1. Microfinance Subsidy: Back before there were impact evaluations the heated discussions in microfinance were about costs and subsidies (and business model, which is really a conversation about cost and subsidy). Those conversations have died down as the focus shifted to impact evaluations--appropriately!--but cost and impact are equally important when it comes to policy choices. Cull, Demirguc-Kunt, and (our very own) Morduch have a new paper that does the painstaking work to accurately measure subsidy in microfinance. They find that subsidy is pervasive and long-lasting, but small: meaning the modest impact of microfinance has to be viewed in terms of even more modest cost. I could write the whole faiV this week just on findings from the paper which is another way of saying: read it! Bob Cull has a short overview of the findings here for those with short attention spans, or a day full of meetings.
2. But Wait, There's More Microfinance: While most eyes have been turned to tracking the growth of digital financial services, the microfinance industry in India is growing rapidly again. The industry association reports 60% year-over-year growth, with the majority coming from the large incumbents like SKS and Ujjivan. Apparently the banking correspondent model is playing a significant role in growth. Let me pause for a moment to roll my eyes at the finding that clients say that 94% of loans are for "income generating activities."
Meanwhile, Jonathan Morduch has a review of Lesley Sheratt's new book on achieving an ethical balance in microfinance, a balance that a 60 percent growth rate calls into question.
3. Financial Inclusion: Back in August I noted a paper about low-take up of no-frills savings accounts in a number of countries. A new paper from Brune et al. using the Malawi commitment savings experiment data to look at what happens with account usage and spending composition when funds are direct deposited into accounts or delivered in cash, and the delay between when the household learns the deposit is coming and when it is delivered. Higher account balances for direct deposit persist for only a few weeks, and there is no meaningful effect on the composition of spending, which, they say, suggests "that households manage cash effectively without the use of formal financial products."
And in a connection that perhaps only my brain makes, here's a new paper about a job training program in Argentina that tracks effects on employment for 4 years. Gains are large in the short run but fade out over time. The effect seems to come from "persistence of employment." In case you're wondering here's the connection I make: financial inclusion matters most to those with income, and the benefits of financial inclusion are related to the volatility of income in the short- and long-term, which, of course, is affected by the persistence of employment.
4. Education: Like financial inclusion, education is about more than access. Liberia, where 60% of children aren't in school, and only 20% of women who reached fifth grade can read a sentence, is experimenting with turning over some schools to Bridge International Academies (not unlike charter schools in the US). Here's a story about an ActionAid visit to one of the schools and the aftermath. There is an external impact evaluation underway of Liberia's experiment. As Justin Sandefur notes in describing the situation and the evaluation, "when the status quo is unacceptable, experimentation is an obligation." But which educational experiments are obligated? 3ie is about to release a systematic review of education intervention evaluations. Here's a paper from David Evans from earlier this year where he notes there are a lot of "systematic reviews" that have little overlap in their systems and come to very different conclusions about which experiments to implement.
5. Evidence-Based Policy: It always comes back to policy eventually, even among anarcho-capitalists. The Liberia situation makes me think of my interview with Angus Deaton where he talks about the issues of where, on whom and what types of experiments are conducted. Here's Deaton's and Cartwright's newest paper on RCTs. Here's the new second edition of Impact Evaluation in Practice (free!). And the Urban Institute is launching a new Evidence-Based Policymaking Collaborative (with Brookings, AEI and the Pew-MacArthur Results First-Initiative) "to create tools to inform evidence-based policymaking at all levels of government." I wonder if a primer on Deaton & Cartwright will be one of those tools?