived the proposed data collection mechanism to be very relevant for developing nations like India, where access to IT infrastructure and power shortages are inherent problems, especially in rural areas, where there is the maximum likelihood of individuals becoming MFI clients. Standardized data collection and handheld devices will be effective in reducing data availability issues, and the resulting infrastructure, supported by credit bureaus, could induce growth for the whole sector. Many limitations in the sector may continue to exist. Currently, there are no facilities for data sharing between banks, MFIs, and MFI credit bureaus. Hence, it will not be possible to identify multiple borrowing across banks and MFIs. Moreover, there are still no penalties for not sharing data. Therefore, implementing appropriate regulations will be essential in the sector, and the RBI should play a dominant role in minimizing inefficacies affecting the functioning of credit bureaus and MFIs.

7. Challenges in Implementation The proposed system design assumes that the seamless exchange of information between different member institutes and the provision of connections between end systems is possible. Some issues identified through the interviews with experts are as follows. First, since MFIs are competing against each other to gain market share and given that sharing information with credit bureaus is not a mandatory part of regulatory compliance in India, some MFIs may not show enough

Designing a Distributed Microfinance Credit Bureau System

315

interest in sharing information on their clientele with competing MFIs. Moreover, commercial banks do not share information with the credit bureaus handling microfinance transactions. Hence, historical information on SHG/SGSY is not available to credit bureaus, which increases the risk of multiple lending. Second, training on data collection and capture formatting needs to be provided in order for the proposed model to be implemented. Third, issues with timely data submission will likely occur, as most of the branches of MFIs are in remote rural areas with poor electricity as well as limited broadband connections. If MFIs restrict their operations to areas with better infrastructures, the question of “mission drift” (not focusing on financial inclusion) will occur. Fourth, branches presently submit data to their controlling offices on a weekly basis, and the offices, in turn, submit the data to credit bureaus. This layering of information processing makes a client wait for at least a fortnight to avail of a loan after it has been sanctioned. However, the trade-off between latency and efficiency should be considered in determining the intervals of data updating in the system. Finally, in the existing model, after a lending decision is made at the branch level, information about the prospective client is sought from the controlling office, which submits a query to a credit bureau. The credit bureau then generates the information and returns it to the controlling office, which in turn provides information to the branch. This model of operations will have to be changed in order to implement the proposed architecture.

Conclusion The IT applications and software models currently used by MFIs in India are relatively sophisticated. However, within the MIS context, these systems continue to be primitive in nature, using traditional models without standardization for data collection. Best practices that have been thoroughly researched and deployed in other contexts have been used to develop the MIS design proposed in this study. The proposed system is expected to result in considerable performance improvements over the traditional systems developed by microfinance credit bureaus. The design guidelines for the proposed system are still in the nascent stage. The system deployment will require several iterations and testing in practice. Moreover, the design requirements mean that collaboration between MFIs will be necessary to improve efficiency in operations and the avoidance of multiple lending. Collaboration with other financial

316

Chapter Fifteen

institutions such as banks will exponentially increase the efficiency of the system and will have an additional policy impact on the sector. With regard to technology, the system requires machine-learning capabilities to implement semantic rules for unique identifiers as well as human intervention. With the increased usage and analysis of historical data, the machine learning aspects are bound to improve, which will lead to higher efficiency levels. The features proposed in the system are thus likely to transform the way the industry operates, and the resulting benefits will directly affect end customers.

Acknowledgements We acknowledge doctoral fellowship and dissertation support grant extended by Indian Institute of Management, Ahmedabad to the first author. Authors put on record the valuable discussion held with Chandra Shekhar Ghosh, Chairman and Managing Director, Bandhan Financial Services (P) Ltd., Shubhankar Sengupta, Chief Executive Officer, Arohan Financial Services (P) Ltd. and Rajesh Singh, Associate Vice President and Sharad Kumar Varma, Credit Analyst, Ananya Finance (P) Ltd. on the subject. Authors express their indebtedness to Prof. Arvind Astha, Prof. M. S. Sriram and two unanimous referees of this book for their valuable suggestions leading to enrichment of this chapter. Errors and omission remain the responsibility of the authors.

Bibliography Akerlof, G. A. (1970). The Market for 'Lemons': Quality Uncertainty and the Market Mechanism. Quarterly Journal of Economics, 84 (3), 488500. Bottomley, A. (1975). Interest Rate Determination in Underdeveloped Rural Areas. American Journal of Agricultural Economics.57(2), 279291. Chen, J. (1999). The Economics of Credit Information. Zagreb International Review of Economics and Business, 2(1), 63-84. Elmasri, R., & Navathe, S. B. (2008).Fundamentals of Database Systems. Boston: Pearson Education. Ghate, P. (2007). Consumer Protection in Indian Microfinance: Lessons from Andhra Pradesh and the Microfinance Bill. Economic and Political Weekly, 42(13), 1176-1184.

Designing a Distributed Microfinance Credit Bureau System

317

Gonzalez, A. (2007). Efciency Drivers of Micronance Institutions (MFIs): The Case of Operating Costs. Microbanking Bulletin, 15, 3742. Guirkinger, C., & Boucher, S. R. (2008).Credit constraints and productivity in Peruvian agriculture. Agricultural Economics, 39, 295308. Hermes, N., Lensink, R., & Meesters, A. (2011). Outreach and eƥciency of micronance institutions. World Development, 39(6), 938–948. Jappelli, T., & Pagano, M. (2000) Information sharing in credit markets: a survey. Working Paper No. 36, Centre for Studies in Economics and Finance, University of Salerno. Jappelli, T., & Marco P. (2002). Information Sharing, Lending and Defaults: Cross Country Evidence. Journal of Banking and Finance, 26 (10), 2017-2045. Jappelli, T., & Pagano, M. (2005) Role and Effects of Credit Information Sharing. Working Paper No. 36, Centre for Studies in Economics and Finance, University of Salerno. Lee, T. W. (1999). Using qualitative methods in organizational research. Thousand Oaks: Sage. Lua, E. K., Crowcroft, J., Pias, M., Sharma, R., & Lim, S. (2004). A Survey and Comparison of Peer-to-Peer Overlay Network Schemes. IEEE COMMUNICATIONS SURVEY AND TUTORIAL, 72-93. Luoto, J. E., McIntosh, C., & Wydick, B. (2007). Credit Information Systems in Less Developed Countries: A Test with Microfinance in Guatemala. Economic Development and Cultural Change, 55(2), 313334. Milgrom, P., & Roberts, J. (1992). Economics, Organization and Management. New Jersey: Prentice Hall. Mellor, J. W. (1968). The evolution of rural development policy. In J. Mellor, J. Weaver, U. Lele, and M. Simon (Eds.), Developing Rural India: Plan and Practice. Bombay: Lalvani Navajas, S., Conning, J., & Gonzalez-Vega, C. (2003). Lending technologies, competition and consolidation in the market for micronance in Bolivia. Journal of International Development, 15(6), 747–770 Padilla, A. J., & Pagano, M. (2000). Sharing Default Information as a Borrower Discipline Device. European Economic Review, 44(10), 1951-1980. Pagano, M., & Jappelli, T. (1993). Information sharing in credit markets. The Journal of Finance, 48(5), 1693-1718.

318

Chapter Fifteen

Reserve Bank of India (2011a): Report of the Sub-Committee of the Central Board of Directors of Reserve Bank of India to Study Issues and Concerns in the MFI Sector. Reserve Bank of India. (2011b). Introduction of New Category of NBFCs – ‘Non-Banking Financial Company-Micro Finance Institutions’ (NBFC-MFIs) – Directions. December 02. Sriram, M.S. (2010a). Commercialisation of Microfinance in India: A Discussion of the Emperor's Apparel. Economic and Political Weekly, 45(24), 65-73. Sriram, M.S. (2010b). Microfinance: A Fairy Tale Turns into a Nightmare. Economic and Political Weekly, 45(43), 10-13. Sarap, K. (1987). Transactions in Rural Credit Markets in western Orissa, India. The Journal of Peasant Studies, 15(1), 83-107. Stigliz, J., & Weiss, A. (1981). Credit Rationing in Markets with Imperfect Information. American Economic Review, 71(3), 393-410. Tanenbaum, A. S., & Steen, M. (2002). Distributed Systems, Principles and Paradigms. Prentice-Hall. USAID. (2006). A Handbook for Developing Credit Scoring Systems in a Microfinance Context. Washington: United States Agency for International Development. Yotopoulos, P. A., & Floro, S. L. (1992). Income distribution, transaction costs and market fragmentation in informal credit markets. Cambridge Journal of Economics, 16, 303-326.

Notes 1

Economic rent is the “excess return” above “normal profit” available in a competitive market (Milgrom & Roberts, 1992). 2 OER = operating expenses/loan portfolio. Operating expenses relate to operations, including all employee expenses, depreciation and amortization, and administrative expenses. 3 Permanent account numbers (PANs), issued by the Income Tax Department of India, are unique alphanumeric combinations given to every juristic entity identifiable under the Indian Income Tax Act of 1961. 4 The Indian government is in the process of issuing 12-digit individual identification numbers, known as aadhaar, which serve as proof of identity and address. 5 High interest rates can be explained through the default premium hypothesis (Bottomley, 1975; Mellor, 1968) which suggests that lenders raise interest rates to cover losses suffered through bad loans. 6 These may be viewed as actions intended to send out strong signals to the market that, unlike loans from government banks that are occasionally waived, loans from

Designing a Distributed Microfinance Credit Bureau System

319

MFIs need to be repaid on time. These actions may thus be intended to impose discipline among borrowers. 7 These figures were given by the research participants, who argued that this would raise the operational costs for MFIs. 8 The authors may be contacted for a sample algorithm that can be used for name checking.

CHAPTER SIXTEEN INSTITUTIONAL WORK IN BUILDING A CREDIT BUREAU FOR MICROFINANCE: THE MFIN CASE STUDY ALOK PRASAD AND VIBHU ARYA

Microfinance in India and the need for a credit bureau In India, “financial exclusion” is a reality with just 50% of adult Indians holding a savings account with a formal financial institution and an estimated under 8% of the population availing of a formal credit facility. Microfinance remains an important policy tool to achieve financial inclusion wherein the range of financial products and services offered are credit, savings, insurance, pension, and money transfer. The two important models / channels of microfinance involving credit linkages with banks in India are (I) Self Help Group (SHG) - Bank Linkage Model: This model involves the SHGs financed directly by the banks viz., Commercial Bank (Public Sector and Private Sector), Regional Rural Bank (RRB) and Cooperative Banks. (II) Microfinance Institutions (MFI) - Bank Linkage Model: This model covers financing of Micro Finance Institutions (MFIs) by banking agencies for on-lending to Joint Liability Groups (JLGs) and other small borrowers. JLGs consist of a group of five women whereas SHGs typically comprise of a 20 women group.

Institutional Work in Building a Credit Bureau for Microfinance

321

Figure 1: Coverage of various channels for financial inclusion in India and their share1. Self Help Group (SHG), Microfinance Institutions (MFI), Regional Rural Bank (RRB), Scheduled Commercial Bank (SCB). Source: Reserve Bank of India data & MFIN Analysis

1. The Non-Performing Assets of Bank - Self Help Group Lending (SHG)2 & Microfinance (JLG)3 Model Given the socio-economic profile of the customer base for financial inclusion which is characterised by low financial literacy, weak credit discipline, ignorance of the importance of credit history and an absence of credit information repository, the risk gets further compounded with multiple channels vying for the same customer profile and the business environment becomes extremely competitive. This in turn fuels the demand for higher credit leading to the risk of higher credit delinquencies in India’s financial system, depicted below4 in Table 1.

322

Chapter Sixteen

The Various Microfinance Models in India SHG-Bank Loan Amount of Linkage Outstanding in NPA Model USD (INR) USD (INR) Million Million Public Sector Bank (SHG) Private Sector Bank (SHG) Regional Rural Bank (SHG) Cooperative Bank (SHG) Total SHG-Bank Linkage Model MFI-Bank Linkage Model Total SHG + MFI

NPA as % of Loan Outstanding

4881 (2,44,060)

303 (15,180)

6.48

280 (14,030)

14.8 (740)

5.30

1722 (86,130)

85 (4260)

4.95

383 (19,160)

26 (1300)

6.84

7268(3,63,400)

430 (22,120)

6.09

2290 (1,14,503.4) 9557 (4,77,883)

50.86 (2,543.4)

2.2

480 (24,023)

4.04

Table 1: Status of Microfinance in India 2012, NABARD

2. The Microfinance Sector in India India has one of the largest microfinance industries in the world (Mix Market). The Self-Help Group (SHG)-Bank Linkage Program (SBLP) which was launched in 1992 has enabled 97 million poor households access to sustainable financial services from the banking system, with an outstanding institutional credit exceeding US $6 billion as of the end March 2012 (NABARD Report on Status of Microfinance in India, 2012). The other model of microfinance, i.e. the Microfinance Institutions model comprising of various entities, such as non-banking financial companies (NBFCs), non-governmental organisations (NGOs), trusts, cooperatives, etc. has also been growing significantly in the recent years. (MFIN Micrometer) Microfinance has established itself as a significant component of the financial system architecture of India; currently, the total number of MFI borrower accounts stands at ~31.8 million which is three times that of India’s regional rural banks (RRBs) and even higher when compared to that of commercial banks. With MFIs operating at 24% of total small loan accounts, the sector’s contribution to financial inclusion continues to rival that of Regional Rural Bank (RRBs 5 ). Microfinance

Institutional Work in Building a Credit Bureau for Microfinance

323

delivers convenient, economical, doorstep credit to women at the bottom of the socio-economic pyramid. 2.1 The Microfinance Sector Association: MFIN (Microfinance Institutions Network) Considering that several MFIs began entering the US $20 million asset base of “systemically important institutions category6”, MFI sector leaders clearly realized the need for ingraining self-regulation and compliance into their sectorial DNA. MFIN was established in December 2009 as a nonprofit society on the model of a Self-Regulatory Organization for the Microfinance Institutions (NBFC-MFIs)—constitutionally, structurally and functionally. MFIN membership comprises only NBFC-MFIs registered with the RBI. MFIN currently has 44 members, constituting over 85% of the microfinance institution business in India. At the fiscal year ending March 2012, MFIN members’ outreach was reported at 31 million low-income borrowers with a loan outstanding of INR 200 Bn. MFIN aims to “help provide inclusive financial services to 100 million low income households by the year 2020, in a responsible and transparent manner.” Its Byelaws state: “To act as a self-regulatory organization for the microfinance sector and to regulate the business of microfinance as carried on by its members.” Microfinance Credit Bureau Timeline 1 MFIN formed 2 Reserve Bank of India invites application for Credit Bureau 3 MFIN engages with applicants and invests in “one” applicant 4 The MFIN-invested applicant earns Credit Bureau license 5 MFIN drives Credit Bureau usage Table 2: Microfinance Credit Bureau Timeline 2.2 The Andhra Pradesh Act (2010) – Damage, Lessons and Mitigants Citing multiple borrowing and its consequences, ranging from over indebtedness to suicides, on 19 October 2010, the Andhra Pradesh (AP) State Government promulgated the Andhra Pradesh Micro Finance Institutions (Regulation of Money lending) Act 2010, which, in the opinion of several legal luminaries, is ultra vires India’s Constitution. India’s Central Bank took the position that duality of regulation is not in the public interest in response to a petition filed by MFIN and an MFI in

324

Chapter Sixteen

Andhra Pradesh High Court. The Andhra Pradesh MFI Act impacted normal business continuity resulting in substantially reduced equity inflows, slowing of fresh bank funding to the microfinance sector and an overall decline in Gross Loan Portfolio (GLP) and client numbers. In Andhra Pradesh particularly, the industry faces an existential crisis. A large number of medium-to-large size MFIN-member MFIs in Andhra Pradesh have accepted the corporate debt restructuring package offered by Scheduled Commercial Banks / Financial Institutions programs, totaling US $1.5 billion. Andhra Pradesh, which was the headquarters of six of the largest ten MFIs, has witnessed the industry repayment rates descend from 98% to single digits and net worth close to zero. Fresh debt and equity funding to Andhra MFIs is non-existent, while MFI disbursals have fallen from the Rs. 50 billion annual to less than Rs. 500 million. Independent studies indicate borrowers have begun accessing high cost funds from various sources including moneylenders. 2.3 The New Normal - The Malegam Committee Report & Need for Credit Bureau A high-powered committee of the Board of Reserve Bank of India, chaired by Mr. Y. H. Malegam, went into the full range of issues confronting the microfinance sector. In its report of January 2011, the committee, inter-alia, recommended a broad framework of self-regulation, a new separate category of NBFC-MFIs, tightening of provisioning norms and a credit bureau. According to the report, “An essential element in the prevention of multiple-lending and over-borrowing is the availability of information to the MFI of the existing outstanding loan of a potential borrower. This is not possible unless a Credit Information Bureau is established expeditiously (YH Malegam, 2011).”7 In advance of the Malegam Committee recommendation, MFIN and the sectorial leadership had already realised the importance and urgency of building a strong credit underwriting process seamlessly across the sector and controlling credit delinquencies within acceptable norms. This led to the need for establishing and deploying credit bureaux to provide member MFI credit grantors with the relevant information and analysis that enables them to make sound credit decisions. The first step to build the credit information reporting began in early 2010.

Institutional Work in Building a Credit Bureau for Microfinance

325

Case Study: The Development of a Microfinance Credit Bureau in India 1. Initiating Microfinance Credit Bureaux Over the last decade, India has witnessed an exponential growth, both in terms of demand and also supply of credit to India’s booming middle socio-economic category of population, largely residing in urban locations. The unintended consequence of this growth was intense competition amongst lenders, multiple borrowing and increased credit delinquencies. Only post the 1997-98 South Asian financial crisis did India’s Central Bank fully realize the need for putting in place an eco-system for a credit information bureau and then enacted a policy for Credit Reporting. 8 In 1999, RBI constituted a working group comprising of representatives from select public sector banks, IDBI, ICICI, Indian Banks' Association and Reserve Bank of India. The objective was to explore the possibilities of setting up a Credit Information Bureau. The group made four major recommendations. Firstly, it suggested that a CIB be set up under the Companies Act, 1956 with equity participation from commercial banks, FIs and NBFCs registered with Reserve Bank of India. Secondly, it felt that a foreign technology partner should be included as a collaborator in setting up of a Bureau. Thirdly, an appropriate legal framework was to be put in place to provide adequate protection to the Bureau as also the credit institutions sharing information with the Bureau. Finally, pending enactment of a master legislation / legal amendments, a beginning could be made for setting up a Bureau that could operate initially by pooling information on suit-filed accounts and transactions on which the borrower has given consent, for sharing amongst the user group.9 On the basis of the recommendations made by the 1999 RBI Working Group, India launched its first credit bureau in 2002 which began operations in 2004. Credit Information Bureau (India) Limited (CIBIL), a collaborative effort of scheduled commercial banks and other large nonbank financial institutions, has a database that today exceeds 250 million records of credit histories on individuals and businesses. Its member base has grown to over 650 members comprised of banks, financial institutions, NBFCs, housing finance companies, credit card companies and other credit institutions.10 As per the Credit Information Companies (Regulation) Act, 2005 every financial institution is required to be a member of at least one credit information company by law and to give data on monthly basis. Hence, the entire credit history of an individual is available in most of the cases,

Chapter Sixteen S

326

enabling creedit scoring byy the bureau. Specified useers can accesss the data outside of banks to enquire e aboutt individualss (insurance, telecom companies eetc.). Credit data d can be used u only for existing custtomers or once the appplication com mes; it cannot be b used for m marketing and selecting customer segments. Evver since, rissk assessmennt has gaineed prime importance for banks andd they make sure s of good ccredit worth and a credit history of ann individual beefore offering g loans. In 20099, India’s NB BFC-MFIs ap pproached CIIBIL to expaand their services covvering the miccrocredit customer. Being a monopoly and a given its own veryy recent vintaage and a larg ge existing meember clientelle, CIBIL appeared m mildly interestted, but repeeated request s and nudgees for an execution pllan cum strateegy have bornee no concrete results. Coincideentally, arounnd the samee time India’’s Central Bank, B the Reserve Bannk of India beegan accepting g applicationss from prospecctive new credit bureauux. NBFC-MF FIs looked at alternative aggencies that co ould serve the needs off the industryy and initiated d discussions with several applicant credit bureaaux, includingg an entrepreeneurial ventuure High Marrk Credit Information Services [H High Mark], that had a stated focuss on the unbanked/unnder banked and a was willin ng to immediaately serve th his market segment andd make approppriate investm ments in technoology and reso ources. Taking into consideration the uniqu ue and challennging data env vironment of microfinaance and needd for a continu uous and long--term investm ment in the microfinancce credit repporting busin ness, MFIN member MF FIs came together as a consortium m under the banner b of speecial purposee vehicle, Alpha Micrro Finance Coonsultants Pvtt Ltd (Alpha)) and invested d INR 20 million in eqquity with Higgh Mark. Thiss investment aassured High Mark M that the industryy is committeed to a microfinance credit it bureau and ready to make approppriate investm ments to get it going.

M MFI MFI

Other Investors

Alpha

High Mark

MFI

MFIN N Invests in Alpha

In nvestor

Figure 2: MFIN Invests in a Credit Bureau

Credit Bureau

Institutional Work in Building a Credit Bureau for Microfinance

327

RBI permitted three new credit bureaux: Experian Credit Information Company, High Mark Credit Information Services and Equifax. Both High Mark and Equifax formally launched the CB services after receiving licences from RBI in Dec 2010 and May 2011 respectively. India now has four credit bureaux; two are largely owned by diversified investors (CIBIL and High Mark), while Experian and Equifax are bureaux with an international lineage. 11

2. Role of MFIN in Building Microfinance Credit Reporting

While the credit bureaux got licensed, MFIN had already spent close to six months with both the member MFIs and credit bureaux in preparing for the launch of its services. MFINs’ critical impetus to the microfinance credit bureau(s) in getting operationalized are outlined below: India’s Credit Bureau Timeline 1999 2004 2009 2010 2011

RBI's Siddiqui Committee on Credit Reporting India's first CB, CIBIL, launched RBI's invites application for new credit bureau The three new credit bureaux formed: Highmark, Equifax and Experian Two credit bureaux’ test first: High Mark and Equifax

Table 3: India’s Credit Bureau Timeline 2.1 Design and Process:12 As a first step, MFIN worked closely with credit bureaux in familiarizing them with the microcredit business and the information and data environment in which it operates. During this process, MFIN along with the credit bureaux closely examined the data availability and collection framework amongst the industry players and arrived at standard data points, key identifiers, data collection processes and reporting structure, among others.

Chapter Sixteen

328

Role of MFIN in Building Microfinance Credit Reporting Design and Process

Data

Bureau Design

Availability

Member Commitment Standardization

Automation Reliability

Drive Usage and Promote Adoption Code of Conduct Committee Compliance Committee Monitoring and Evaluation

Pricing, Competition and Scope Best Pricing

Foster Competition Include Non Member MFIs

Table 4: Role of MFIN in Building Microfinance Credit Reporting In parallel, MFIN worked closely with its members to understand their preparedness vis-à-vis participation in the credit bureau and prepared them accordingly. This involved an intense process of engagement including educating, capacitating, preparing and implementing the IT preparedness plan with each and every member. Initially a number of MFIN members were unclear about the benefits of joining the credit bureau and indeed had negative perceptions and a certain reticence on sharing borrower information. MFIN along with the credit bureau organized a number of interactions, collectively and individually with the senior management of the MFIs to educate them on the benefits of credit bureaux for the industry and to get their buy-in. MFIN reached out to its members to explain the Principles of Reciprocity, essentially a set of guidelines governing the sharing of MFI client credit performance and related data with credit bureaux. MFIN is continually collaborating with MFIs and credit bureaux for standardizing reporting formats, and facilitating data collection and upload process. 2.2 Data Not all MFIs have a standard Management Information System (MIS). The data collection and management process among MFIs is varied, including the definitions and the way data is captured. MFIN worked closely with its members in ensuring that all members have a system in place to collect borrower data, as required by the credit bureau. In many instances, this meant changing loan application forms and other data capture sheets at the MFI level and operationalizing the changes.

Institutional Work in Building a Credit Bureau for Microfinance

329

It may be noted that a number of MFIs were using manual record keeping. MFIN worked with such members to move to automated system to be able to share data with the credit bureau. The staffs of MFIs were trained on the data points and software and provided with on-site and off-site technical support to accurately upload the data / interpret queries. MFIN facilitated the CB preparedness plan for all the members and ensured its implementation. This, of course, is an ongoing process as a few members, especially smaller MFIs or ones with manual systems are still in the process of regular data sharing with credit bureaux. 2.3 Driving Usage and Promoting Adoption MFIN, through successive Board directives and Code of Conduct compliance requirements mandated all its members to join the Credit Bureaux and provide data. The terminal date for full compliance on this was 31st March 2012. MFIN also formed a MFIN Credit Bureau Task Force for oversight of credit bureau related matters. The Credit Bureau Task Force closely tracks the members progress vis-à-vis credit bureau regarding sign-up, data submissions and usage. The Credit Bureau Task Force also looks at data submission and report quality issues and provides feedback to Credit Bureau for improvement, on a regular basis. MFIN has mandated all its members to join two credit bureaux, namely, High Mark and Equifax. As of Dec 2012, all MFIN member MFIs are members of at least one bureau, sharing complete loan data with credit bureaux and running credit bureau checks prior to sanctioning new loans with a match rate of over 70% The data sharing process is done on the principle of 'Reciprocity', according to which only such members, who have submitted their credit data, may get access to Credit Information Reports (CIRs) of these credit bureaux. These reports help CIBIL MFIN members in taking objective and quick lending decisions. The first Code of Conduct adopted by MFIN (in April 2010), put a three lender limit and mandated use of a credit report for verification purposes. As a result, the MFIN member’s loan appraisal processes incorporates the use of CB Reports for credit decisions. On Dec 2, 2011, the Reserve Bank of India introduced a new category of NBFCs—‘Non-Banking Financial Company-Micro Finance Institutions’ (NBFC-MFIs) which inter alia was to deal with issues of multiple-lending, over-borrowing and ghost-borrowers. The RBI stipulated that NBFC-MFIs can lend to individual borrowers who are not a member of a Joint Liability Group (JLG)/Self Help Group (SHG) or to borrowers that are members of

330

Chapter Sixteen

JLG/SHG. A borrower cannot be a member of more than one SHG/JLG. Not more than two NBFC-MFIs should lend to the same borrower. There must be a minimum period of moratorium between the grant of the loan and the due date of the repayment of the first instalment. The moratorium shall not be less than the frequency of repayment. For example, in the case of weekly repayment, the moratorium shall not be less than one week. Recovery of any loan given in violation of the regulations should be deferred till all prior existing loans are fully repaid (RBI, 2011). In the absence of the credit bureaux, it would be impossible for MFIs to ensure compliance with the above guidelines. MFINs role was to ensure member committed using credit bureaux. 2.4 Pricing & Competition MFIN negotiated a fair price to ensure that price does not become a deterring factor to joining the Credit Bureaux, especially for the smaller MFIs. The fair price is one tenth of what the country’s leading credit bureau charges its institutional clients. Moreover, the joining fee was kept nominal and staggered based on the asset size. The Credit Bureaux also offered heavy discounts on reports and free trial periods to demonstrate the value of credit reports to MFIs and hence its usage. For instance, CIBIL charges banks and financial institutions US $1 for making a query, whereas High Mark and Equifax charge MFIN member microfinance institutions between US $0.5-$0.10 per query. Given the desirability of greater competition within credit information providers, MFIN has also been in discussion with other credit bureaux to start offering services to MFIs. As a result, the other two credit bureaux, namely, CIBIL and Experian are actively considering entering the microfinance space. Increased competition will ensure better products and services, enhanced capabilities and improved risk management by MFIs. Under the technical support arrangement with IFC, MFIN is now working to obtain NGO-MFIs participation in credit bureaux. So far, 32 NGO-MFIs have signed-up with Highmark. Given the large number of NGO-MFIs, the majority of which are still on manual systems, this will be a focus area for 2015. Collaboration with Sa-Dhan, the Association representing non-profit MFIs, has been instituted for this purpose.

Institutional Work in Building a Credit Bureau for Microfinance

331

3. Issues and Challenges In summary, notwithstanding the substantial progress made in a fairly short span of time, major challenges lie ahead and, unless they are addressed the intended impact of credit bureaux on microfinance, borrowers and lenders will remain limited. 3.1 Widening & Deepening Credit Reporting Coverage A number of key micro-credit market segment lenders remain outside of any credit reporting database, such as NGO-MFIs, Self Help Groups (SHGs), Regional Rural Banks (RRBs) and Cooperatives. Unless the data from these key credit providers is also captured in the credit reporting system, lenders will not have a full picture of borrowing at an individual level. This would also severely limit the efforts of MFIN members to avoid multiple lending and overleveraging of borrowers. In this regard, challenges of credit reporting at SHGs level need special mention. First, there is considerable overlap among SHGs and MFIs borrowers as they target the same individual borrowers. Secondly, data on individual members of a SHG is not being captured or reported to credit bureaux by banks. Thirdly, SHGs data maintenance is, for the most part, inadequate. Fourthly, individual level data of SHG members is not available, so the debt profile of a borrower remains incomplete. Hence, for adequately addressing the issue of multiple lending and over-indebtedness, SHGs need to furnish details of their members to banks and regularly update that data. In turn, the banks need to provide SHG member level data to the credit bureaux. The huge, untapped market at low-income market segments is increasingly attracting mainstream commercial players, especially from those with consumer lending experience. As of now, the two databases, one from microfinance and other for mainstream commercial lending sector are not integrated. Hence, a lender of either category is not aware of the complete debt profile of a potential borrower. While, presently, the overlap between the two market segments is insignificant, going forward, as low income borrowers move up the economic ladder and access mainstream lending, it becomes important that credit databases get shared between credit bureaux.

Chapter Sixteen

332

3.2 Data Quality Issues While both credit bureaux have invested significantly in software and development of ‘matching’ algorithms to match a borrower with a high degree of accuracy, the challenges of data quality remain. Indians do not have a universal unique identity number (UIDAI13) and only recently has the government initiated an ambitious program for providing identification for each resident across the country which would be used primarily as the basis for efficient delivery of welfare services. It would also act as a tool for effective monitoring of various programs and schemes of the Government. Accuracy of data captured on key indicators such as date of birth, name, address, is still uneven and somewhat inconsistent across MFIs. This challenge is more acute for SHGs and NGO-MFIs. With regard to SHGs particularly, it is to be noted that full data on individual members of a group is hardly captured. And, without reliable means of uniquely identifying borrowers, the data available has little or no value.

Recommendations The willingness of a system of private-sector agents to induce compliance with common rules of conduct from each agent, without resorting to exogenous rule-enforcing mechanisms, can either be the result of agents having incentives to undertake behaviour that conforms to the collective interest, or the outcome of agents having an incentive to mutually monitor behaviour (Biagio Bossone & Larry Promisel).14 Three factors led to building the dedicated microfinance credit bureau in India: 1) The need to build credit discipline; 2) The enterprise and cooperation of MFI social entrepreneurs on the principles of reciprocity suitably directed and led by the MFI Industry Association, MFIN; and 3) exemplary world class collaboration between various stakeholders. India managed to build the world’s largest repository of microcredit borrower loan records with multiple credit bureaux, nearing 85 million and growing, in a record 24 months. MFIN played a key role in building microfinance credit information reporting in India.

Institutional Work in Building a Credit Bureau for Microfinance

333

Appendix A: About the organization MFIN With a vision to be an engine of inclusive growth for India and help provide financial services to 100 million low income households by the year 2020, in a responsible and transparent manner, MFIN was established in December 2009 and is the primary representative body and the SelfRegulatory Organization (SRO) for all RBI regulated NBFC-Microfinance Institutions (MFIs) which constitute over 85 per cent of the microfinance business in India (excluding SHGs). The MFIN central secretariat is based in the National Capital Region (NCR; Gurgaon, Haryana), with six state/regional chapters, one each for Andhra Pradesh, Tamil Nadu, Bihar, Maharashtra, Northern Region (Rajasthan, NCR, Haryana, Punjab, Himachal Pradesh, Jammu & Kashmir, Uttarkhand, Uttar Pradesh), and Central Region (Madhya Pradesh, Chhattisgarh).

High Mark High Mark brings proprietary state-of-the-art technology to address nuances inherent to Indian credit data & draws global best practices from international leaders in bureau technology. It also operates world's largest rural and microfinance database, which is poised to be the next frontier of credit growth. High Mark provides cost-effective, flexible and end-to-end bureau solutions to support your credit risk and decision management requirements.

Notes 1 RRB (Regional Rural Bank), MFI (Microfinance Institution), SHG (Self Help Group), SCB(Scheduled Commercial Bank), PAC (Primary Agricultural Cooperative Society 2 Status of Microfinance in India, 2011-12, NABARD 3 A Joint Liability Group (JLG) is an informal group comprising preferably of 4 to10 individuals but can be up to 20 members, coming together for the purposes of availing bank loan either singly or through the group mechanism against mutual guarantee. 4 Status of Microfinance in India, 2011-12, NABARD 5 MCRIL 6 In 2006, in view of the systemic risk arising from access to public funds such as bank borrowings, CPs, etc, by NBFCs, and their interconnectedness with the financial system, the focus of regulatory concern widened to include non deposit

334

Chapter Sixteen

taking NBFCs also. Accordingly, non deposit taking NBFCs with an asset size of Rs. 100 crore and more as per the last audited balance sheet were defined as systemically important, (NBFCs-ND-SI) and a regulatory framework was put in place for them vide Circular No 86 dated December 12, 2006. 7 The YH Malegam Committee Report, RBI, January 2011 8 Arun Thakral (2011) 9 Report of The Working Group to Examine the Role of Credit Information Bureaux in Collection and Dissemination of Information on Suit-Filed Accounts and Defaulters (RBI, 2002) 10 CIBIL 11 Sugandh Saxena 12 Ibid 13 The Unique Identification Authority of India (UIDAI) was established in 2009 by the Government of India, to provide a universal way of uniquely identifying Indian residents in the form of AADHAAR, a 12 digit unique identification number(UID) that will be provided after getting the demographic and biometric information of an individual. 14 Biagio Bossone and Larry Promisel.

BIOGRAPHICAL PROFILE OF AUTHORS

Dr. Sharam Alijani is Professor of Strategy and Entrepreneurship at NEOMA Business School and a member of the research team of ICEAMS (NEOMA) and HABITER (Reims Champagne-Ardenne University). His teaching and research activities focus on topics of social innovation, entrepreneurship, microfinance, sustainability and urban studies. He has served in different executive capacities in the industry and taught courses as visiting faculty in France, China and the United States. His latest article entitled “Entrepreneurial Capability and Leadership” has appeared in Encyclopedia of Creativity, Invention, Innovation, and Entrepreneurship (Springer, 2013). He holds a Ph.D. in Economics from the University of Paris Est, France. Sunder Annamraju specializes in systems and technology within the Financial Services industry. He has guided and managed several technology-enabled change programs in large-scale environments within top-tier global banks like Citibank, UBS and Deutsche Bank. He is a graduate of the Indian Institute of Technology, Delhi and the Indian Institute of Management, Calcutta. He holds a Master’s degree in Knowledge Engineering from the National University of Singapore. He is a Chartered IT Professional of the British Computer Society and a member of British Mensa. His current interests are in the area of Serious Games and he operates a consultancy in this field. He can be contacted at [email protected]. Vibhu Arya is the founder of The Flat Pyramid, an inclusion and ‘markets for the poor’ consulting practice. The Flat Pyramid is aimed at architecting inclusive public-policy and resource allocation decisions within political, economic, and social systems and institutions. His work spans across Microfinance, Housing Microfinance, Branchless Banking, and Impact Investing. He has completed assignments for Ayani, The Microfinance Institutions Network (MFIN), GIZ, ShoreBank International, Access Development Services and Habitat for Humanity. Vibhu has previously worked for a decade with Baxter, BHP Billiton, General Electric and Citibank, where he managed the microfinance institution’s lending business for North, East and West India.

336

Biographical Profile of Authors

Dr. Arvind Ashta holds the Banque Populaire Chair in Microfinance at the Burgundy School of Business (Groupe ESC Dijon-Bourgogne), France. He offers courses in Microfinance and researches institutional aspects of Microfinance, technology in Microfinance, and CSR. He has taught Microfinance as visiting faculty in Chicago, Pforzheim, Brussels, Barcelona and Hertfordshire. He has edited a book on Advanced Technologies for Microfinance. He has a number of publications in international journals and guest edits special editions of various journals devoted to microfinance. He is on the editorial review board of Cost Management, Strategic Change, and International Journal for Technology and Human Interaction. He can be contacted at [email protected]. Dr. Djamchid Assadi is Professor and member of the research team of the Banque Populaire Chair in Microfinance, Burgundy School of Business, Dijon, France. He is a specialist in (online) strategy and marketing. His research focuses on the impact of non-economic factors on the buying behavior and strategic behavior and “peer-to-peer” relations including “social lending.” He teaches the course “Microfinance Strategy and Marketing.” He has taught at several universities in France and the United States. He has written many books, several book chapters, articles and papers presented at numerous conferences. He holds a Ph.D. in Marketing Strategies and Communication from the University of Paris at Dauphine, Paris, France. Dr. Bryan Barnett is an independent microfinance consultant and serves as a banking adviser with the Office of Technical Assistance of the U.S. Department of the Treasury. He has completed projects for USAID, the Bill and Melinda Gates Foundation, the Grameen Foundation and the Microfinance Information Exchange (MIX). He previously served in various roles at Microsoft Corporation, as an analyst and project manager with investment firm Vulcan Northwest, and was a founder and VicePresident of ApexLearning, a pioneering online learning company. Vitalie Bumacov is currently conducting a doctoral research with Oxford Brookes University, UK on the use of poverty scoring and credit scoring in micro lending. He also works as consultant in access to agriculture finance for Millennium Challenge Account Moldova. His microfinance career started in Colombia, but most of his assignments have concerned MFIs in Eastern Europe and Asia.

Management Information Systems for Microfinance

337

Mikhail Cherkas is a Partner with MF Strategy, a management consultancy focusing on financial services and development finance. He previously worked as a consultant for Internationale Projekt Consult (IPC) with the EBRD’s Russia Small Business Fund, and as a senior manager with Citibank, Renaissance Capital and KMB. He was also General Manager of KMB Leasing, which he developed into one of the largest SME leasing providers in Russia. Mikhail holds an M.B.A. from IMD, Lausanne; a Master’s degree in Business from Novosibirsk State Technical University; and a B.Sc. in Technical Trade from the Harbin Institute of Technology, China. Dinos Constantinou is the Managing Partner of MF Strategy, a management consultancy focusing on financial services and development finance. He previously worked in strategy consulting with Gemini Consulting; SME finance with IPC; and in banking with Barclays and Dresdner Bank. Dinos holds an M.B.A. from IMD in Lausanne, an M.A. in European Economic Studies from the College of Europe in Bruges, and a B.Sc. in Economics from Bristol University. He has written on corporate finance and business strategy in areas such as technology, energy and financial services and is currently working on a book on corporate governance. Dr. Mawuli Couchoro is Senior Lecturer at the University of Lomé in Togo. He holds a Ph.D. in Economics from the University of Poitiers. His research generally focuses on the microeconomics of development, with microfinance as one of its applications. He is a member of the Centre for Training and Research in Economics and Management (Centre de Recherche et de Formation en Sciences Economiques et de Gestion (CERFEG)), University of Lome, a member of the design and management committee for the Masters Program in Microfinance at the Catholic University for West Africa in Togo, and associate researcher with the Chair in Microfinance, ESC-Dijon (France). Professor Karl Dayson is the Director of the Salford Institute for Public Policy at the University of Salford, UK. A sociologist by training, his research interests include microfinance in Europe, financial exclusion and inclusion, credit, money and social businesses. He has helped establish over a dozen community-owned financial institutions in the UK and advised the UK government, local authorities and housing providers. He is the co-author of the European Code of Conduct for Microcredit Providers issued on behalf of the European Commission and is currently designing

338

Biographical Profile of Authors

the Code's implementation and compliance processes. He tweets @KDayson predominantly on microfinance governance and performance issues and can reached at [email protected]. Dra. Glòria Estapé-Dubreuil is Professor at the Department of Business, Universitat Autònoma de Barcelona, Spain, and member of the Business Efficiency and Competitiveness Research Group. Her basic degree is in Mathematics, and her Ph.D. is in Economics and Business Administration. She has also had a number of years of experience working with third sector organizations. Her present research interests lie in the areas of microfinance, third sector organizations management, and the development of models and applications of quantitative analysis to decision-making issues. She has a number of publications both in national and international journals, and actively participates in international conferences. Dr. Marc Ingham is a Professor of Strategy and Innovation and holds the Chair in Responsible Management and Innovations (Management et Innovations Responsables) at the Burgundy School of Business (Groupe ESC Dijon- Bourgogne). He is a research associate at CRECIS (Louvain School of Management) and he is or has been a professor, visiting professor or lecturer on four continents. His current research interests cover the strategic integration of responsible management and innovations. He holds a degree in Economics from the Université Catholique de Louvain and a Ph.D. in Management Sciences from the Université de Paris Dauphine. Dr. Jacques Bongolomba Isoketsu was born on September 5, 1960 at Basankusu in Equateur province in the Democratic Republic of Congo. He obtained a Ph.D. in demography in 2011, at the University Paris 1 Panthéon-Sorbonne. He holds several masters degrees: in Ethics and Organization; Demography and Population Dynamics; Social and Societal Audit, Tax Administration; and Development Studies. He has served as Tax Auditor, Assistant to the Higher Institute of Management (DRC), and consultant to social and international organizations. His research focuses on microfinance, poverty, economic analysis, population aging, pensions, family structures, taxation of business, ethics and organizations, and corporate social responsibility.

Management Information Systems for Microfinance

339

Saleh Khan is an experienced international development and microfinance professional. Over the last decade he has worked with some of the leading global development organizations to formulate, manage and evaluate large scale projects in countries ranging from South Asia to West Africa. He was responsible for managing two top-tier microfinance institutions in Nigeria and Ghana on behalf of a large private equity investment fund. Saleh currently works in the International Development practice of PricewaterhouseCoopers (PwC) and is part of the firm’s global development consulting effort. He helps drive PwC’s global microfinance advisory services. Dr. Sudeep K. Krishnan is a Fellow of the Indian Institute of Management, Ahmedabad (IIMA). IDEA Telecom Centre of Excellence Ph.D. Fellowship, IIMA Best Thesis Proposal, and Industrial Finance Cooperation of India Thesis Proposal Award were conferred to him for his thesis on open innovation in the IT sector. He has also presented papers at international conferences, doctoral colloquiums, participated in open innovation European projects, and was selected as Leader of Tomorrow at the 42nd St. Gallen Symposium. He currently works as Senior Manager with the Decision Analytics team of EXL Service Holdings. His interests include Business Analytics, Open Innovation, and IT Strategy & Design. Dr. Raghavan Kunigahalli has more than 25 years of experience working with software, educational, banking and financial services institutions. Currently, Dr. Kunigahalli is a Corporate Enterprise Architect at Navy Federal Credit Union. Dr. Kunigahalli served for several years as a Vice President and Information Officer at American International Group (AIG), and as a lead systems architect at the Bank of New York. Dr. Kunigahalli received his doctoral degree from the University of Wisconsin–Madison. Dr. Kunigahalli has authored numerous publications including newspaper articles, journal and conference papers and handbook chapters. He has also delivered several lectures and presentations in international conferences on similar topic areas. Frédéric Lanet studied Economics in Montpellier, France. Since 2000 he has worked for Airdie, a French solidarity-based funder (MFI) that is member of France Active, a local network that has helped poor people create businesses over the last 25 years. He started as a micro business financial analyst, before he was made responsible for financing social businesses. Currently, Frédéric is deputy director of Airdie. He is also

340

Biographical Profile of Authors

president of the Association “APIJE” which helps unemployed people find work. Nitin Madan holds a Master’s degree in Poverty and Development from the Institute of Development Studies (University of Sussex). He has over 10 years of management and consultancy experience in microfinance. He was part of the founding team of Rajasthan Shram Sarathi Association, a reputed financial inclusion action research project for seasonal migrant laborers in South Rajasthan (India). He has been involved in providing technical support to MFIs on operations, product and process design and re-engineering. He currently consults with MFIs globally on Social Performance Management. Abu Saleh Mohammad Musa is the Team Leader of MicrofinanceEXPOSURE, a virtual knowledge-sharing platform for microfinance practitioners. He combines strong analytical, financial and management skills with expertise in participatory research, financial management, product development, technology innovation, microenterprise and business development. He is good at providing training, program oversights including budget management, and developing technical and financial proposals for projects targeting to improve the livelihoods of the poor. He is a Certified Service Provider (CSP) for MicroSave in Market Research for Microfinance Toolkit. Mr. Musa has completed his M.B.A. in Finance from the University of Dhaka, Bangladesh. Krishna Nyapati is a Director of Microsense Pvt Ltd, an Internet services company based in India, and also serves as a Board Member of the Centre for Budget and Policy Studies, a think-tank based in Bangalore. He holds a degree in Engineering from the Indian Institute of Technology, Madras and a graduate diploma in Management from the Indian Institute of Management, Kolkata. His areas of interest include Management Information Systems, software development, Quality Management, sustainable technology and public policy, and he has been active in teaching, publishing and making presentations at several international conferences related to these areas. He serves as the Chairman of TIDE, which received the Ashen Global Energy Champion Award during 2008. Dr. Debdatta Pal teaches Economics at Indian Institute of Management Raipur. He is a Fellow (Agriculture) of Indian Institute of Management Ahmedabad. Before joining the doctoral program, Dr. Pal served at Indian

Management Information Systems for Microfinance

341

Bank, a nationalized scheduled commercial bank, for three years as Manager looking after rural credit portfolios. His current research interests include understanding rural credit market imperfections in an agricultural household framework, empirical assessment of varying access to institutional finance by the rural households, impact assessment of development initiatives and choice modeling. Rupal Patel holds an M.B.A. in Finance and Non Profit Management from Rutgers University School of Business in the U.S. Rupal has taught social entrepreneurship at Rutgers University and has worked on the development of a water entrepreneur initiative in Gujarat. She has been actively involved in providing handholding support to MFIs on social performance management, from implementation of poverty assessment and design of reporting systems, to data analysis. Her work in the financial inclusion space ranges from sector assessments, conducting financial ratings, to providing training. Prior to this, her experience included consultancy and audit in the corporate sector. Alok Prasad is a veteran banker with over 30 years of banking and financial services experience. He is currently the CEO of MFIN, the primary representative body and the Self-Regulatory Organization (SRO) for the Reserve Bank of India (RBI) regulated NBFC-Microfinance Institutions NBFC-MFIs. He was formerly the Country Director of Citi Microfinance Group (India) and the Head of Strategy, Business Development for the Global Consumer Group Citibank India. Prior to joining Citibank Mr. Prasad served with distinction in the RBI for over 15 years. He was also a member of the start-up team of the National Housing Bank, where he played a pivotal role in the formulation of policies for the development of the housing finance sector in India. Dra. Ma. Rosa Rovira holds a Ph.D. in Business Administration. She is currently Associate Professor at the Business Department, Universitat Autònoma de Barcelona, Spain. She teaches financial accounting, environmental business accounting and reporting, and environment management tools. Her research areas include corporate social responsibility (CSR), sustainability reporting, Global Reporting Initiative (GRI), environmental management, as well as the Eco Management and Audit Scheme (EMAS). She has been appointed as project coordinator in several projects studying the integration of CSR issues through the process of producing sustainability reports (according to GRI G3) or the

342

Biographical Profile of Authors

applicability of the GRI-G3 sustainability reporting—a sector supplement for non-profit organizations. Sandeep Mysore Seshadrinath is a doctoral student at the School of Business and Economics at Loughborough University, UK. Sandeep’s work focuses primarily on organizational emergence and legitimacy, and innovation. Prior to starting his doctoral studies, Sandeep worked for a policy think-tank with a research focus on decentralized e-governance and rural infrastructure projects. His experience includes an internship with a micro-finance institution and employment with a global software services firm. He holds a B.Eng degree in Information Science Engineering and a Postgraduate Diploma in Rural Management. Professor Satchidananda Sogala, Ph.D. is Director of Chidambara Research, a global think-tank on development and also the Founder and CEO of Srichid Group. He has more than 35 years of experience in the financial domain and in technology solutions, having worked at Reserve Bank of India (Regional Director), International Institute of Information Technology (Research Director), IBM and HP (Head of Risk Research). With an M.S. in Financial Markets (University of Illinois at U-C) and a Ph.D in Economics, Dr. Satchidananda is also the originator of Ganaseva, the innovative technology platform for the delivery of financial & other services, and a comprehensive solution for watershed management. He has authored over 20 papers and holds several items of IP in watershed management, health care and health insurance management. He is also the President of the World Association of Risk Management Practice and Research. Frances Sinha is co-founder of the development consultancy EDA Rural Systems and the specialist rating agency M-CRIL, based in India. A graduate from Oxford University and the London School of Economics, Frances has led teams and consultancies for research, evaluation, impact assessment and training, throughout Asia and Africa. She has been closely involved with the development of social performance and double bottom line management and reporting in microfinance, including social performance management and training with the Imp-Act consortium, social performance assessments and ratings with M-CRIL, and social performance reporting with country networks, the MIX and the Social Performance Task Force.

Management Information Systems for Microfinance

343

Gaurav Sinha has a decade-long experience of working on policy and operational aspects of improving livelihoods and enabling access to financial services for the poor. He has undertaken independent assignments supported by RBS (formerly ABN-AMRO), SIDBI, etc. on various aspects of institution development, including incubation of small organizations in microfinance. He has conducted research and assessment studies for many organizations including IFC, Tata-Trusts, C-QuestCapital, and Water.org. Moreover, he has designed and delivered educational programs on microfinance for NABARD & IIM and has several publications to his credit. Gaurav is a certified trainer of MicroSave on several microfinance toolkits. Currently, he works with Azim Premji Foundation. Aishwarya Srinivasan is a full-time M.B.A. student at the W P Carey School of Business at Arizona State University in the U.S. with focus on Financial Markets and Management. Prior to starting her graduate studies, Aishwarya worked for the Essar Group, one of India's largest business conglomerates, for two and a half years as Manager in the corporate finance and treasury department handling fundraising and risk management activities. She has also undertaken an internship with Equirus Capital Pvt Ltd, a mid-sized investment bank, and worked for a software services firm. She holds a B.E degree in Electronics Engineering from University of Mumbai, India and a Postgraduate Diploma in Management with a finance and marketing focus from Institute of Management Technology, India. Rakesh Sud, from St Columba’s School, Delhi, is a graduate in Economics (Hon) from St. Stephens College, Delhi, an associate member of the Institute of Chartered Accountants of India and a graduate member of the Institute of Cost Accountants of India. He has passed his CISA, DISA and CISSP Exams and is currently doing a Ph.D. He is Director at ACMC (Acharya Center for Management Consultancy) at AIMS. He has had over three decades experience in employment, entrepreneurship, teaching and consultancy. He has worked in Nigeria, Tanzania and UK and India. He is active on LinkedIn with 30,000 first level connections. Godfrey Supka is the Global Operations Manager of Fern Software, the microfinance information systems specialist based in Belfast, Northern Ireland. Founded in 1979 by Eamon Scullin, Fern now has over 300 microfinance and credit union clients in over 30 countries. Godfrey has more than 20 years experience in microfinance and banking systems,

344

Biographical Profile of Authors

mostly in East Asian and Middle Eastern markets. Godfrey is the author of the IBS Intelligence review on Microfinance Information Systems. He is a graduate of the Boulder Microfinance Program and has an M.B.A. from Henley Business School. Dra. Consol Torreguitart-Mirada is Professor at the Department of Business, Universitat Autònoma de Barcelona, Spain. She has a M.B.A. in Political Science and a Ph.D. in Economics and Management. She offers courses in the area of organization and business administration. She has researched in the fields of ethical finance, microfinance institutions and the social economy. Her research interests include gender studies, the use of the ITC in the enterprises of the social economy, and the impact of microcredit in the real economy. She has a number of publications in international journals and she has contributed to several chapters of books.