Consumer Protection in Indian Microfinance

Consumer Protection in Indian Microfinance

Economic and Political WeeklyMarch 31, 20071177MFIs from the purview of a proposed amendment to the statemoneylender’s act providing for an interest rate ceiling.In subsequent months there were a series of further news reportsin the local papers critical of MFIs. Finally, on March 8, inresponse to a demonstration involving stone-throwing by a groupof irate borrowers led by a local politician outside a SHAREbranch in Krishna district, demanding the return of house titledeeds which had been retained as security for housing loansSHARE was making, the collector of Krishna district moved onthe branches that night.On March 20, Sa-Dhan initiated damage control by issuingan interim code of conduct to be followed by member MFIs,which was publicised through a press-conference. The code ofconduct was finalised in January 2007 after consultations withinthe sector (see below). The RBI expressed its concern to the stategovernment that the action it had taken could have wider reper-cussions by vitiating the MFI repayment culture in other partsof the state, jeopardising a reported Rs 680 crore outstandingon loans made by the banks to MFIs in AP. It set up a co-ordinationforum to discuss issues of concern to stakeholders andresolve them as soon as possible. At a meeting of the forum onApril 20 it was claimed by government representatives that theMFIs were “eating into the SHG movement”, and that theirpractices were “barbaric” and posed a serious law and orderproblem. Spandana and SHARE announced a reduction in theirinterest rates below even those suggested in the code of conduct,including those on current loans outstanding, to 15 per cent. Itwas left to a respected MFI leader to point out that this rate wasunacceptable to other Sa-Dhan members because it was notsustainable, and that the state government had no business tostipulate rates for NBFCs which are regulated by the RBI andnot the state governments. However the state government con-tended that rates should be reduced further, especially as lowinterest rates for the poor were a major policy initiative of thestate government, as discussed below.In late September, ICICI Bank, the major lender, offered tobreak the stalemate by resuming management of its portfolio inthe 46 villages that had been identified by the district authoritiesas being particularly distressed, where it would reschedule theloans and collect them though the village organisations (VOs)– the second tier of the Velugu structure of institutions, see below– instead of through the originating MFIs, at the lower rate ofinterest already agreed to by the MFIs. The arrangement waswidely seen as a compromise, with lenders conceding to the long-standing demand of the government that they should work throughthe VOs, at a much lower interest, and the government allowingnormal operations to resume in the rest of the district. While theon-time repayment rate recovered to about 40 to 50 per cent inthe rest of the district, the initiative to collect repayments throughthe VOs has reportedly been unsuccessful. Both the MFIs andthe banks have had to write off losses. Fortunately, however,repayments in other districts were not affected, contrary to initialapprehensions.Enabling Causes of the CrisisWhile it is somewhat arbitrary to divide the causes of the APcrisis into the categories of enabling and underlying, it is a usefulanalytical device. In the absence of more definitive empiricaldata, the numbers in the following discussion are taken mostlyfrom two quick, small, sample surveys of borrower perceptions,carried out by APMAS, a respected capacity building andqualityassessment resource institution set up to support the SHGmovement in AP. The first of these surveys sampled 40 Spandanaborrowers in neighbouring Guntur district [APMAS 2005] andthe second, conducted a year later 130 borrowers in Krishnadistrict who had borrowed from either SHGs, or MFIs, or fromboth [APMAS 2006].The most important enabling (or contextual) cause was thenear-saturation of coastal Andhra with microfinance. AP has notonly covered a very large proportion of poor families under theNABARD-sponsored SBLP, but has with World Bank assis-tance federated the groups into a three-tier structure by clusteringSHGs into VOs, and federating VOs into Mandal Samakhyasat the mandal level (World Bank assistance pays for extra staffand funds including the Community Investment Fund). Ninety-two per cent of poor households in AP had already been coveredby March 2005, according to the annual report of the AP ruraldevelopment department, and the project aimed to cover the restby end 2005 [Government of AP 2005]. For the MFIs, the highproportion of poor landless or near-landless agricultural labourfamilies in the coastal districts, and the high population densitygenerally, provided strong demand conditions, and importantoperational cost-saving advantages. Spandana actually startedlife in Guntur district, next door to Krishna district. Given thehigh coverage of both Velugu and the MFIs in coastal Andhra,the Guntur survey found that dual membership was as high as67 per cent, and that multiple membership in Velugu, Spandanaand SHARE, was 32 per cent. A year later, the Krishna surveyfound that multiple membership in all three had increased to asmuch as 82 per cent. Further, despite the presence in the areaof two large MFIs (the two largest in the country) both expandingrapidly, reportedly with considerable rivalry, new local MFIswere still springing up and joining them.Despite all this competition, the Krishna survey reported thewidespread presence of informal lenders, referred to locally as‘girigiri’bankers,5 indicating that demand was still not satiated.Eighteen per cent of borrowers had taken loans at some time frommoneylenders to pay for MFI instalments. However, despite themuch publicised “comeback” of moneylenders, 80 per cent ofrespondents said dependence on moneylenders had decreased.It is interesting that while increasing loan size is the most promi-nently reported means of competition in the literature,6 and bothMFIs were increasingly making individual loans above a sizeof about Rs 10,000, Spandana reports that in view of difficultiesbeing experienced by borrowers, its average loan size had actuallycome down during the year ending with the crisis.7Anotherinteresting consequence of intense competition for clients pointedout by the Krishna study was a “softening” of practices, suchas fines for late arrival at weekly meetings.The second enabling cause was the rapid expansion of banklending to MFIs that took place after 2003, with the introductionof ICICI Bank’s “partnership” model. The innovation underlyingthe partnership model was for the bank to take the loans tofinal borrowers directly on to its own books, as against theearlier practice of the bank “bulk” lending to the MFI, withtheMFI then on-lending at its own risk. Instead, under thepartnership model, the MFI provided loan origination, dis-bursement and collection services for a fee. While in practicethis meant no change in procedures, it greatly reduced MFI equityrequirements.Thearrangement is described in some detail inNair et al (2005). It enabledgreatly stepped up lending to MFIs,especially by ICICI Bank, which accounts for over two-thirdsof total bank lending to MFIs in the country, which is estimatedto have doubled every year in the three years leading up to March2006. With the financial constraint on expansion lifted, the MFIs

Economic and Political WeeklyMarch 31, 20071178were now free to grow as rapidly as they could recruit and lendto new borrowers, who, as it turned out in coastal Andhra, verymore often than not existing borrowers of Velugu or other MFIs.The third enabling or contextual cause was the political invest-ment the current state government had made in cheap credit forthe poor as an important part of its election platform, through thePavla Vaddi or literally “quarter interest” rate scheme which derivesits name from the fact that the state government stands committedto subsidise any amount above 3 per cent that the SHGs may haveto pay the banks for linkage loans (currently about 12 per cent).The subsidy is transparent, since it is financed out of the budget(and is not a cross-subsidy as in many other previous schemes whereit was financed by larger bank borrowers) and is “collective”, inthe sense that it is deposited into the group’s bank account, andthat too only after completion of timely repayment, which is madea condition.8Politically it must have been jarring to see MFImicrofinance not only flourishing, but growing rapidly in the stateat much higher interest rates.Fourth, there was, and still is, widespread lack of publicawareness of features of MFI microfinance such as doorstepdisbursement and collection in weekly instalments, which leadto the seemingly “usurious” interest rates they have to charge.This lack of understanding is shared equally by the bureaucracy,politicians and the media and was clearly on display in theexaggerations, misinformation, and screaming headlines9 thatfollowed the crisis. Thus there was a massive external or publicrelations failure on the part of the sector, and one of the majorchallenges before it is a public education programme.Underlying CausesRush to Grow at All CostOne of the longer term causes was clearly the “quest fornumbers” relating to outreach and profitability that is the mainmotivation of many MFIs. While extending the depth and breadthof outreach is clearly central to microfinance’s mission of makingan impact on poverty through financial inclusion, and whilesustainability is essential if MFIs are to attract lenders andinvestors in order to grow, the crisis serves as a useful reminderthat there are other just as important client-centred consumerprotection objectives such as transparency in dealings withborrowers, and being careful not to saddle them with more debtthan they can handle. These are goals that apply equally tominimalist (define) as well as more holistic microfinance.10Ina sense it is the industry that shares the responsibility for buildingup a climate of expectations that celebrates the inter-relatedachievements of rate of growth of outreach, efficiency, and fieldworker productivity without always remembering that these can:first, lead to short-cuts in client selection and training, fieldworker training and sensitisation, and loan size determination;second, be used as the only criteria for incentive payments tofield workers: and third, put a degree of pressure on them thatleaves no time for issues affecting client satisfaction, other thanloan turn-around time, and progression in loan size, etc.11MFIs have a social as well as a financial mission. Unfortunatelysocial performance is not as easy to measure as financialperformance, although a welcome development in India is thenew social rating tools being developed by the main MFIrating agency, Microfinance Credit and Ratings International(M-CRIL), to measure social performance defined as the “ef-fective translation of mission into practice, in line with acceptedsocial values”. These tools include surveys to obtain feedbackon client protection issues, client satisfaction, reasons for drop-ping out, and similar variables. Interestingly, mission drift in theliterature usually refers to moving up-market to increase breadthof outreach at the cost of depth of outreach. This has not beenan issue at all in the present crisis, and it is generally acceptedthat MFIs following the Grameen Bank methodology in particularhave an extremely poor clientele.12Rather, the mission drift herewas to ignore the rest of the social mission at the expense ofthe financial mission.MFI PracticesInterest RatesA second underlying cause relates to the operational practicesof the MFIs concerned, which were partly the outcome of therush to grow and become profitable as soon as possible. Thesepractices were probably no different from those of many otherMFIs growing rapidly in other areas, but since they have receivedextensive publicity there is more information on them, allowingusto use them to discuss issues that apply to the sector as a whole.While they were not the real reason for the state government’saction, “usurious interest rates and hidden costs”, or what arereferred to as “deceptive” interest rates in the literature [CGAP2004] were given prominence in the early days after the closure,before many state government officials realised that NBFCs arefree to set their own rates. Indeed they do not seem to have beenaware that the state moneylenders act was no longer valid,13 sothat non-NBFCs were also legally free to do so.Spandana’s and SHARE’s effective interest rates before thecrisis are reported to have been 31 and 28 per cent respectivelyfrom 2002-03 to 2005-06, when they were reduced to 28 and24 per cent. These rates are not very different from the modal rateof 2 per cent a month rate charged by SHGs, and below half thatof the informal sector (for ‘girigiri’bankers they are 6.7 per centa month). They include the loan processing fee, but not a one-timemembership fee. The MFIs also required borrowers to make asecurity deposit as cash collateral.14Third, Spandana deducted1 per cent of the loan amount or Rs 50 towards a life insurancepremium in which the borrower’s spouse received the loan amount,less the outstanding balance if she died, on which Spandana madea surplus [Roth et al 2005]. However this too does not increasethe effective interest rate, whether or not one agrees with the viewexpressed by the press and government that the surplus from thescheme should have been distributed to borrowers.15It is truethough that borrowers lacked a clear understanding of their entitle-mentsunder the scheme, which some studies report is generallythe case with insurance products distributed by MFIs in India.A practice that has raised some eyebrows was charging bor-rowers who wanted to prepay their current loan in order to availof a larger one, interest on the entire period remaining on thecurrent loan. This can be a rather high penalty, depending on theperiod in case. As with many other questionable practices theMFIs were accused of, there is no data on how frequently it wasresorted to. The MFIs, like most others in India and Bangladesh,were also using flat rates. Transparency demands MFIs shouldprominently disclose the effective interest rate as an APR.16Coercive Collection PracticesThe set of accusations that received the most prominencehowever related to coercive collection practices, leading toborrowers having to “abscond”, or migrate out of the village,

Economic and Political WeeklyMarch 31, 20071179and even in some cases, allegedly, commit suicide.17The respon-dents in the Krishna survey felt that (i) joint liability (the grouppaying on behalf of the defaulter), (ii) compulsory attendance,(iii) fines and (iv) keeping all members waiting until repaymentsare made, are the chief means (in that order of importance) ofensuring a “cent per cent recovery” rate. Means that wouldgenerally be regarded as “abusive”18or at least questionable werementioned by respondents in the followingorder of frequency19(i) adjusting overdues against the security deposit, (ii) holdingthe weekly meeting in front of the defaulter’s house, (iii) MFIstaff sitting in front of a defaulters door, (iv)offensive languageused by group leaders or staff, and (v) putting up a loan overduenotice in front of a defaulters house.20The point at which peer group pressure becomes coercive isan extremely difficult one and is discussed further below. Howeverone clear lesson of the crisis is that the policy of 100 per centrepayment and “zero tolerance” for default carried a very highcost in terms of client dissatisfaction, and provided amplematerial to be exploited by interested parties. Clearly there isa need for flexibility to accommodate cases of extreme distressin which a borrower is unable to pay because of critical illness,hospitalisation etc. A second lesson is that there is a great needfor action research to provide answers to the question of howflexible MFIs can afford to be, even in cases of lesser distress(such as failure of a business) in rescheduling loans, withoutaffecting repayment discipline generally, and how much opera-tional costs would go up to introduce such flexibility.21 Third,an additional response should clearly be much wider use ofemergency loans and risk funds.OverlendingThe third set of accusations was that MFIs were “dumpingmoney on borrowers” who were finding it difficult to repay, andhaving to borrow from moneylenders at a higher cost in orderto stay in good standing with the MFI. This is an extremelycomplicated issue calling for much further field research and isdiscussed further below. While the banks are in a position to lendto salaried borrowers whose total income is relatively easy toassess, MFIs lend almost entirely to the self-employed whoserelevant income is that of the household as a whole. Such incomesare very variable, and very hard to assess. There was no doubtsome overlending but we need detailed case studies of howsomeborrowers got into repayment difficulties. This is the kindof research still neglected in India as elsewhere, and needseconomic anthropologists willing to live in villages for prolongedperiods and to use participant observation and other methods toget over the limitations of survey-based research and “quick anddirty” studies.Unattractive Features of the SHG Model for BorrowersAthird underlying cause of the crisis lay in certain featuresof the main alternative model of microfinance available toborrowers, the SHG programme. These features made Veluguless attractive to many borrowers as a source of micro-credit(although they may have continued to value many of the otherservices and benefits accruing from Velugu), and led them toborrow also from the MFIs. Most of these features are inherentdisadvantages of the SHG model itself, although on balance thedisadvantages may be dominated by the advantages,22the mostimportant of which has to do with empowerment.23In the Guntursurvey of Spandana borrowers, two-thirds of whom were alsoSHG members, inadequate loan size was the most frequently citedproblem in borrowing from SHGs, followed by the long waitingperiod for loans. In contrast, timeliness of loans was cited mostfrequently as an advantage of borrowing from MFIs. In thesecond, Krishna district, survey, cumulative borrowing fromSHGs was found to be about 40 per cent of that from the twoMFIs (standardised for number of loans), although there was lessdifference in loans outstanding given the longer tenor of SHGloans. MFIs deliver loans of larger average size in a more timelyfashion because (i) they borrow in bulk from the banks forrelending, whereas SHGs have to wait for the last member inthe group to repay her loan to the group before the group canrepay its loan to the bank, and (ii) when they do get a loan, itssize is tied to group savings, and depends on the assessment ofthe local bank manager on the absorption capacity of the groupfor a subsequent loan.Interestingly, the Krishna district survey ranked the availabilityof individual loans as the most frequent response among reasonsfor enrolling in MFIs, ahead of timeliness and large loans.24Interestingly, again, the most frequent response (90 per cent) forproblems with MFI procedures was weekly instalments, justahead of the high rate of interest, and well ahead of “rigidityeven in genuine cases” (19 per cent). “More pressure and mentaltension” on account of weekly repayment was an importantdrawback also reported in the first survey.25On balance, given these advantages and disadvantages, only10 per cent of Guntur district borrowers said they would dis-continue MFI membership after repaying their current loan, aproportion that increased to 21 per cent in the Krishna survey.However as the survey report explains, many members whointend to leave, end up “prolonging” their membership becausea member who leaves and then changes her mind and want torejoin again, either has to wait for another drop-out in her originalgroup who she can replace, or organise a new group. In additionshe loses her seniority and is on a par with new members in respectof loan size. So the incentives are structured so as to keepmembership alive.Likewise, despite dissatisfaction with SHG membership asa credit institution, very few members actually quit and“defect”.This is becauseSHG membership too confers manyadvantages,with access to a number of developmentprogrammes, services and benefits being contingent on mem-bership.26 Thus there was no “poaching” as such, because bothmodels were restricting competition by making it costly forborrowers to switch. The impact of competition is more onregularity of loan repayments and monthly savings in the SHGs.The Krishna survey found overdues amounting to almost halfof total loan amount outstanding in the SHGs that membersbelonged to. The proximate cause of the state government actionagainst the MFIs was the increasing frustration being experiencedby Velugu managers over the effect MFI lending to SHG bor-rowers was having on Velugu performance as shown in indicatorssuch as the recovery rate within groups, and of on-timerepayments by the groups to the banks, and to the CommunityInvestment Fund (CIF).27Lack of Contact and CommunicationA final underlying cause was that the MFIs, in their self-absorption with growth at the cost of other objectives, under-estimated the degree to which they were antagonising severalpowerful players in the local political economy. Among thesewere informal lenders such as, reportedly, chit fund organisers,28

Economic and Political WeeklyMarch 31, 20071180and local girigiri bankers, politicians who were embarrassedby the fact that the MFIs were disbursing much more moneyin their constituencies than they could lay their hands onthemselves through their constituency funds, some bureau-cratswhofelt uncomfortable about anything major happeningin their district, even if it was good, without their “blessings”,or at least their knowledge. Even retail traders of consumerdurables are said to have felt threatened by the fact that Spandanahad earlier set up a consumer store so as to be able to purchasegoods in bulk and enable borrowers to benefit from lower pricesusing vouchers they had the option to accept instead of cashloans.29And finally of course there were a number of highlydiscontented borrowers, as noted above, some of who werechampioned by local politicians. This powerful combinationof interests may explain the lack of public support when thecrisishit.Towards a Modus Vivendi30Thus the AP crisis was caused essentially by a combinationof four factors, (i) the fact that the state itself was an activepromoter of microfinance through a politically-attractive andhigh visibility programme, Velegu, (ii) competition between thisprogramme and MFIs, (iii) the fact that this competition wasdamaging Velugu, and (iv) that the state government intervenedto protect the model it was promoting, or in other words let itspromotional role dominate the role of neutral umpire betweenmodels. With the growing state government involvement in theSBLP in other states too, the potential for conflict will increase,and the need to evolve a harmonious modus vivendi betweenthe two models becomes all the more urgent.MFIs will have to become much more adept at buildinglong-term healthy relationships with government and otherstakeholders at local and higher levels. As Sa-Dhan’s code ofconduct states, MFIs should aim to complement the activitiesof the SBLP. They will be adding the most value by expandingto areas underserved by both models, as indeed Andhra MFIshave been doing increasingly since the AP crisis. In those areaswhere they operate side-by-side with the SHG model however,they should show greater sensitivity to the need not to affectcredit discipline in SHGs. It would not seem unreasonableto expect an MFI to satisfy itself before lending to an SHGmember, that she will be in a position to repay not only the MFIloan, but also any dues she might have to the SHG. However,at the same time, in such areas, government agencies such asDRDAs also need to respect the right of borrowers to exercisechoice. There are many borrowers who genuinely need thekindof larger and more timely loan MFIs have a comparativeadvantage in offering.Greater information sharing and much closer contact willhave to be part of the solution. It is essential though, thatno matter how closely a state government is associated withthe SBLP, it does not let its role as a player dominate thatofanimpartial umpire between the two models in the event ofconflict. Moreover, it should respect the fact that as financialinstitutions MFIs are crucially dependent on public confidence,which is very hard to restore once damaged, and any actionagainst erring MFIs should be taken by the proper authorityaccording to law. The sector should in the future be preparedto challenge the legality of actions such as that taken by theAP government in Krishna district in March 2006 in thecourtsif necessary. It needs of course to also put its own housein order.Code of Conduct and Some Remaining QuestionsAmong the positive effects of the March 2006 episode wasthe fact that many of its lessons for MFI practices were recognisedand codified in Sa-Dhan’s interim code of conduct which waslater finalised after widespread consultations within the sector.The interim code emphasised among other things the need (i)toavoid over-financing of the same household by different MFIs,(ii) make interest rates more transparent, (iii) ensure that staffdo not use abusive language or intimidation tactics while col-lecting repayments, (iv) ensure high standards of corporategovernance by including on MFI boards eminent independentboard members, and (v) stay in touch with government authori-ties, banks and the media on a regular basis.The interim code was finalised and supplemented with a state-ment of core values released on January 2007, reiterating interalia that services will be delivered in an “ethical, dignified,transparent, equitable and cost effective” manner. Clients shouldbe educated about the terms of loans offered to them, and a balancestruck between respect for a client’s dignity, an understandingof her vulnerability and a “reasonable pursuit” of loan recovery.Sa-Dhan would set up an independent Ethics and GrievanceRedressal Committee to look into cases of non-compliance bymember MFIs, leading to expulsion if warranted.While this is an important step forward, many difficult ques-tions remain, on which much soul-searching and hard thinkingstill needs to be done. For instance, what constitute abusivecollection methods? Group lending is built on the foundation ofjoint and several liability (or social collateral, also referred toas a collateral substitute). This applies equally to SHGs, wherethe lender is not concerned with the identity of the individualsto whom the loan is distributed by the group, and to smallerGrameen-type groups where loans are recorded in the names ofindividual borrowers. Joint liability, although theoreticallyenforceable in the latter case through the courts (because othergroup members sign the necessary documents), is in practiceenforceable only morally through peer group pressure.The strictly moral component of peer group is the sense ofobligation that group members feel to repay their loans. In India,as in Bangladesh and perhaps elsewhere, poor village womentake the oath they are administered when becoming membersextremely seriously. The other more opportunistic componentof peer group pressure derives from the incentive to repay asthe condition for receiving larger loans or continuing to receivecredit at all.The not-easy-to-answer question is, can MFIs rely on thesetwo components alone, and if not what other methods are ethical?There was much talk during the AP crisis of MFIs keepingborrowers standing in the sun waiting for the last member of thecentre to turn up with her loan payment. A field worker froma rival MFI felt this was unacceptable (as most people would).But when asked “how about keeping them waiting sitting down,in the shade?” she replied: “Well perhaps, for a while…” In thememorable words of a professor at the Indian Institute ofManagement, Bangalore, “you don’t recover loans by smilingat people”. However, at what point does embarrassing a borrowerfor keeping everyone waiting by coming late to a meeting withher repayment, become unacceptable humiliation?Second, what constitutes over-lending? It is not easy to definethe tipping point beyond which a borrower over-borrows. Clearlywhen borrowers default involuntarily they have over-borrowed,but not all over-borrowing leads to default, as when a borrowersellsa valuable asset, or cuts back on educational expenses, or is forced

Economic and Political WeeklyMarch 31, 20071181to work when ill in order to meet the next weekly instalment.Group lending leaves it to the good sense of group leaders totake a call on whether to chip in for a “deserving” defaulter. Butwhat are the ethical considerations when they are not preparedto put up for a wilful defaulter who then migrates out of the villageor worse. Instances of excessive peer group are not the monopolyof MFIs. There are documented cases of SHG members contem-plating suicide, and it is standard practice in the bishisor informalcredit unions of Maharashtra for the entire membership of perhaps50 to a 100 people to visit and embarrass a defaulter in her house,perhaps even restraining her inside temporarily.Third, when does healthy competition become unhealthy?What constitutes poaching – merely lending to a borrower whenshe has a loan outstanding to an existing lender? A libertarianin these matters would advocate respecting her freedom of choiceto avail of a larger or more timely loan. Is there an obligationon the new lender not to lend to her until she has fully repaidher existing lender? Or is the lender only obliged to ensure thatshe will be able to pay back both loans? Is it necessarily irrationalto borrow from a higher cost moneylender for a few days to repayan MFI? “Double dipping” is said to be widely practised inBangladesh. What have been the effects?Hopefully clearer answers will emerge to these questions asthe work of ethics committee, and that of the ombudsman ap-pointed under the proposed microfinance bill, gets going, andexperience accumulates.Proposed Microfinance BillConsumer ProtectionA draft of the Micro Financial Sector (Development andRegulation) Bill, 2007 had at the time of writing been presentedto the cabinet and should be presented to Parliament in the budgetsession. While one of its main objectives is to allow MFIs toaccept savings, the bill has two important components whichshould greatly strengthen the consumer protection regime. First,it provides for the redressal of grievances through an ombudsman,and through officers to be appointed “at places easily accessible”to clients, with powers to issue directions to MFIs. Second, itseeks to promote the orderly growth of the sector, by settingperformance and accounting standards, creating a reporting systemand data base, supporting research, promoting consumer educa-tion, disseminating information relating to fair practices, andlaying down a code of conduct.Taken together these activities should lead to the creation of amoreaccountable and transparent microfinance sector generally,with the last three contributing to the consumer protection regimeinparticular. Enforcement will be the responsibility of the ombuds-man and his organisation, complementing the self-regulatoryefforts of the sector itself. By making the enforcement of consumerprotection clearly the responsibility of the new regulator, NABARD,the bill will also provide comfort to MFIs by removing theoccasion for episodes similar to those that took place in Krishnadistrict in future. In that sense, some good has come out ofit.We will discuss next one of the main objectives of the bill, whichis to enable MFIs to offer a savings service to their members.SavingsIndia’s attitude to savings mobilisation by non-banks has beenmore restrictive than elsewhere, an attitude strengthened byperiodic scams which affect the savings of the urban middleclasses and which therefore receive widespread publicity in thepress. There is also a widespread misconception that the poorare too poor to save, and that they need credit, not savings facilitiesand services. On the contrary, savings is probably a more widelyfelt need than credit, and takes place through a variety of savingsmechanisms and institutions in the informal sector, such as smallcommunity chit funds, or informal credit unions (such as thebishis of Maharashtra), or by investing in livestock or ornaments(which can later be liquefied through the pawnbroker andmoneylender), or by lending to a friend in need so that the lendercan borrow reciprocally when required. The phenomenal growthof the SHG movement in which rural women meet with unfailingregularity once a month to save small sums ranging from Rs 10to 50 attests to the importance of the almost universal need tosave. Like the rest of us, the poor are looking for savings serviceswhich are convenient, safe, liquid, and can preferably be usedto leverage loans.In restricting MFIs (although not SHGs) to the delivery ofcredit the poor have until now been deprived of an essentialfinancialservice that MFIs couldoffer, and have been forcedto rely instead on less convenient, riskier, lower yielding, andoften socially less productive savings instruments (such asornaments). Except for SHGs, and MFIs registered as coop-eratives, whichaccount for a very small albeit growing pro-portion of total microfinance lending, neither (i) unincorporatedMFIs registered as societies or trusts (the majority of MFIs innumber, although not as a share of total borrowers and loans)or (ii) those incorporated as NBFCs or non-profit S25 companies,are allowed to mobilise savings, even from their borrowermembers.There isunderstandable reluctance to allow MFIs tomobilise public deposits, for sound prudential reasons. But thevast majority of MFI members are net borrowers of the MFI atany one time. They borrow to finance their larger investmentrequirements, but simultaneously save small amounts regularlyto finance their liquidity requirements, provide for emergencies,build up a cushion to tide over the lean seasonwhenagriculturalwage employment is scarce, and aggregate savings intoamountslarge enough to make useful investments, repair thehut, send a daughter to high school, or a son to the big city tolook for work.Recognising this, the draft microfinance bill allows what it calls“Microfinance Organisations” or MFOs to collect “thrift” fromits members through the group mechanism after obtaining acertificate of registration from NABARD, the proposed regula-tory authority. In order to be eligible, the MFO must have beenin existence for at least three years and have net owned fundsof at least Rs 5 lakh.31Quite apart from the benefit to member-savers, there will be benefits to the MFI – their cost of fundswill come down, and members’ sense of ownership in their MFIwill increase, since a small part of the loans they will be gettingwill be their own money coming back to them, giving them afurther incentive to maintain high repayment rates. BecauseIndian MFIs have had to rely until now exclusively on fundsborrowed from the banks their financial expense ratios (cost offunds as a proportion of total costs) are the highest in the world[Ghate 2006]. In Bangladesh on the other hand interest rates areslightly lower than in India, because as much as a third of thefunds base comes from member savings. Many Bangladeshi MFIsalso pay a slightly higher rate to savers than the banks, becauseof the cost efficiencies that ensue from economies of scope inbeing able to use existing field staff, who have to meet borrowersonce a week anyway to collect loan repayment instalments anddisburse fresh loans. Grameen Bank mobilises in fact more

Economic and Political WeeklyMarch 31, 20071182savings than it disburses as credit (as does Bank RakyatIndonesia,further examples of how savings are as valued by poorclients as credit).Two DeficienciesWhile the proposed legislation is therefore a great step forward,for savings related as well as other reasons, there are two de-ficiencies in the draft bill. First, it excludes MFIs registered asNBFCs and S25 companies, which account for nearly all the largeMFIs and a larger part of total microcredit in the country. Theirnumber is steadily increasing as they are joined by more and moreNGO-MFIs transforming themselves into companies in order toescape the vicissitudes of state government policy towards MFIs(as financial companies they will be regulated by the RBI). Yet,NBFC-MFIs are much stronger than NGO-MFIs in terms ofcapitalisation, liquidity, audit and disclosure (because of regu-latory requirements laid down for all NBFCs) and are betterequipped to handle the extra accounting requirements entailedin collecting and servicing savings. The irony is that NBFCs theyare in principle even allowed to accept even public deposits onmeetings certain requirements. However, these requirements areso stringent that none of the MFI-NBFCs have succeeded inqualifying yet. By being placed outside the purview of theproposed bill, they will not be able to mobilise thrift from theirmembers either.The argument adduced for keeping them outside the purviewof the bill is that NBFCs are already governed by comprehensiveregulations framed by the RBI. However the same argument(the need to avoid duality of regulation) is applicable todistrict, state andurban cooperative banks which are governedby the Banking Regulation Act in respect of banking activities,while conforming tothecooperative law in other respects. Likethem,NBFCswouldbe governed by the microfinance bill inrespect of thrift activities without any dilution of their capital,reserve, or liquidity requirements as NBFCs, until they qualifyunder NBFC regulations to mobilise not just thrift from membersbut public savings.A second precedent (of dual regulation) is contained withinthe bill itself, which includes cooperatives in the definition ofMFOs, many of which are savings and credit cooperatives,whether primarily agricultural such as the PACS, or non-farm.Cooperatives as mutual organisations are already allowed toaccept savings from their members. Indeed the term thrift in thesense of small, periodic, usually compulsory payments, waspopularised in India by the Cooperative Development Founda-tion, a well known NGO in Andhra Pradesh which spearheadedthe movement for cooperative reform, and lobbied with the APlegislature to enact the Mutually Aided Cooperative SocietiesAct or APMACS Act in 1995. This was followed by the enactmentof similar acts in 8 other states and a central multi-state act toprovide for a new type of “mutually aided” or “mutual benefit”cooperative which would be much truer to cooperative principlesthan traditional cooperatives under existing state acts, andenjoymuch more autonomy and freedom from governmentinterference of the kind that has led to the deterioration oftraditional cooperatives.CDF has set up 450 “mutually aided”, women and men’s thriftcooperatives under the new act in Warangal district of AP, whichexercise a powerful demonstration effect on how savings andcredit cooperatives should function. With meticulous accounts,good governance, compulsory monthly thrift ranging fromRs20to50, and membership of about 550 each, they havemobilised own funds or member savings in various forms to thetune of Rs 37 crore (by March 2006), of which Rs 32 crore hasbeen lent back to members. CDF is strongly opposed to MACSbeing brought under the bill since they are already empoweredto mobilise thrift under a carefully circumscribed and adequatelyfunctioning regulatory regime.32It fought a long battle to rescuecooperatives from the ill-advised attempt over the years to assistcooperatives with subsidies so as to convert them into vehiclesfor political patronage, and is understandably worried aboutcoming under the oversight of an organisation it fears may notunderstand the MACS ethos. There is a strong case therefore forexcluding cooperatives from the purview of the bill insofar asit relates to thrift.The bill does indeed empower the central government toexempt a particular class of microfinance organisation fromany or all theprovisions of the bill, and it will have to do sofor SHGs of which there were 2.23 million by March 2006, andwhich cannot possibly be expected to apply for registration tomobilise savings individually. However CDF and the MACScommunity are understandably reluctant to depend on the goodsense of government to exempt them as a class after the billbecomes law. As a vocal civil society organisation CDF hasexperienced considerable opposition from government on a hostof issues in the past, the latest instance of which is the APgovernment’s desire to foreclose the MACS option to milkcooperatives, on which CDF has obtained a stay from the APHigh Court. CDF is on strong legal grounds for insisting thatas mutual organisations already allowed to mobilise thrift by stategovernments under whose jurisdiction cooperation belongs as astate subject, cooperatives should be kept out of the bill fromthe outset.The difficulty with this is that one of the purposes of the billas mentioned above is to create a data base in the public domaincontaining essential information such as the the number and typeof MFOs of different kinds, the number of borrowers (and nowsavers), the quantum of lending and savings (or in other wordssomething as basic as the size of the sector),and a host ofother variables, the lack of which is an important gap in theknowledgebase essential for sound policymaking for the sector.33MACSarelikely to become increasingly important players inthe sector since AP and Orissa are already using the MACS formof incorporation to register federations of SHGs in their respectivestates, and are likely to be followed by other states. However,despite the reporting requirements in the MACS Acts, data onMACS is as deficient as it is on societies and trusts. It wouldseem advisable therefore to find a way of including cooperativesin respect of thereporting requirements of the bill, and perhapsalso those relating to its consumer protection provisions,since not all cooperatives, whether traditional or MACS, haveas good andresponsive a governance as those organised by CDF.Also, it is not clear whether the bill has been brought intoconformance with the ongoing effort to reform the cooperativemovement generally, to be underpinned by huge loans from theWorld Bank and ADB. At the very least, this and other issuesneed to be discussed much more widely and openly than theyhave been so far.To conclude, it would seem a great opportunity lost nottobring MFIs registered as NBFCs and S25 companies into thebill in respect of their microfinance activities, although not inrespect of reserve, accounts, audits and other prudentialrequirements which would continue to be governed by NBFCregulations. Second, it would seem desirable to officiallyannounce the intention, priortoenactment of the bill, of

EPW