When asked if he would like to have dinner at a highly-regarded restaurant, Yogi Berra famously replied “Nobody goes there anymore. It’s too crowded”. This contradictory situation of very low take-up combined with large overall use is common with some financial products – for example, the response rate to direct mail credit card solicitations had fallen to 0.6 percent by 2012, yet lots of people have credit cards.
It is also a situation we recently found ourselves in when working on a financial education experiment in Mexico with the bank BBVA Bancomer. They worked with over 100,000 of their credit card clients, inviting the treatment group to attend their financial education program Adelante con tu futuro (Go ahead with your future). Over 1.2 million participants have taken this program between 2008 and 2016, yet only 0.8 percent of the clients in the treatment group attended the workshop. A second experiment which tested personalized financial coaching also had low take-up, with 6.8 percent of the treatment group actually receiving coaching.
In a new working paper (joint with Gabriel Lara Ibarra), we discuss how the richness of financial data on clients allows us to combine experimental and non-experimental methods to still estimate the impact of this program for those clients who do take up the program.
- Data Colada on how to properly pre-register a study: “it may be helpful to imagine a skeptical reader of your paper. Let’s call him Leif. Imagine that Leif is worried that p-hacking might creep into the analyses of even the best-intentioned researchers. The job of your preregistration is to set Leif’s mind at ease. This means identifying all of the ways you could have p-hacked – choosing a different sample size, or a different exclusion rule, or a different dependent variable, or a different set of controls/covariates, or a different set of conditions to compare, or a different data transformation – and including all of the information that lets Leif know that these decisions were set in stone in advance”…but on the other hand “it should contain only the information that is essential for the task at hand… We have seen many preregistrations that are just too long… you don’t need to say in the preregistration everything that you will say in the paper. A hard-to-read preregistration makes preregistration less effective…” – comes with a nice example table of what bad specifications and good specifications look like.
Yesterday I posted a round-up of the research presented at NEUDC, a major conference on development economics. Although most economic research aspires to uncover principles relevant across multiple contexts, empirical research happens at a place and time. I mapped out the distribution of research presented at NEUDC, fully recognizing that this makes no claim to be representative of the profession as a whole.
Below, I charted the number of studies per country (for all countries that had at least two studies). If a study used data from multiple countries (up to four), I counted each of them. If a study used a data set that spans 30 countries, I didn’t use it.
Did you miss this year’s Northeast Universities Development Consortium conference, or NEUDC? I did, unfortunately!
NEUDC is a large development economics conference, with more than 160 papers on the program, so it’s a nice way to get a sense of new research in the field.
Thankfully, since NEUDC posts submitted papers, I was able to mostly catch up. I went through 147 of the papers and summarized them below, by topic. If a paper you loved or presented isn’t in the rundown, feel free to add a brief summary in the comments. (Why 147 instead of 160? I skipped a few macro papers and the papers that weren’t posted.)
These links should take you to your topic of interest: Agriculture, cash transfers and asset transfers, credit and insurance, crime, conflict, violence, and war, culture, norms, and corruption, education, elections and political economy, firms, governance, bureaucracy, and social capital, health (including WASH), jobs (including public works), marriage, methodology, migration, mobile phones and mobile money, poverty, inequality, and shocks, psychology, taxes, and traffic.
This week, I leave you with this short 2003 paper in the Journal of Economic Perspectives by Kaushik Basu. It both follows somewhat from my last post, is related to the day's news, and relevant for thinking about principles for intervention in labor markets for a host of issues that our colleagues deal with in developing and developed economies...Here is the abstract - but you can read the paper in 30 minutes...
- A simple growth chart poster has a surprisingly large impact on reducing stunting in Zambia according to this IPA brief.
- David Rinnert and Liz Brower of DFID provide a typology of ways evidence is used by policymakers on the Oxfam from poverty to power blog – with examples ranging from the more nebulous “help decision-makers acknowledge the full body of evidence “ to the more concrete, but still hard to pin-down “fed directly into the decision-making”.
- No Sugar: MIT news summarizes and provides background into Noam Angrist’s attempt to scale up Dupa’s Sugar Daddies work in Botswana
- A new journal for replication studies in economics- International Journal for Re-Views in Empirical Economics (IREE)
With tomorrow being Halloween, I thought it perfect timing to discuss a paper about death and zombies. Small firms are an important source of income for the poor in developing countries, and the target of many policy interventions designed to help them grow. But we don’t actually know much about their death, with no systematic evidence available as to the rate of small firm death, which firms are more likely to die, and why they die. Indeed firm death often ends up being hidden in the attrition numbers of much of our data, and out of 35 published RCTs on interventions for small firms in developing countries, only 13 either report a firm death rate or look at death as an outcome.
My new working paper (ungated version) (with Anna Luisa Paffhausen) aims to provide systematic evidence on small firm death in developing countries. We spent several years cleaning and putting together data on more than 14,000 small firms from 16 firm panel surveys in 12 countries, enabling estimation of the rate of firm death over horizons as short as 3 months and as long as 17 years. Detailed questions added to nine of these panel surveys also enable us to dig deeper into cause of death.
- job market series 2017