Microfinance institutions (MFIs) provide a public good: they provide income-creating financial services to un-bankable people. If MFIs create and deepen markets where none existed before, there may be a case for public support. While subsidies are generally not favorably seen in financial sector development, being difficult to target and possibly distorting the local financial market, there may be situations where the net social benefits of micro-finance may exceed those of not doing anything and of alternative anti-poverty programs. Under such circumstances longer-term public support may be justifiable. This book is based on a study of forty-five MFIs carried out by ILO, in partnership with the Universities of Geneva and Cambridge. The application of factor analysis and cluster analysis shows that MFIs form clusters in terms of social and performance. Within each cluster there is one institution that is most efficient on both scores. Public support should ensure that the relative efficiency of MFIs is enhanced, it should not prod MFIs to modify their mission and position between poverty outreach and profitability.