Efficiency of Microfinance Institutions in India: Are They Reaching the Poorest of the Poor?

Transcription

1 Article of Institutions in India: Are They Reaching the Poorest of the Poor? Vision 20(1) MDI SAGE Publications sagepub.in/home.nav DOI: / Prabhjot Kaur 1 Abstract institutions are being criticized for their shift in the objective of serving the poor to the profit-making institutions. Serving the poor is a relatively costlier issue. Thus, microfinance institutions are being blamed for serving the better off clients. The focus of this study is to look at the financial efficiency of microfinance institutions in India and analyse their efficiency to reach women and the poorest of the poor. Application of Data Envelopment Analysis shows, on an average financial efficiency of Indian MFIs is much higher than their social efficiency. However, study does not find evidence for the presence of trade-off between financial and social efficiency. Key Words Andhra Pradesh microfinance crisis, Data Envelopment Analysis, of microfinance institutions Introduction institutions (MFIs) are losing the faith of poor and, particularly poor women, as they are progressively heading towards the commercialization of their operations. It is, therefore, immensely important that we measure social and financial efficiency of these institutions and analyse what went wrong with them. Andhra Pradesh microfinance crisis, 2010 is seen as biggest tragedy in the history of microfinance in India. Crisis was triggered by the events like over indebtedness among the clients, usurious rate of interest charged by MFIs, unethical collection behaviour of the staff and mounting suicides tendencies among the borrowers. Thus, after the microfinance crisis in Andhra Pradesh, which eventually spread to almost all other states, analysing the efficiency of MFIs in Indian context has become all the more important. In the past, studies have been conducted to measure the efficiencies of MFIs operating in different regions like Latin America, Asia, Africa, MENA region etc. (Gutiérrez-Nieto, Serrano-Cinca, & Molinero, 2007; Haq, Skully, & Pathan, 2010; Hassan, Sanchez, & Ngene, 2012). In a recent study, Lebovics, Hermes, and Hudon (2015) have measured the financial and social efficiency of Vietnamese MFIs and confirm absence of trade-off between social and financial objectives. There is clearly dearth of studies that have addressed the issues related to social and financial efficiency of MFIs operating in India. One study that has emerged recently in Indian context is done by Kumar and Sensarma (2015). Their study has analysed the efficiency of Indian MFIs during the period They find Indian MFIs face trade-off while reaching the poor; however, no such trade-off is observed while reaching the women. In this study, an attempt has been made to gain insight of how MFIs in India are performing on financial and social efficiency front after the microfinance crisis. For the purpose of analysis, latest and complete data were available for the financial year Thus, year has been chosen for the achievement of the objective. This study also aims to provide benchmarking of MFIs so that inefficient MFIs can identify the peers whose practices can be followed to improve upon their efficiency. The rest of the article is structured as follows. The second section gives briefing of microfinance crisis in India. The next section reviews the debate on trade-off between outreach and sustainability of MFIs and literature on efficiency of MFIs. The fourth section gives model specification and measurement of efficiencies for different specifications. Results and 1 Faculty of Management Studies, Delhi, India. Corresponding author: Prabhjot Kaur, Faculty of Management Studies, Delhi, India. prabhjot.fms@gmail.com

2 Kaur 55 interpretation of DEA and conclusion are given in the fifth and sixth section, respectively. in India In India, as per the legal status, mostly non-banking financial companies/intermediary (NBFCs/NBFIs), and nongovernmental organizations (NGOs) are providing microfinance services. Credit unions, cooperatives and rural banks are other institutions that provide microfinance services to the clients. In 2010, before the outbreak of microfinance crisis in Andhra Pradesh, Indian microfinance sector was the largest in the world. With the growth of 62 per cent per annum in terms of number of clients and with around 27 million accounts, Indian microfinance sector was the largest in the world. However, microfinance crisis in Andhra Pradesh in 2010 has reversed the situation. Increasing rate of suicides of borrowers led to enactment of the Andhra Pradesh Institutions (regulation of money lending) Act, The Act restricted MFIs from collection of loan instalments from door to door and asked them to conduct monthly meeting instead of weekly. Use of coercive methods of collection was made punishable. Message that was perceived by the borrowers was that they need not to repay the loan, and in any case, state government is there for them to rescue. Act affected micro-lenders profitability, loan recovery and turned their operations unviable. The crisis in the microfinance sector has left some microfinance companies with negative net worth, due to which they are unable to borrow from the banks and hence, their credit disbursals are affected by the crisis. Banks were also affected by the crisis as 80 per cent of loans borrowed by the MFIs were from banks only. Funding constraints, negative perceptions combined with higher operating costs are posing challenges to the growth of the industry. Structure of Indian microfinance is also changing, as there is wide trend of transformation of NGOs into for-profit NBFCs (M-CRIL, 2010). In this study for the fiscal year 2012, out of 87 MFIs studied, 52 are NBFCs, that is on an average approximately 60 per cent of the sample comprises NBFCs. For large loans and investments, commercial banks and equity investors prefer NBFCs to NGOs, which eventually has resulted into this transformation trend. crisis in Andhra Pradesh and its spill over effects on others states has brought the sector under intense scrutiny. Against the backdrop of microfinance crisis The Micro Finance Institutions (Development and Regulation) Bill, 2012 was tabled in the national Parliament on 29 May However, the Parliamentary Standing Committee on Finance scrapped the bill in February 2014, on the ground that it is sketchy and lack consensus and consultations with the stakeholders. So now, all eyes are on the drafting of the new microfinance bill to protect the interests of the clients and to provide better regulatory environment for the operations of MFIs in India. Sustainability versus Outreach of Institutions With the incidents like Andhra Pradesh and increasing commercialization of microfinance operations debate on sustainability and outreach of MFIs in India is also gaining ground. Financial sustainability approach or institutional approach focuses on the sustainable operation of MFI by charging reasonable rate of interest to cover the costs of lending. This approach emphasizes on increase in revenues from interest income and fee and cutting down the operational cost. Institutional approach asserts financial sustainability holds the key to serve the large number of poor. On the other hand, the poverty lending or welfarist approach supports loans to the poor at the rate affordable to them. Lending to poor is a relatively costly affair, as they frequently need loan of small amounts. Thus, as per this approach, reaching the poorest of the poor, as better known as the depth of outreach and achieving sustainability goes against to each other. In the microfinance literature, this phenomenon is known as mission drift and trade-off between social and financial goals of MFIs. Schreiner (2002) describes as depth of outreach and financial sustainability as two polar opposite targets since depth of outreach implies increased service to the poor, while emphasis on financial sustainability may force MFIs to either cut back on the disbursement of smaller loans or rely more on subsidies. It has been argued that since poorest clients need frequent loans involving small amounts, thus it is relatively costlier to serve the poorest. However, Gibbons and Meehan (1999) contend that even among the poorest loan clients, average loan size tends to increase considerably over the years depending on their repayment history. They refute the trade-off between outreach and sustainability and assert that financial sustainability depends on the strength of the financial programme on the type of clients served. Quayes (2012) analyses 702 MFIs from 83 countries. He categorized MFIs into high and low disclosures firms and showed a positive complementary relationship between financial sustainability and depth of outreach for the high-disclosure MFIs. Thus, this study negates the trade-off between financial sustainability and depth of outreach; however, trade-off exists between breadth of outreach and financial sustainability. Whereas Fajonyomi, Jegede, and Akinlabi (2012) analyse 80 MFIs from the period from southwestern Nigeria and they confirm strong positive relationship between sustainability and breadth of outreach. In this study, trade-off is measured by considering their efficiency to reach the poorest of the poor and efficiency to realize their financial goals. of MFIs in India in post-crisis period is measured using Data Envelopment Analysis (DEA). of Institutions and Data Envelopment Analysis of production units lies in the judicious use of inputs to produce maximum output. In context of MFIs,

3 56 Vision 20(1) Balkenhol (2007) has defined efficiency in microfinance as how well an MFI technically transforms inputs (such as assets, staff and subsidies) to produce the maximum outputs (such as number of loans, financial self-sufficiency and poverty outreach). They emphasized since MFIs are performing both financial and social functions, hence, efficiency should be measured from both the perspectives. Conventionally, ratio analysis has been used widely to measure the efficiency of financial institutions. However, inability of ratio analysis to capture the multidimensional aspects of financial activities adequately, encouraged shifting the focus on some better techniques. Ratio analysis hardly gives information on the best practice and benchmark and further, scale of economies cannot be incorporated in the ratio analysis (Athanassopoulos & Ballantine, 1995; Worthington, 1998). of MFIs can better be measured by using parametric techniques like Stochastic Frontier Approach, Distribution Free Approach and Thick Frontier Approach; and non-parametric techniques like DEA, Free Disposal Hull etc. Berger and Humphrey (1997) conducted a very comprehensive study and tried to cover almost all the studies accessible to them at that time, which used frontier analysis to measure the efficiency of financial institutions. Out of 130 studies analysed by them, 62 studies applied DEA. Balkenhol (2007) also supports suitability of DEA for measuring the efficiency of MFIs as objectives of MFIs vary from focusing on the poor to profit making. DEA is a mathematical programming-based approach for measuring relative efficiency of decision-making units (DMUs) that have multiple inputs and outputs (Charnes, Cooper, & Rhodes, 1978). 1 One major advantage of DEA is that it does not require assumption about the form of production function. Gutiérrez-Nieto et al. (2007) used model specification methodology suggested by Serrano-Cinca and Mar Molinero (2004), and used 21 specifications under DEA to calculate efficiencies of MFIs for every possible combination of two inputs; credit offices and operating expense; and three outputs; interest and fee income, gross loan portfolio and number of loans outstanding. They further used Principal Component Analysis with Property Fitting to understand why a MFI achieves a level of efficiency under a given specification. Haq et al. (2010) used DEA to measure the efficiency of 39 MFIs across Africa, Asia and Latin America. Considering the fact that uncontrollable variables may lead to waste and mismanagement of resources, they incorporated percentage of rural population as an uncontrollable social economic factor in the second stage of their analysis. Their study found NGO MFIs to be most efficient under production approach and hence, suggested promoting these MFIs in the developing countries as they were achieving outreach and sustainability objective simultaneously. However, some studies (Annim, 2012; Nghiem, Coelli, & Rao, 2006) calculated and compared the efficiency of MFIs by using both DEA and Stochastic Frontier Analysis. Model Specification For the purpose of this study, data have been collected for the Fiscal Year 2012 for 87 MFIs operating in India from the MIX Market database. For the year 2012, data for 91 MFIs were available but, after omitting the MFIs with missing fields, only 87 MFIs have been studied. MFIs are expected to reach the poor (outreach) while being financial sustainable, though findings of several studies highlight that achieving both objectives simultaneously is not easy. Studies (Cull, Demirgu c_-kunt, & Morduch, 2007; Hermes, Lensink, & Meesters, 2011; Olivares-Polanco, 2005) find a trade-off between outreach and sustainability. Considering the conflicts between financial and social objectives, in the present study, selection of variables is made in a way to reflect both financial as well as social efficiency of MFIs in India. Variables are selected after going through the literature on social and financial efficiency and debate on using intermediation and production approach under DEA. The intermediation approach uses outputs measured in dollars, whereas in production approach labour and capital are generally used as inputs to generate deposits and loans as outputs (Avkiran, 1999). In India, not all MFIs are allowed to accept deposits. Thus, intermediation approach that uses deposits as input to produce loans cannot be used here. Selection of inputs and outputs has been made on the lines of studies done by Flückiger and Vassiliev (2007) and Guitierrez-Nieto et al. (2009). In this study, to measure the efficiency of MFIs three inputs used are total assets, operating cost and number of loan officers. Financial efficiency is calculated using two outputs: gross loan portfolio and revenue generated by the MFIs, as success of generating these two outputs depends on how efficiently MFIs are utilizing their financial resources. To know the sustainability, it is imperative to know the costs, thus operating cost and number of loan officers were taken as two inputs. Bassem (2008) considered number of women clients as a proxy for the breadth of outreach. Women are generally poorer compare to their male counterpart, thus, the number of women reached is considered as one of the index for measuring social output/efficiency. Further, poverty index as suggested by Guitierrez-Nieto et al. (2009) has been constructed and, is multiplied by number of active borrowers of the respective MFIs to assess the number of poorest clients MFIs are serving. Hence, the number of poorest clients reached by MFIs is another index used to measure social efficiency of MFIs. ALBi - Min(ALB) P i = 1- Range(ALB) where P i is poverty index for ith DMU/MFI, ALB is average loan balance per borrower, Min (ALB) is the minimum value over all i, Range (ALB) is the maximum value of ALB over all i minus the minimum value of ALB over all i. Annexure 3 gives the description of various inputs and outputs used in the study.

4 Kaur 57 For the ith DMU, out of j DMUs, using x n and y m inputs and outputs, respectively, the input-oriented technical efficiency (TE) denoted by i under constant return to scale is obtained by solving the following linear programming, suggested by Charnes et al. (1978) as Subject to j j = 1 TE i Min, = i m m y - y 0, m= 1,... M j mj mi $ j i x - m x $ 0, n= 1,... N i ni j nj j = 1 m 0, j= 1,... J j $ i i d(0,1] As indicated by Gutiérrez-Nieto et al. (2007), efficiency of MFIs varies with the specifications of input and output variables, in this study, six specifications have been considered by combining three inputs with four outputs as shown in Table 1. Description of inputs and outputs is given in Annexure 3. Further, Table 2 summarizes the descriptive statistics of inputs and outputs used in the study. Performance of key MFIs on output variables is shown in Figures 1 4. i i i Results and Interpretation calculated from financial perspective under financial efficiency specification by combining all the three inputs with two outputs, gives average efficiency of approximately 84 per cent. Hence, under this specification inputs can be reduced from the present level by 16 per cent while maintaining the same level of output. Out of 87 MFIs, 11 MFIs, viz. Bandhan, Disha Microfin, FFSL, MMFL,, SEWA Bank, SML, Sanghamithra, Trident and WSE have scored perfect 1, and hence, are relatively efficient on the financial front (Table 3, column 3). HiH and JFIL have the lowest financial efficiency score of 0.30 and 0.34, respectively. Counting the number of times efficient MFIs is referred to as benchmarks/peers for other inefficient MFIs, Muthoot scored the highest, that is 68 times followed by FFSL 33 times (see Annexure 1). The DMU that appears in the references sets more than others do, known as the global leader (Avkiran, 1999). Hence, under financial efficiency specification Muthoot turned out to be the global leader. Considering the combination of three inputs and just revenue as output, named as revenue efficiency specification, in Table 3, eight MFIs scored perfect 1. Trident is 100 per cent efficient under financial efficiency specification, but shows the lowest score in generating revenue with mere 13.6 per cent efficiency. Similarly, SML is financially efficient, but efficiency in generating revenue is just 54 per cent. Table 1. Specifications of Used Under DEA Model from Financial Perspective Inputs Outputs a. Financial efficiency Total assets, operating cost, number of loan officers Revenue, gross loan portfolio b. Revenue efficiency Total assets, operating cost, number of loan officers Revenue c. Loan efficiency Total assets, operating cost, number of loan officers Gross loan portfolio from Social Perspective a. Social efficiency Total assets, operating cost, number of loan officers Number of active women borrower, number of the poorest reached b. to reach women Total assets, operating cost, number of loan officers Number of active women borrowers c. to reach the poorest Total assets, operating cost, number of loan officers Number of poorest reached Source: Author s own. Table 2. Descriptive Statistics of Inputs and Outputs Inputs and Outputs Minimum Maximum Mean Standard Deviation Assets 440, ,599, ,985, ,685, Operating cost 21, ,786, ,047, ,494, Loan officers 4 8, , Gross loan portfolio 185, ,457, ,129, ,387, Revenue 61, ,825, ,998, ,492, No. of women borrowers 1,310 4,433, , ,629 No. of poorest reached 224 4,094, , ,651 Source: Author s calculation based on the data collected from MIX market database (

5 58 Vision 20(1) Performance of Key MFIs on output variables for the financial year 2012 Figure 1. Gross Loan Portfolio ($) Figure 3. Revenue ($) Source: Mix market database. Figure 2. No. of Female Borrowers Figure 4. No. Poorest Reached Analysing loan efficiency, SEWA Bank, and WSE are relatively efficient in generating loans. Whereas Muthoot, Bandhan, Disha Microfin and SCDC, Sanghamithra, MMFL were efficient financially and in revenue generation but not in placing loans, their loan efficiency is 36.9, 39.9, 44, 49.8, 59.7 and 60.7 per cent, respectively. Now turning attention to social perspective gives miserable results for MFIs in India. Overall, average social efficiency of MFIs in India is 32.5 per cent. Their average efficiency to reach women and the poorest is low at 32 and 30 per cent, respectively. As against, 11 efficient MFIs under financial efficiency specification, there are just two MFIs that are on efficient frontier under social efficiency

6 Kaur 59 Table 3. Score of MFIs Under Six Specifications MFI/DMU Name (1) Legal Status (2) Financial (3) Social (4) Revenue (5) Loan (6) to to Reach Reach Women the Poorest (7) (8) AML NBFI ASA India NBFI Adhikar CU/CO* Annapurna Mahila Credit NBFI Co-op Society Annapurna NBFI Pvt. Ltd Arohan NBFI Arth NBFI Asirvad NBFI Asomi NBFI BASIX NBFI BFL NBFI BJS NBFI BMVS NBFI BSS NGO Bandhan NGO Belstar NBFI CDOT NGO CMML NGO Cashpor MC NBFI Chaitanya CU/CO DBIS CU/CO DCBS CU/CO Disha Microfin NBFI ESAF NBFI Equitas NBFI FFSL NBFI Fusion NBFI GFSPL NBFI GMF NGO GMSSS NGO GU NGO GUARDIAN NGO GVMFL NGO Grameen Sahara NBFI HiH NGO IDF Financial Services NBFI IMPACT NGO IRCED NGO JFIL NBFI JFSPL NBFI Jagaran MF NBFI KBSLAB Rural Bank LBT NGO M-power NGO MFS NGO MMFL NBFI Mahasemam NBFI Mahashakti NBFI Muthoot NBFI NEED NBFI NFPL NGO Navachetana NBFI Nightingale NGO PRAYAS NGO RGVN(NE) NBFI (Table 3 continued)

7 60 Vision 20(1) (Table 3 continued) MFI/DMU Name (1) Legal Status (2) Financial (3) Social (4) Revenue (5) Loan (6) to to Reach Reach Women the Poorest (7) (8) SCDS NGO SCNL NBFI SEWA Bank NBFI SKDRDP NBFI SKS NGO SMCS NGO SMGBK NGO SMILE NBFI SML NGO SMPL Other SNFL NGO SVCL NBFI Sahara Utsarga NGO Sahayog NGO Saija NBFI Samasta NBFI Sanghamithra NBFI Sarala NBFI Seba Rahara NBFI Sonata NBFI Spandana NBFI Suryoday NBFI Swadhaar NBFI NBFI UFSPL NBFI Ujjivan NBFI Utkarsh NBFI VFS NBFI WSDS NGO WSE NGO YFS NBFI YMF Other AVERAGE Source: Author s calculation based on DEA. *CU/CO refers to Credit Unions and Cooperatives. specification and 92 per cent of the MFIs have registered efficiency below 50 per cent. Bandhan, Disha Microfin and Muthoot are financially efficient, but shows remarkably poor social efficiency performance, their social efficiency is low to the tune of 27.8, 28.6 and 31.8 per cent, respectively. Talking about efficiency to reach to the women and the poorest, only 8 per cent MFIs have shown efficiency above 50 per cent. This clearly shows the poor the performance of MFIs in India is, when it comes to serve women and the poorest. WSE is the only MFI, who scored perfect 1 under all six specifications and hence efficient from financial as well as from social perspective. Analysing efficiency from different specifications using DEA gives us a broader view of institutions performance and provide new insights of focus areas to improve upon their efficiency. For example, take the case of Bandhan. Looking at just the financial efficiency dimension, Bandhan is located on efficient frontier. However, looking at the social efficiency dimension, it is found to be performing dismally low. Hence, this requires a further introspection of the reasons accountable for the difference in efficiencies under different specifications. Possible reasons for Bandhan to be relatively inefficient on social efficiency front could be found in its spread of operations in east and northeast regions. Bandhan, headquartered in West Bengal, has around 400 branches in the eastern states like Bihar, Chhattisgarh, Jharkhand and Odisha. It has also spread its operations in northeast states like Assam, Tripura, Meghalaya, Sikkim, Manipur and Mizoram. Owing to the geographical constraints, for financial institutions, it is difficult to reach in east and far northeast regions. P. Vijaya Bhaskar, Executive Director of RBI, throws a light on the problem of financial exclusion in central, eastern and northeastern regions, and highlights that these three regions are home to 64 per cent of all financially excluded farmer households in the country. 2 Lack of infrastructure and connectivity to the rest of the country is a big hurdle in the

8 Kaur 61 northeast region. Thus, relative inefficient use of resources for not reaching the poorest may be attributable to the location of branches in the remote areas. A similar introspection can be made in case of Trident. Hyderabad-based is efficient under all the specifications except in the case of revenue efficiency, which is showing efficiency below 15 per cent. Possible reason could be found by analysing the financial statement of the MFI that shows as against income of ` lakh for the fiscal year 2011, the MFI had recorded a total income of ` lakh for the fiscal year This drastic fall in income is attributable to the Andhra Pradesh microfinance crisis, Thus, analysing efficiencies using DEA gives wider scope to understand the performance of MFIs from different perspectives. Further, this study provides the identification of peer MFIs that can be looked upon to improve upon the efficiency of individual studied MFIs (see Annexures 1 and 2). For example, to be financially efficient AML can follow Muthoot, SML and. Correlation between Efficiencies Examining the correlation between financial efficiency and social efficiency gives moderate but positive correlation (see Table 4). Institutionalist approach asserts that financial and social efficiency are complementary to each other. In order to serve social objectives, MFIs first have to be financially sustainable. The conclusion follows the findings of Gutiérrez-Nieto, Serrano-Cinca, and Molinero (2009) that support given the choice of financial efficiency and social efficiency, institutions would first prefer financial sustainability in order to be capable of serving the poor on continual basis. There are just two socially efficient MFIs, and they are financially efficient too. However, reverse is not true. Revenue efficiency and loan efficiency are negatively related to social efficiency and efficiency to reach the poorest, respectively, though correlation is not statistically significant. Thus, correlation matrix helps to understand the underlying mechanism by which MFIs are distributing their funds among different objectives and setting their priorities. Thus, this study could not find Figure 5. Scatter Diagram Depicting the Relationship Between Financial and Social efficiencies. Source: Author s own. the evidence of existence of trade-off between financial and social efficiency. Kumar and Sensarma (2015) confirm a trade-off while reaching the poorest of the poor for MFIs in India; however, their study negates the presence of any trade-off while reaching the women borrowers. Scatter diagram (Figure 5) further illustrates the patterns of relationship observed between social and financial efficiency. Conclusions Analysing efficiency of MFIs in India from financial and social perspective gives eye-opening results. Average social efficiency of MFIs is 32 per cent as against average financial efficiency of 84 per cent. Their efficiency to reach the poorest and women is as low as 30 and 32 per cent, respectively. The results show on social efficiency front, on an average MFIs can reduce their inputs by 68 per cent and can still reach the same number of women and poorest clients. MFIs like Bandhan have shown 100 per cent efficiency on financial front but on social efficiency front, their efficiency is even less than 30 per cent. MFIs in India are performing their financial objectives in a much better way than their social objectives. However, looking at the positive correlation between social and financial efficiency suggests social objectives of MFIs follow their financial objectives. Table 4. Pearson Correlation Between Efficiencies Financial Social Revenue Loan to Reach Women to Reach the Poorest Financial efficiency Social efficiency 0.414** Revenue efficiency 0.687** Loan efficiency 0.599** 0.811** to reach women 0.436** 0.993** ** to reach the poorest 0.356** 0.964** ** 0.952** Source: Author s calculation. Note: **Correlation is significant at the 0.01 level (2-tailed).

9 62 Vision 20(1) Therefore, study finds lack of evidence of existence of tradeoff between social and financial efficiency as suggested by welfarist approach. However, it can be said that MFIs in India give priority to financial objectives over social objectives. Thus, financially sustainable MFIs should gear up their resources on the second bottom line, that is, social objective. They should step up their efforts to benefit the poorest and women, which is one of their core objectives. The study has policy implications for donors and rating agencies as it is always in their interest to know how efficiently funds are being distributed among financial and social objectives. Further, study has relevance for the individual MFIs too, as benchmarking of MFIs has been done to identify the peer for every studied MFI to improve their financial and social efficiency. Limitation of this study is that it pertains to only financial year However, on social and financial efficiency front, availability of more data in coming years will facilitate comparison and give broader view of how shift in objectives, if any, has happened in India. Acknowledgement The author is thankful to Dr Soma Dey, Faculty of Management Studies, Delhi, for her invaluable suggestions and comments. Annexure 1. Peer Group Under Financial Specification DMU Name AML ASA India Adhikar Annapurna Mahila Credit Co-op Society Annapurna Pvt. Ltd Arohan Arth Asirvad Asomi BASIX BFL BJS BMVS BSS Bandhan Belstar CDOT CMML Cashpor MC Chaitanya DBIS Peer Group in Case of Financial Muthoot, SML, FFSL,, Trident FFSL, Muthoot, Sanghamithra, Trident SCDS, FFSL, MMFL, Muthoot, Sanghamithra. Bandhan, MMFL, Muthoot Disha Microfin, FFSL, MMFL, Muthoot Bandhan FFSL, Disha Microfin, FFSL, MMFL, Muthoot, FFSL, Muthoot, Sanghamithra, Trident DMU Name DCBS Disha Microfin ESAF Equitas FFSL Fusion GFSPL GMF GMSSS GU GUARDIAN GVMFL Grameen Sahara HiH IDF Financial Services IMPACT IRCED JFIL JFSPL Jagaran MF KBSLAB LBT M-power MFS MMFL Mahasemam Mahashakti Muthoot NEED NFPL Navachetana Nightingale PRAYAS RGVN(NE) SCDS SCNL SEWA Bank SKDRDP SKS SMCS SMGBK SMILE SML SMPL SNFL SVCL Sahara Utsarga Sahayog Saija Samasta Sanghamithra Sarala Peer Group in Case of Financial Disha Microfin Disha Microfin, Muthoot Disha Microfin, FFSL, MMFL, Muthoot FFSL Disha Microfin, FFSL, MMFL, Muthoot FFSL,, Trident Bandhan, Muthoot, SML, Trident, Trident FFSL, MMFL, Muthoot, Sanghamithra SCDS FFSL, Muthoot, Sanghamithra, Trident FFSL, Bandhan, Sanghamithra Disha Microfin, Muthoot FFSL, MMFL Disha Microfin, Muthoot FFSL, Muthoot Bandhan, Muthoot, SML Disha Microfin, Muthoot FFSL, SCDS FFSL, MMFL, Muthoot, Sanghamithra SEWA Bank FFSL, Sanghamithra, Trident, WSE FFSL, SCDS, Bandhan, MMFL, Muthoot, Sanghamithra Bandhan, SML, Bandhan, Muthoot, SML SCDS Bandhan, MMFL, Muthoot FFSL, Disha Microfin, FFSL, MMFL, Muthoot Sanghamithra (Annexure 1 continued)

10 Kaur 63 (Annexure 1 continued) DMU Name Seba Rahara Sonata Spandana Suryoday Swadhaar UFSPL Ujjivan Utkarsh VFS WSDS WSE YFS YMF Peer Group in Case of Financial Disha Microfin, FFSL, MMFL, Muthoot Bandhan, SML, Disha Microfin, MMFL. Disha Microfin, Muthoot, Trident FFSL, Bandhan, Muthoot, SML WSE FFSL, Annexure 2. Peer Group Under Social Specification DMU Name AML ASA India Adhikar Annapurna Mahila Credit Co-op Society Annapurna Pvt. Ltd Arohan Arth Asirvad Asomi BASIX BFL BJS BMVS BSS Bandhan Belstar CDOT CMML Cashpor MC Chaitanya DBIS DCBS Disha Microfin ESAF Equitas FFSL Fusion GFSPL GMF GMSSS GU GUARDIAN Peer Group in Case of Social, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE DMU Name GVMFL Grameen Sahara HiH IDF Financial Services IMPACT IRCED JFIL JFSPL Jagaran MF KBSLAB LBT M-power MFS MMFL Mahasemam Mahashakti Muthoot NEED NFPL Navachetana Nightingale PRAYAS RGVN(NE) SCDS SCNL SEWA Bank SKDRDP SKS SMCS SMGBK SMILE SML SMPL SNFL SVCL Sahara Utsarga Sahayog Saija Samasta Sanghamithra Sarala Seba Rahara Sonata Spandana Suryoday Swadhaar UFSPL Ujjivan Utkarsh VFS WSDS WSE YFS YMF Peer Group in Case of Social, WSE, WSE, WSE, WSE, WSE, WSE WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE, WSE WSE

11 64 Vision 20(1) Annexure 3. Description of Inputs and Outputs Variables Inputs Total assets ($) Operating cost ($) Loan officers Outputs Revenue ($) Gross loan portfolio ($) Number of active women borrowers Number of poorest reached Explanation Total assets, adjusted for inflation and standardized provisioning for loan impairment and write-offs Expenses related to operations, including all personnel expense, depreciation and amortization, and administrative expense Persons directly related to loan-related activities and responsible for arranging and monitoring client loans Financial revenues from the loan portfolio, income from interest, fee etc. All outstanding principals due for all outstanding client loans. This includes current, delinquent, and renegotiated loans, but not loans that have been written off. It does not include interest receivable Number of women borrowers with loans outstanding adjusted for standardized write-offs Calculated using the index for ith MFI as ALBi - Min(ALB) Pi = [1 - ]# ( Number of active borrowersi) Range(ALB) Note: Definition of inputs and outputs is adopted from MIX market database. Notes 1. Each unit under analysis under DEA is known as DMU, hence, for the purpose of efficiency analysis, each MFI will be treated as DMU. of DMUs is represented on the scale 0 to 1 or can also be in percentage terms. score 1 or (100 per cent) means DMU is on efficient frontier and score less than 1 implies DMU is relatively inefficient in comparison to other DMUs. 2. Excerpts are taken from the speech on, Financial inclusion in India An assessment, delivered by RBI Executive Director P. Vijaya Bhaskar on 10 December Copy of speech is accessed from Reserve Bank of India s official website. 3. Disclosure of financial statements of for the year is accessed from website References Annim, Samuel Kobina. (2012). efficiency: Tradeoffs and complementarities between the objectives of microfinance institutions and their performance perspectives. European Journal of Development Research, 24(5), Athanassopoulos, A. D., & J. A. (1995). Ratio and frontier analysis for assessing corporate performance: Evidence from the grocery industry in the UK. Journal of the Operational Research Society, 46(4), Avkiran, Necmi K. (1999). An application reference for data envelopment analysis in branch banking: helping the novice researcher. International Journal of Bank Marketing, 17(5), Balkenhol, B. (Ed.) (2007). and public policy: Outreach, performance and efficiency. International Labour Organization, UK: Palgrave Macmillan. Bassem, B. S. (2008). of microfinance institutions in the Mediterranean: An application of DEA. Transition Study Review, 15(2), Berger, A. N., & Humphrey, D. B. (1997). of financial institutions: International survey and directions for future research. European Journal of Operational Research, 98(2): Charnes, A., Cooper, W.W., & Rhodes, E. (1978). Measuring efficiency of decision making units. European Journal of Operational Research, 2(6), Cull, R., Demirgu c_-kunt, A., & Morduch, J. (2007). Financial performance and outreach: A global analysis of lending microbanks. Economic Journal, 117(517), F107 F133. Fajonyomi,O. S., Jegede, C. A., & Akinlabi, B. H. (2012). Relationship between outreach and sustainability of microfinance banks in southwestern Nigeria. International Journal of Sustainable Development, 5(11), Flückiger, Y., & Vassiliev, A. (2007). in microfinance institutions: An application of data envelopment analysis to MFIs in Peru. In Bernd Balkenhol (Ed.), and public policy: Outreach. Performance and (pp ). International Labour Organization, UK: Palgrave Macmillan. Gibbons, David S., & Meehan, Jennifer W. (1999). The Microcredit Summit s challenge: Working toward institutional financial self-sufficiency while maintaining a commitment to serving the poorest families. Journal of, 1(1), Gutiérrez-Nieto, B., Serrano-Cinca, Carlos, & Molinero, C. M. (2007). institutions and efficiency. International Journal of Management Science, Omega 35(2007), Gutiérrez-Nieto, B., Serrano-Cinca, Carlos., & Molinero, C. M. (2009). Social efficiency in microfinance institutions. Journal of the Operational Research Society, 60(1), Haq, Mamiza, Skully, Michael, & Pathan, Shams (2010). of microfinance institutions: A data envelopment analysis. Asia-Pacific Financial Markets, 17(1), Hassan, M. K., Sanchez, B., & Ngene, G. (2012). Scales and technical efficiencies in Middle East and North African (MENA) micro financial institutions. International Journal of Islamic and Middle Eastern Finance and Management, 5(2), Hermes, N., Lensink, R., & Meesters, A. (2011). Outreach and efficiency of microfinance institutions. World Development, 39(6),

12 Kaur 65 Kumar, N., & Sensarma, R. (2015). of micro finance institutions in India: A stochastic distance function approach. Retrieved from Publications/WorkingPapers/184fullp.pdf?t=16 Lebovics, M., Hermes, N., & Hudon, M. (2015). Are financial and social efficiency mutually Exclusive? A case study of Vietnamese microfinance institutions. Annals of Public and Cooperative Economics, DOI: /apce M-CRIL Review (2010). contributes to financial inclusion. Retrieved from Backend/ModulesFiles/NewsEvents/M-CRIL-- Review-Nov2010.pdf Nghiem, H. S, Coelli, Tim, & Rao, Prasada (2006). The efficiency of microfinance in Vietnam: Evidence from NGO schemes in the north and the central regions. Paper presented at second international conference on environmental, cultural, economic and social sustainability, Vietnam, 9 12 January. Olivares-Polanco, F. (2005). Commercializing microfinance and deepening outreach: Empirical evidence from Latin America. Journal of, 7(1), Quayes, Shakil (2012). Depth of outreach and financial sustainability of microfinance institutions. Applied Economics, 44(26), Schreiner, M. (2002). Aspect of outreach: A framework for discussion of social aspect of microfinance. Journal of International Development, 14(5), Serrano-Cinca, C., & Mar Molinero, C. (2004). Selecting DEA specifications and ranking units via PCA. Journal of the Operational Research Society, 55(5), Worthington, A. C. (1998). The application of mathematical programming techniques to financial statement Analysis: Australian Gold Production and Exploration. Australian Journal of Management, 23(1),