While most of such impact studies have acknowledged one or more methodological problems such as selection bias, lack of control groups, and inability to gather longitudinal data, they all seem to have emphasized the need to measure changes in such dependent variables as incomes, assets, productivity, and general well-being as a result of inputs such as financial and nonfinancial services. Indeed, they all seem to echo Hulme and Mosley's argument that "the ultimate test of any institution is not whether it exists or sustains itself, but whether it manages to do something useful" (1996, p. 86).

Assessing Performance at the Institutional Level

But the above approach has been criticized by those who believe that effectiveness of such services as savings and credit needs to be gauged not by assessing their specific changes or impacts on borrowers but by their impact on the financial system at large. According to Rhyne, "If program evaluations can be freed from the burden of proving that finance matters.. they can concentrate on evaluating the quality of the services and their institutional settings" (1994, p. 107).

Such an alternative approach to performance evaluation is captured in Jacob Yaron's framework for assessing the success of microfinance institutions. Yaron (1994) recommends employing two key indicators of program success-institutional outreach and financial sustainability. In the spirit of his approach, institutional outreach is measured in terms of its breadth as well as depth. While the breadth of outreach is assessed by measuring such variables as the number of people who are provided financial services and the kinds of instruments offered to them, the depth of outreach is generally measured by the average loan size and the gender distribution of the portfolio. In this regard, the smaller the average loan size and the more female clients, the higher the confidence that services indeed are being provided to a poorer clientele.

As part of this approach to program evaluation, Rhyne (1994, p. 111) has recommended the assessment of another variable-the quality of service provided. According to her, "The strongest and simplest test of service quality.. is willingness of clients to pay, which serves as a basic indicator of the value or the benefit of the service." This market test is supplemented by measuring other quality variables such as client transaction costs and other service terms like collateral requirements for loans and liquidity of deposits.

A second, widely employed measure for assessing the success of a microfinance program is its level of sustainability. According to Yaron, selfsustainability "is achieved when the return on equity, net of any subsidy received, equals or exceeds the opportunity costs of funds" (1992, p. 5). The extent to which an institution has achieved self-sufficiency is measured by the calculation of an SDI, which is "a ratio that measures the percentage increase in the average on-lending interest rate required to compensate..for the elimination of subsidies in a given year while keeping its return on equity equal to the approximate nonconcessional borrowing cost" (Yaron, 1992, p. 5).

Yaron's approach to assessing the success of microfinance programs has been widely adopted. For example, it was employed by Gonzales-Vega et al. (1997) in their study of the BancoSol program in Bolivia; by Khandker, Khalily, and Khan (1995) in their study of the Grameen Bank; and by Christen, Rhyne, and Vogel (1994) in their study of 11 microfinance programs in Indonesia, Bangladesh, Kenya, Senegal, Niger, Costa Rica, Columbia, the Dominican Republic, and Bolivia.