Week of January 16, 2017

Editor's Note: I've been traveling so much at the beginning of the year that I literally forgot that last Friday was Friday. Rather than publish a few days late, I decided to save up for an extra large faiV this Friday, figuring that many of you will appreciate lots of fodder to distract from the events of the day.

1. In Memoriam: In the last edition, I linked to some remembrances of Tony Atkinson. The philosopher Derek Parfit died the same day as Atkinson, and was equally concerned with inequality though in a quite different way. Here's a 2011 profile in the New Yorker by Larissa MacFarquhar that provides a very good overview of Parfit's thinking (by the way, if you haven't read MacFarquhar's Strangers Drowning, it is probably the ne plus ultra of counterprogramming today). Here's Dylan Matthews remembrance and explainer at Vox--with the truly remarkable note that a volume of essays discussing his work On What Matters was published before the book itself came out. Here is a review of Reasons and Persons from 1984 and Parfit's essay "Why Anything? Why This?" both from LRB.

2. Reproduceability, Replication and Meta-Analysis: There are all sorts of arguments about what the defining features of science are, but I think most of them include reproduceability and the ability to make accurate predictions. At the ASSA meetings there were a number of sessions on these issues in economics. Here's a look at published replications in development economics and an overview of the state of replication in the field in general. Here is Eva Vivalt's review of the dispersion of estimates of impact in development impact evaluations (in lieu of the in progress paper she presented on the rate of false positives and false negatives that builds on her earlier work). And here's Rachael Meager's job market paper on using Bayesian Hierarchical Analysis to understand and predict heterogeneity in treatment effects using the microcredit impact literature. And here's Ioannidis et al. on the limited power of most economics papers

3. Household Finance and Cashflow: Expect to hear a lot in this space about cashflow and how it affects households (hey did you know you can pre-order The Financial Diaries, the book about the US Financial Diaries work?). Here's a paper looking at how changing minimum payments affects how much people pay on their credit cards finding a large but not exclusive role for liquidity constraints, estimating that US consumers would save $570 million a year (or credit card companies would lose $570 million a year in earnings) if all companies used the most conservative minimum payment calculation. Here's a look at excessive sensitivity to payday in Iceland--people spend a lot more when they receive paychecks and it's not explained by illiquidity. Here's recent work in Malawi varying timing of paydays (weekly vs. monthly, Friday vs. Saturday) finding that monthly payments helped recipients save more. And here's a video of Meiping Sun discussing the very large effects of the New York City MTA imposing a $1 fee on the purchase of fare cards

4. Cash and Cashlessness: Two big questions remain in cash transfers (whether recipients use them well is settled as far as I'm concerned): when and where do conditions help, and what are the long-term effects of cash. Here's a study from Lesotho on soft conditions--a clear message that the cash should be spent on the "interest and needs of children" yields significant changes in how the funds are spent. Here's work by Baird, et. al. building on their earlier work examining the impact of conditions in cash transfers, finding that the benefits of cash transfers fade out after two years with interesting implications for the role and application of conditions. And while we're talking about cash (albeit in a very different application of the term), here's a Guardian piece on the exclusionary impact of moving away from cash in developed countries. Very soon we're going to have to develop new vocabulary to distinguish between "fully liquid" aid and literal, physical cash.

5. Microfinance Innovation: NextBillion is moving to a new editorial approach by focusing it's coverage on a different topic each month. January is about microfinance and features three pieces by me on innovation in microcredit. The latest is considers the implications of thinking about microcredit as a vaccine or as an antibiotic. All three of the pieces are extensions of my recent paper The Case for Social Investment in Microcredit, which if you haven't read it yet, you should. And sticking at NextBillion (and setting up the next item) here is Graham Wright's take on five ways microfinance needs to innovate in response to digital finance.

6. Tech and FinTech: I confess that this most recent paper on the impact of One Laptop per Child in Peru made me giggle: Getting a laptop makes it less likely that teachers use a more effective approach, less likely that kids do homework, and less likely that they do chores. The effect sizes are huge so I'm skeptical but I don't recall seeing so comprehensive a set of bad news in a long time. The news isn't nearly so bad for FinTech of course, but that doesn't mean that all signs are positive. Here's Graham Wright reflecting on some areas of concern, particularly the troubling statistics on digital credit out of Kenya. Bangladesh is lowering the ceilings on mobile money transactions (both amount and frequency) due to unspecified suggestions of "abuse." And to illustrate the huge gap in thinking about FinTech in the US and in developing contexts here are 10 fairly ridiculous predictions for FinTech in 2017.    

7. Algorithms: Wrapping up on the tech front: In contrast to the late '90s thinking about how the internet was going to a massive boon to consumers, this piece suggests that algorithms could dramatically reduce price competition.

8. Microenterprise: Have I mentioned that my book is out? Yeah, I guess I have. One of the interviewees is David McKenzie--we spend a lot of time talking about his experiments, with de Mel and Woodruff, to understand microenterprises, in particular work in Sri Lanka on whether it's possible to help microenterprises "level-up" or to change their growth trajectory. David's most recent paper on the issue is now out as well, finding that subsidies to employ workers don't have a lasting impact on employment, profitability or sales. The search for ways to help microenterprises continues.

9. Parsimony: Behavioral economics has illustrated the many, many impediments to "rational" and consistent behavior but it some ways those many, many ways inhibit our ability to study and understand specific issues: does each study need to do extensive work to understand each participants risk aversion, hyperbolic preferences, limited attention, etc. etc.? A new paper by Stango, Yoong and Zinman attempts to make such work a bit easier by looking for simpler ways to measure and model behavioral phenomena, with some success and some failure.

10. Deworming: No, there will never be a last word. Owen Ozier has a new paper on long-term effects of deworming that sheds a lot of light on the reasons why long-term effects can be large while short-term effects are small or negligible. And in case you haven't taken the time yet, here's David Roodman's two long posts (one and two) on the deworming literature--a great useful distraction if there ever was one.

Video: Esther Duflo's Ely Lecture on "The Economist as Plumber"

Video: Esther Duflo's Ely Lecture on "The Economist as Plumber"

Return to the Weekly faiV