Thus donors/shareholders that are providing funds to ongoing programs need to demand financial information and provide appropriate technical assistance on a regular basis, with a special emphasis on employing standardized practices; for instance, in reporting loan recovery, income and costs, and portfolio risk (N. Bhatt, 1998). Lack of accurate financial reporting is the key reason why program inefficiencies are often passed on unchecked, year after year, as a result of which institutional failure often comes as a surprise to many stakeholders. Nonetheless, donors/shareholders also have to be cautious in not imposing operational efficiency as the only criterion in evaluating a program. Otherwise, program officials may have an incentive to divert their efforts away from serving the poor, who might be the original target population.

Program inefficiencies also can result if the clients face incompatible incentives. It is now well known that clients value convenience and flexibility in accessing financial services more than they value paying a low-interest rate for such services (Otero & Rhyne, 1994). Programs that have not assessed the characteristics and needs of their clients often impose high transaction costs on them, as a result of which client noncooperation renders their operations nonviable (Bhatt & Tang, 1998b). One example is the difficulties encountered by a microfinance program that tried to implement the group lending model in Burkina Faso. A study by the World Bank (1997) indicates that most of the women surveyed preferred individual to group loans, to avoid possible social animosity among themselves. In some communities where the peer group method is socially acceptable, prospective borrowers may seek out the least risky peers to form groups. Hulme and Mosley (1996) provide evidence of such "social exclusion" in Sri Lanka's SANASA and Bangladesh's BRAC initiatives. In these programs, members identified some people in their villages as being too poor, and hence unsuitable for group membership.

These and other similar experiences from around the world indicate that when borrowers face incompatible incentives, forcing them to form groups would create unnecessary burdens on them or undesirable consequences for the program, without enhancing the operational efficiency of the program (Bhatt & Tang, 1998a, 1998b).

Conclusions

The future success of microfinance as a development tool will depend heavily on the ability of public, private, and nonprofit organizations to develop a diverse set of institutions to meet the different financial needs of various segments of low-income populations. The design of such institutions must be informed by a thorough understanding of the causes of poverty, as well as the specific reasons for the lack of viable formal financial intermediaries in specific communities. What are the major obstacles faced by the target populations, especially the entrepreneurial and marginally self-employed poor, in their attempt for economic and social advancement? To what extent are these obstacles related to the lack of financial services? Would potential clients need additional nonfinancial services, such as training, technical assistance, and health and human services, to be able to make productive use of the loan, and to what extent would such services impact a client's cash flow and a program's subsidy dependence? A focus on such basic market research and needs assessments must be emphasized in feasibility studies before the plan for any microfinance program is drawn up.

Microfinance programs should not be designed without considering their opportunity costs. As most microfinance programs are still subsidy dependent, a dollar spent on supporting a microfinance program means a dollar less for another potential development intervention. To date, little systematic study has been done to estimate the external efficiency of microfinance programs. Part of the reason for the lack of such studies is probably the difficulty in quantifying and comparing the benefits and costs of alternative interventions. Although exact quantification may not be feasible, or even desirable, in many cases, it is important for the development community to understand, at least in terms of qualitative models, the major types of costs and benefits that must be considered when evaluating the external efficiency of a microfinance program.

Last, but not least, as the development community begins to emphasize internal efficiency for microfinance programs, more research is needed to examine what incentive structures are most conducive to such operational efficiency. What institutional arrangements or governance structures are needed to achieve incentive compatibility among donors/shareholders, program managers, and clients? The New Institutional Economics literature (Eggertsson, 1990; Furubotn & Richter, 1997; Lin & Nugent, 1995; Ostrom, Schroeder, & Wynne, 1993; Tang, 1992; Williamson, 1985), which emphasizes information, transaction cost, and incentive problems in economic organizations (Bhatt & Tang, 1998a; Prendergast, 1999), can potentially offer useful perspectives, but more work is needed in adapting those perspectives to the context of microfinance delivery.

Nitin Bhatt is Manager at Grant Thornton LLP's Entrepreneurial Consulting Services Practice in Los Angles and former Executive Director of the University of Southern California's (USC) Business Expansion Network. He is the author of Inner City Entrepreneurship Development: The Microcredit Challenge (ISC Press, 2001), and his work in the area of enterprise development has been published in such journals as Harvard Business Review, World Development, Economic Development Quarterly, Public Administration and Development, and International Journal of Technical Cooperation. He has served as an advisor or board member for various nonprofits, foundations, and government entities involved in entrepreneurship and economic development. A Rotary Foundation Scholar, Mr. Bhatt holds an MBA in entrepreneurship and venture management and a Ph.D. in policy, planning, and development from USC.

Shui-Yan Tang is associate professor and Director of the MPA Program at the USC School of Policy, Planning, and Development. He has done extensive theoretical and empirical work on community organizations, microfinance, and environmental policy. He is the author of Institutions and Collective Action: SelfGovernance in Irrigation (ICS Press, 1992), and articles in such journals as The China Quarterly, Land Economics, Public Administration and Development, Public Administration Review, Journal of Public Administration Research and Theory, Public Productivity and Management Review, and World Development.