Financing microfinance the ICICI Bank partnership model

Bindu Ananth

March 2005

Published in Small Enterprise Development Vol 16 No.1, March 2005

Centre for Micro Finance, 8 Floor, West Wing, Fountain Plaza, Khaleel Shirazi Estate, 31/2 A, Pantheon Road, Egmore, Chennai

th

Capital constraint is an issue impeding scaling up of microfinance in India. Based on an analysis of traditional financing models and ICICI Banks experience in India, it analyses the partnership model of financing microfinance institutions (MFIs). This model is unique in that it combines both debt as well as mezzanine finance to the MFI in a manner that lets it increase outreach rapidly, while unlocking large amounts of wholesale funds available in the commercial banking sector in India. The paper also discusses building links to capital markets for financing microfinance through securitization. It concludes by highlighting certain key enablers for an environment of rapid microfinance growth including regulator support for hybrid models of outreach and 1 investments in training and funding of initial expenses for new/emerging MFIs.

Bindu Ananth is a Manager at ICICI Bank. She has worked in ICICI Banks microfinance practice for over three years and was the coordinator of the Centre for Micro Finance at the time this paper was written. The views presented in this paper are solely those of the author and do not in any way reflect the views of ICICI Bank.

Centre for Micro Finance -MFI Strategy Unit-June 2006

Financing microfinance -- the ICICI Bank partnership model

BINDU ANANTH
Capital constraint is an issue impeding scaling up of microfinance in India. Based on an analysis of traditional financing models and ICICI Bank's experience in India, it analyses the `partnership model' of financing microfinance institutions (MFIs). This model is unique in that it combines both debt as well as mezzanine finance to the MFI in a manner that lets it increase outreach rapidly, while unlocking large amounts of wholesale funds available in the commercial banking sector in India. The paper also discusses building links to capital markets for financing microfinance through securitization. It concludes by highlighting certain key enablers for an environment of rapid microfinance growth including regulator support for hybrid models of outreach and investments in training and funding of initial expenses for new/emerging MFIs. MICROFINANCE IN INDIA has received a recent impetus with a growing number of commercially oriented microfinance institutions (MFIs) emerging and banks increasing their exposure to this sector considerably. Despite this, there is still a large gap between estimated demand and total supply. Countries encounter varying challenges in the pursuit of universalization of access to basic financial services. In India, several enabling factors exist: a vibrant commercial banking sector, the presence of operationally sound MFIs and a regulatory environment that permits multiple models of building access to credit. This article seeks to focus on one particular aspect that is impeding scale in India: that of access to the capital required for building a scaled microfinance industry. Some of the insights may be useful for other countries as well. One of the key assumptions of this article is that the route to universalization of microfinance, certainly in the case of India, is to work with high-quality, autonomous (both for-profit and not-for-profit) local MFIs. Commercial banks directly building outreach to microfinance clients is not a model that is discussed here. In India, high-performing MFIs grow at sub-optimal rates simply because of a capital constraint. Often, wholesale funds for on-lending and equity are not distinguished when debating availability. Owing to priority sector targets for banks, which require banks to lend at least 40 per cent of their net bank credit in any given year to agriculture and weaker sections, including lending to historically disadvantaged communities and micro-credit, there has been a logic for the flow of wholesale funds to this sector, however limited it has actually been. However, there are no `natural providers' of risk capital for microfinance and this has constrained the growth of several MFIs. (In fact, risk capital or equity is a constraint, not just for microfinance in India, but also for the small-scale enterprise and infrastructure sectors).

Banks in India usually work with high-quality MFIs to reach poorer clients, rather than serving them directly themselves

Centre for Micro Finance -MFI Strategy Unit-June 2006

Under the first model, the bank lends directly to the self-help groups

The costs of promoting the selfhelp groups is borne by the NGO, and usually covered by donor grants

A discussion on risk capital should distinguish between explicit and implicit capital (Mor and Sehrawat, 2003). Explicit capital is the capital committed by the promoter of a project and implicit capital is that committed by the lender to a project usually in the form of economic capital or Tier 1 or 2 capital adequacy. In the Indian microfinance context, it must be noted that the scarcity has been of explicit capital that the promoter (in this case the MFI) would bring. Banks, on the other hand, have an ability to contribute the implicit capital, at least in the volumes required by microfinance. The article discusses a model that overcomes this challenge. Recommendations by the government to set up `equity funds' for this sector do not adequately address key issues such as rules governing allocation of these funds among competing MFIs, incentives for MFIs to use these funds carefully and a business model for the fund itself. Models of microfinance that have prevailed tend to use risk capital inefficiently. The few MFIs in India which have raised commercial equity have delivered return on equity figures of less than 5 per cent, primarily because of low levels of leverage. The fact that several MFIs have chosen to establish themselves legally as non-banking finance companies (NBFC) also imposes a demand for regulatory capital. (In order to register as an NBFC, an MFI needs entry-level capital of Rs20 million, approximately US$500 000 at an exchange rate of Rs40 = US$1.00.) The preference for an NBFC format may be largely explained by the desire of MFIs to provide savings facilities for clients. In a later section, the article briefly discusses how savings may be facilitated without the MFI adopting an NBFC format. A comparison of three financing models for microfinance In order to examine the merits of the partnership model, it is important to understand the traditional financing approaches for microfinance in India and to compare them according to the two key dimensions: capital and incentive alignment. Model 1. The self-help group (SHG)--bank linkage model The predominant model in the Indian microfinance context continues to be the SHG--bank linkage model that accounts for nearly 20 million clients. Under this model, the self-help promoting institution (SHPI), usually a non-governmental organization (NGO), helps groups of 15-20 individuals through an incubation period after which time they are linked to banks. The bank lends to the groups after the incubation period, and this linkage may be single- or multi-period. The SHPI typically receives no, or below cost, consideration either from the bank or the clients for the function of group promotion. They meet this expense out of external grant sources. The cost estimates for this vary from $35--250 per group. Once the groups have been linked to the bank, the SHPI often supervises the loan portfolio. However, there is no specific incentive for the SHPI to play this ongoing role, which has associated expenses in terms of employee time. Pricing to the clients under this model is not based on full costs, as the costs of promotion and those of transactions handled by the SHPI are not included. Capital allocation. Since the SHPI in this case does not play the role of a financial intermediary, it does not have to allocate capital against the lending under this programme. In most cases, the SHPIs in this programme are trusts or societies that have no desire to take credit risk. Capital needs to be allocated only by the bank, which bears the entire credit risk.

Centre for Micro Finance -MFI Strategy Unit-June 2006

The incentive of the NGO to originate high-quality groups and supervise portfolio is weak

There is a double counting of the capital requirement to cover the loan portfolio

Incentive alignment. In this model, the bank bears 100 per cent of the credit risk. It has no recourse to the SHPI, which has originated the groups, in the event of default on loans. Therefore, at least in theory, the incentive of the SHPI to consistently originate high-quality groups and supervise existing portfolio is weak. One of the reasons why this might not be reflected in repayment rates under the SHG--bank linkage programme (which are reported to be high) could be that portfolio sizes per SHPI are small and not quite comparable to that of MFIs. Also, for several of the larger NGOs, repayment is seen to reflect on reputation and hence they ensure that supervision levels remain high even if it means raising resources from donors and other agencies. Repayment rates might alter with a significant scaling up of portfolio per SHPI and bank branch. Model 2. Financial intermediation by the microfinance institution In this model, the MFI borrows from commercial sources and on-lends to clients (groups/individuals). This is a recent shift that has been facilitated in part by the participation of commercial banks in the microfinance sector and in part by the lack of resource options for growing MFIs, given that they cannot take deposits and face limited availability of grant funds. Most MFIs in India started operations with grants and concessional loans and gradually made the transition to commercial funding. For instance, Bharatiya Samruddhi Finance Ltd (BSFL), one of India's leading MFIs, financed much of its growth in the initial years with concessional loans from funding agencies. This was followed, from 2001 onwards, by BSFL raising equity from various domestic as well as international sources. Capital allocation. Here, capital allocation happens at two stages for the same portfolio that is financed. An illustration will make this clearer. If an MFI estimates a loan requirement of, say, $250 000 for its clients, it approaches a bank for that amount. The bank views it as a lending `to the MFI' (organization-based lending), not finance for the underlying pool of borrowers (asset-based lending). Accordingly, pricing is a function of the rating of the MFI. In most cases this is likely to be poor, owing to low levels of equity capital, in contrast to rating on the loans pool, which is derived from historical loss rates. The bank allocates capital as relevant to the rating obtained by the MFI. Now, when the MFI on-lends the $250 000 million to its clients, it further allocates capital against this portfolio to take care of unexpected losses on this portfolio and to satisfy capital adequacy norms that may govern it. This, in effect, results in a double counting of capital requirement for that particular portfolio of microloans -- once by the bank and again by the MFI. The final pricing will therefore include capital charges at both these levels. Incentive alignment. The incentives of the MFI to maintain supervision levels are high in this case, as it bears 100 per cent credit risk on the portfolio. The banks' incentive is directed to ensure MFI solvency. Model 3. The partnership model -- MFI as the servicer In 2002, an internal analysis by ICICI Bank revealed that despite consistent evidence of viable demand from clients, access to MFIs was constrained due to the organization-based financing model adopted until then. Owing to the limited number of rural branches, the SHG--Bank Linkage model was not considered a scaleable route for ICICI Bank. The partnership model pioneered by ICICI Bank attempted to address the following key gaps:

Centre for Micro Finance -MFI Strategy Unit-June 2006

A model was needed to separate the risk of the MFI from the risk inherent in the loan portfolio

The MFI must still share with the bank the risk of the portfolio, up to a certain defined limit

To separate the risk of the MFI from the risk inherent in the microfinance portfolio. To provide a mechanism for banks to incentivize partner MFIs continuously, especially in a scenario where the borrower entered into a contract directly with the bank and the role of the MFI was closer to that of an agent. To deal with the inability of MFIs to provide risk capital in large amounts, which limits the advances from banks, despite a greater ability of the latter to provide implicit capital. The model has been conceptualized and executed with the following key characteristics: Loan contracts directly between bank and borrower. This feature is similar to the SHG--bank linkage model, and means that the loans are not reflected on the balance sheet of the MFI. The MFI continues to service the loans until maturity, however, so the bank relies on the MFI's field operations for collection and supervision. The key difference, then, between this and the financial intermediation model does not lie in the operating methodology (it is in fact identical), but in the manner in which the financing structure has been designed. This structure primarily attempts to separate the risk of the MFI from the risk of the underlying portfolio. Why is this important? Because when the bank lends to the MFI, it has no recourse to the underlying borrowers. On the other hand, if the bank lends directly to the borrowers without the funds entering the MFI's balance sheet, it has recourse to the borrowers. So, at least in theory, if the particular MFI goes bankrupt or closes down for any other reason, the bank can appoint another agency to recover the dues from borrowers. In addition, since the loans are not reflected on the balance sheet of the MFI, its own requirement for regulatory capital ceases to exist. Therefore, the lending paradigm shifts from being organization based to being asset based. This shift has crucial implications for rating, pricing and consequent marketability. Alignment of incentives with a first-loss guarantee structure. In order to preserve MFI incentives for portfolio quality in the new scenario where its role is closer to that of an agent, the structure requires the MFI to provide a guarantee (typically a `first-loss default guarantee') through which it shares with the bank the risk of the portfolio, up to a certain defined limit. A first-loss default guarantee (FLDG) makes the provider of the guarantee liable to bear losses up to a certain specified limit, say the first 10 or 20 per cent of loss on the portfolio. It is different from partial guarantees, where the guarantor is liable for a fraction of losses, say 50 per cent of all losses on the portfolio. In terms of incentive compatibility, an FLDG forces the guarantor to prevent any losses at all, as it is affected adversely right from the start. The quantum and pricing of the FLDG will reflect the operating capability and maturity of the MFI. In this model, the MFI collects a `service charge' from the borrowers to cover its transaction costs and margins. The lower the defaults, the better the earnings of the MFI as it will not incur any penalty charges vis--vis the guarantee it provides. Over a longer period of time, returns from the partnership model can thus permit the MFI to build its core Tier I capital through retained earnings. The bank receives a fixed amount of interest on its loan. It must be noted that the bank accepts a fixed pay-off, while passing on the dynamic benefits or losses of `higher than expected recovery' or `lower than expected recovery' (losses limited to the band defined by

Centre for Micro Finance -MFI Strategy Unit-June 2006

The leverage on the overdraft facility that the MFI is able to achieve improves tremendously

With the partnership model, ICICI Bank is working with more than 30 Indian MFIs with loans outstanding of around $150 million

the FLDG) to the MFI. The asset that the bank has thus acquired has a risk-return profile that is now similar to a AAA asset, for which secondary markets are easier to access. Transfer of implicit capital from the bank to the MFI through an overdraft facility. Executing the partnership model at scale begs one question -- how will MFIs provide the risk capital implied in the FLDG in large amounts? In other words, how does it resolve the capital issue for the MFI? ICICI Bank has evolved a very innovative structure that combines the provision of both debt as well as mezzanine finance in response to this challenge. Along with advancing the credit to meet the demand of the clients, ICICI Bank provides an overdraft (OD) facility to the MFI equivalent to the amount which the MFI is liable to provide as the FLDG. The OD represents funds committed, but not utilized, and is drawn only in the event of default. On default, the MFI is liable to pay a penal rate of interest on the amount drawn down from the OD facility. The OD facility in effect, assumes the character of mezzanine equity, permitting the MFI to leverage it for wholesale funding. In this manner, the bank enables the MFI to provide explicit capital. The bank's own implicit capital allocation alters only to the extent that it has to now allocate capital against the OD limit advanced to the MFI, separate from the capital that it will allocate against the microfinance portfolio. Given that the FLDG is a fraction varying between 5 and 20 per cent of the loan portfolio, the leverage that the MFI is able to achieve improves tremendously. This has implications for MFI profitability and therefore return on equity, without any real difference in operating methodology. With this structure that combines the provision of both wholesale funding and mezzanine equity, MFI growth is only limited by its capability to grow field operations in a sustainable manner. Financing ceases to be the binding constraint. This can be a key driver of scale for mature MFIs operating in an environment of high client demand. An MFI that expects to have very low rates of default can grow operations even as a `shell company' with no equity. All it needs is a robust operating methodology and long-term funds to finance the field operation until it breaks even. This model may prove critical in `unleashing' the wholesale funds of Indian banks. This excess of wholesale funds may be seen in a very high proportion of savings deposits being held by banks as investments in Government of India securities -- 40 per cent for the whole banking system as against a regulatory requirement of 25 per cent. As the partnership model scales up, there will be several aspects to analyse. For instance, by working with several MFIs in different geographical regions, can the bank derive diversification benefits and therefore reduce the capital allocation against the OD as well? How does providing insurance cover against systemic risks such as rainfall affect the OD/FLDG requirement? These are issues to research and track. With the partnership structure, ICICI Bank is working with more than 30 MFIs in India accounting for loans outstanding of approximately $55 million in December 2004. With the traditional financing structure (second model), ICICI Bank's lending to MFIs did not exceed $5 million in 2001-02. The vision is to work with 200 MFIs in the country, each serving half a million clients each with average loan sizes of $60. Securitization -- paving the way for capital markets access Microfinance assets originated under the partnership model facilitate participation of a wider investor base. This is achieved through the process of

Centre for Micro Finance -MFI Strategy Unit-June 2006

Microfinance assets originated under the partnership model can be offered to a wider investor base

Insurance can be bundled with the loans to cover borrowers' critical illness, death, accident and rainfall failure

securitization. This may involve sale of portfolio by the originating bank to another bank in the initial phases. When the microfinance pools become larger in size, issuance of securities that are backed by microfinance assets become conceivable. Securitization is a process through which homogeneous illiquid financial assets are pooled and repackaged into marketable securities. Securitization involves isolation of specific risks, evaluation of the same, allocating the risks to various participants in the transaction (based on who is best equipped to mitigate the respective risks), mitigating the risks through appropriate credit enhancement structures and pricing the residual risk borne by the originator. The aim of securitization typically is to ensure that repayment of the securities issued to investors is dependent upon the securitized assets and therefore will not be affected by the insolvency of any other party, including the entity securitizing the assets. This reveals why a partnership model of financing that separates risk of the MFI from the risk of the microfinance loans is a pre-cursor for securitizing microfinance loans. An analysis of microfinance portfolios against desired attributes for securitization reveals a reasonably good fit. Some of the pertinent characteristics are described below: Microfinance assets represent weekly steady cash flows from clearly identifiable borrowers. An analysis of the past portfolio details of the top MFIs in India reveals portfolio losses to be consistently under 5 per cent. This can be explained by various factors: joint liability models preventing adverse selection and moral hazard issues, high levels of supervision by MFIs and alternative sources of finance for the poor being non-existent or very expensive. Microfinance portfolios comprise of several small loans made to individuals for diversified purposes, including livestock, small businesses and consumption. Loan sizes per client typically range from $25--500. Recent initiatives in bundling insurance products with the loan contract that result in the insurer bearing loan liability when specified events occur (critical illness, death, accident and rainfall failure) further contribute to de-risking the portfolio against systemic and large idiosyncratic shocks that might impair the individual's ability to pay. Ideally for securitization, receivables that are being securitized must be periodic. With most MFIs following weekly repayment schedules, this criterion is easily satisfied. Most MFIs advance loans that are fairly similar in terms of maturity, interest rates and risk profiles. For example, almost no MFI in India directly finances crop farming which presents a significantly different profile (often adverse) from lending small amounts for livestock. Credit enhancement for microfinance portfolios For several banks and investors, microfinance represents an unfamiliar asset class. While microfinance practitioners have evolved a methodology over several years based on client- and household-level insights, these may not be obvious. For example, when ICICI Bank was in dialogue with rating agencies for the rating of its microfinance portfolio, some of the questions that were raised included, `Do poor households have any surplus to repay loans in the first place? What happens if the business for

Centre for Micro Finance -MFI Strategy Unit-June 2006

In order to bring in `mainstream investors', it may be necessary to provide additional credit enhancement

The partnership model of financing allows the MFI to build outreach much more rapidly

which the money is borrowed fails or the cow financed by the loan dies? What if there is a serious rainfall failure?' Therefore, in order to facilitate securitization and the participation of mainstream investor', it may be necessary to provide `additional comfort' to these investors through various forms of credit enhancement. Over a period of time, with sufficient familiarity and experience of investing in microfinance portfolios, the levels of credit enhancements can be tapered. Credit enhancements could be: Originator provided. Here, the originator (MFI or the bank) provides a guarantee or cash collateral either in part or full. Structural. Structural credit enhancement is achieved by distribution of risks among investors with different risk appetites and through tranched securitization. Tranching is possible with issuance of multiple tranches of securities with a pre-determined priority in their servicing, whereby first losses are borne by the holders of the subordinated tranches. In certain structures, the originator retains a subordinate tranche. In this case, the subordinate tranche could be retained by the originating MFI or bank. Third-party provided. There could be specialized third-party entities that provide credit enhancement. In the USA, there are examples of third-party agencies contributing significantly to catalysing new asset classes. For mortgage financing, the relevant examples are the Federal Housing Administration and the Federal Home Loan Mortgage Corporation. The objective was to develop a secondary market in mortgage financing. The Small Business Administration (SBA) similarly issues guarantees, pools loans given to small businesses and securitizes them for sale to investors. An effort to create a specialized entity that will provide credit enhancement for microfinance portfolios is underway in India. Modelled along the lines of the agencies described in the previous paragraph, Grameen Capital India is a joint initiative of ICICI Bank and the Grameen Foundation, USA. There is scope for the participation of equity investors as well as multilaterals to catalyse this effort. The opportunity for MFIs A model of financing that starts with the partnership model of financing and culminates in securitization significantly relaxes the capital constraint that was outlined in the introductory paragraph. Firstly, it permits the MFI to build outreach much more rapidly. This is facilitated by certain aspects of the model -- one, the overdraft facility overcomes the problem of explicit capital shortfall. Secondly, the MFI is able to achieve more leverage for a given amount of risk capital because the lending bypasses the MFI's balance sheet. The process of securitization releases capital for the origination of fresh assets at a greater frequency than what would be possible if MFIs held assets on their own balance sheet to maturity. The MFI also experiences `rating arbitrage', in that it is conceivable through structuring and credit enhancement for the asset pool to obtain a rating that is higher than what a generic organization rating would have yielded. The improved rating results in lower cost of financing for the asset pool than would have been possible otherwise. If we just examine ICICI Bank's own experience, the lending rate has fallen from an average of 12 per cent to 8.5 per cent (without including the commitment fee on the overdraft facility) with the transition in financing structure. MFI then make an additional mark-up of 8--18 per cent to cover transaction costs.

Centre for Micro Finance -MFI Strategy Unit-June 2006

Box 1. Microfinance securitization Share, a Grameen Bank replicator, has been one the leading MFIs in India with a good track record, growth rate and scale of operation. In two transactions with Share, ICICI Bank has securitized the receivables of microfinance loans from Share amounting to $5.25 million consisting of loans made by Share to microfinance clients. ICICI Bank bought this microfinance loan portfolio against: A consideration calculated by computing the NPV of receivables amounting to $5.25 million at an agreed discount rate. Partial credit protection provided by Share to ICICI Bank in the form of a first-loss default guarantee amounting to 8 per cent of the receivables under the portfolio. Subsequently, ICICI Bank sold the securitized portfolio to a private sector bank in India. Implications for the financial system Beyond the benefits to the MFIs outlined above, pursuing capital markets access for microfinance also has implications for the larger financial system. In the Indian context, it creates a sustainable model to originate `priority sector assets' which are in short supply. The Reserve Bank of India's report on Advances to Agriculture and Weaker Sections (2004) reveals that only four out of 30 private sector banks and seven out of 27 public sector banks met the target for lending to `weaker sections'. It catalyses development of a secondary market for microfinance whereby some entities specialize as originators and others emerge as buyers. Therefore, those with unique competencies to originate assets can potentially earn a premium for that function. On the other hand, banks with no originating capability, instead of building branch networks from scratch, can rely on existing originators. Microfinance securitization may also create new asset classes for investors. This refers not only to the underlying microfinance loans, but also potentially the pool of overdraft facilities that represent MFI risk. Creating a conducive environment for capital markets access A number of areas need further examination for these processes to unfold. Servicer risk and capabilities. Servicer risk is the risk of the MFI defaulting on its commitment to monitor the programme and collect receivables. This might arise from a variety of factors, including bankruptcy. The separation between portfolio risk and MFI risk has not been considered possible among analysts of microfinance in the past, because the quality of the loan is as much a function of the organization as of the borrower, such that it has been assumed that you could not transfer the loan servicing from one organization to another in the case of failure of the originating organization. However, there have been no studies to look at the contribution of various MFI supervision aspects (separate from joint liability effects, for instance) on portfolio performance. This is an aspect that will have to be studied in greater detail. The servicer risk debate does raise serious challenges for the MFI in terms of operating capabilities. The systems and procedures of the MFI are critical in determining on-going servicing ability. Emerging MFIs need to invest sufficient time and energy in ensuring that their MIS systems are able to provide detailed portfolio data with minimum time lags.

The partnership model should help banks achieve their target for lending to `weaker sections'

Centre for Micro Finance -MFI Strategy Unit-June 2006

Agency models for increasing access to other financial services may be possible as well

Training continues to be an important factor for MFIs. As they gradually move into a phase of rapid expansion with standard operating models, they must add seamlessly to the field workforce without great delays. Legal format. The legal format chosen by the MFI must be one that permits it to assign loans, provide guarantees and retain earnings, for the model to be executed smoothly. It appears that a trust or society format is less suited, given this objective. Rating agency participation. Most securitization issues are rated by an accredited credit-rating agency. The existence of professional rating provides comfort to investors. The rating applies to the securities that are issued to investors and indicates the likelihood of payment of interest and payment of principal in full and on time. Rating agencies need to engage with the microfinance sector from a wholly commercial perspective in order to give investors sufficient comfort. Pre-operating expenditure for new MFIs. The partnership model tries to solve the problem of wholesale funds. However, it does not solve the problem of working capital finance for MFIs -- e.g. to expand into new areas, or for early stage financing of emerging MFIs. There may be a role for venture capital companies and other investors who are willing to take MFI risk initially and then exit through arrangements with the banks interested in financing these MFIs or with MFIs directly. Regulatory support. This is necessary in order to take the partnership model to scale. This could involve recognizing the MFI engaged in a partnership model as playing a facilitative role and exempting it from needing an NBFC licence. The regulator has taken an important step in the development of a secondary market by giving securitized paper the same recognition as directly originated assets vis--vis priority sector requirements. This will eventually lead to specialization of a few players (with comparative advantage in origination capabilities) as net originators. Similar to the partnership approach for credit, the Regulator may explore supporting agency models for increasing access to other financial services as well, specifically savings and insurance. This has the advantage of leveraging the risk-management capabilities of mainstream providers (like banks, mutual funds and insurance companies) while relying on the operational strengths of MFIs. Widening the range of services facilitated by the MFI also has positive implications for its operational self-sufficiency. For instance, Prudential ICICI Mutual Fund is exploring a product that will replicate the features of a savings bank account through a money market mutual fund, which is then offered to microfinance clients through the MFI. Individual accounts are maintained by the company for reconciliations and the MFI plays a purely facilitative role in terms of cash handling and data capture. With the emergence of the partnership and similar models, it becomes conceivable to think of MFIs emerging as thinly capitalized profitable local financial institutions playing a vital role in distribution of financial services to the poor and interacting with multiple providers in the back-end. The entry and growth barriers cease to be that of capital, but those of operating capabilities. This is a challenge that can be addressed with the support of training and donor agencies.

REFERENCES Mor, N. and S. Sehrawat (2004) `Sources of infrastructure finance',

iciciresearchcentre.org

Centre for Micro Finance -MFI Strategy Unit-June 2006

10