Small loans have helped millions of people but have reached only a small fraction of the poorest, with relatively high interest rates.

Judge as of 2015. The claim has three parts: (a) microloans have helped millions; (b) their interest rates are relatively high; (c) they have reached only a small fraction of the poorest. True: microfinance research up to 2015 (for example CGAP, the World Bank, or randomized evaluations) supports all three, meaning tens of millions of borrowers with at least some benefit, interest rates above those of mainstream lending, and limited reach into the poorest households. Mostly true: all are broadly supported with a minor overstatement. Mixed: one of the three parts is contradicted, for example evaluations find no measurable benefit to borrowers. Mostly false: two parts are contradicted. False: the evidence shows microloans reached most of the poorest at low rates.

Set before the research began.

1. E1 Six Randomized Evaluations of Microcredit: Introduction and Further Steps American Economic Journal: Applied Economics (Banerjee, Karlan, Zinman) · January 2015

   “Summarizing and interpreting results across studies, we note a consistent pattern of modestly positive, but not transformative, effects.”

2. E2 Mapping Pathways out of Poverty: The State of the Microcredit Summit Campaign Report, 2015 Microcredit Summit Campaign (A Project of RESULTS Educational Fund) · 2015

   “At the end of 2013, the microfinance community reached 211 million clients, 114 million of whom were living in extreme poverty.”

3. E3 Helpdesk Research Report: Evidence on microcredit for the ultra-poor GSDRC (Governance and Social Development Resource Centre) for DFID · 16 January 2013

   “Nevertheless Cull et al (2009) find that in practice it appears most non-governmental MFIs charge more than the Nobel Prize winner Mohammed Yunus’s suggestion of an optimum interest rate of 10 per cent or at the most 15 per cent.”

4. E4 Helpdesk Research Report: Evidence on microcredit for the ultra-poor GSDRC (Governance and Social Development Resource Centre) for DFID · 16 January 2013

   “Over the years evidence has mounted that the ultra-poor are commonly not being reached by mainstream microcredit services; while microfinance does serve very poor clients, in general they do not participate in large numbers (Simanowitz and Walter 2002).”

Part (a) is supported: E2 reports 211 million microfinance clients at the end of 2013, and E1, a synthesis of six randomized evaluations, finds modestly positive but not transformative effects. Part (b) is broadly supported by E3, which says most NGO microfinance lenders charge more than the 10–15% rate Yunus suggested, though it does not compare them directly with mainstream lending rates. Part (c) is supported by E4, which finds the ultra-poor are commonly not reached by mainstream microcredit. E2 reports 114 million clients living in extreme poverty but gives no share of the world's poorest reached, so it neither confirms nor refutes "only a small fraction" [E2].