Research Papers  (Comments are most welcome!)


ABC, 123: The Impact of a Mobile Phone Literacy Program on Educational Outcomes (with Jenny Aker, Tufts and Travis Lybbert, UC Davis) (published in: American Economic Journal: Applied Economics, American Economic Association, vol. 4(4), pages 94-120, October, 2012.)

This paper reports the results from a randomized evaluation of an adult education program in Niger, in which some students learned how to use simple mobile phones (Project ABC). Students in ABC villages achieved test scores that were 0.19-0.26 standard deviations higher than those in standard adult education classes, and standardized math test scores remained higher seven months after the end of classes. These results suggest that simple information technology can be harnessed to improve educational outcomes among rural populations.


Guns and Roses: The Impact of the Kenyan Post-Election Violence on Flower Exporting Firms (with Rocco Macchiavello, Warwick and Ameet Morjaria, Harvard) (Revise and resubmit, Review of Economics and Statistics) [paper]

This paper studies how export-oriented firms react to ethnic violence and civic unrest. Predictions derived from a model of firms reaction to violence are tested using Kenya flower exporters during the 2008 post-election crisis. The violence reduced exports primarily through the absence of workers and resulting production losses. Interestingly, the institutional arrangements that firms developed to export in integrated non-traditional agricultural value chains matter: firms with direct contractual relationships in export markets and members of the business association had higher incentives and lower costs of reacting to the violence and suffered smaller production and workers losses. Model calibrations suggest that the average firm operated at a loss during the violence and absent workers suffered welfare losses at least three times larger than weekly earnings.


Learning without Teachers?  Evidence from a Randomized Experiment of a Mobile Phone-Based Adult Education Program in Los Angeles (with Jenny C. Aker, Danielle Miller, Karla Perez, Susan L. Smalley), July, 2014 (submitted)  [paper]

We evaluate the impact of a entirely mobile- phone-based adult education program (Cell-Ed) in Los Angeles. We find that students’ reading scores are substantially increased over a four-month period, equivalent to a 2-4 year increase in reading levels. Because there was substantial differential attrition between the treatment and control groups, I extended Lee (2009) bounds to a phase-in design, which led to a substantial tightening of the bounds. Both the substance and the method were the subject of this Development Impact blog post.

The Impact of Village Savings and Loans Associations on Household Welfare: Evidence from a Cluster Randomized Trial (with Helene Bie Lilleor, Jonas Helth Lonborg and Ole Dahl Rasmussen), September, 2014 (Submitted, Journal of Development Economics) [paper]

Using a cluster randomized trial we investigate the impacts of a community managed microfinance intervention aiming to do exactly this. We assess the impact of village savings and loans associations (VSLAs) in 46 villages from Northern Malawi. Based on a panel survey of 1700 households from the treatment and control villages we find that the intervention had positive and significant effects on savings, the number of meals consumed in a day and total household expenses. This effect is linked to an increase in the use of fertilizer and an increase in business income. These results are robust to using household-specific fixed effects and adjusted regressions.

Learning from Peers: Experimental Evidence of Group Learning in Senior High School in Ghana. (with Kim Lehrer) December, 2013 [paper]

This paper uses a framed field experiment to examine the role of peer effects and group composition in high schools in Ghana. Students are randomly assigned to practice Sudoku individually or in groups of sizes 2,3 or 4. We find that, on average, group assignment has large positive effects on achievement. However, group work penalizes the initially best performing group members. Peer effects operate within groups: initial peer achievement has a positive impact on student achievement. We find no differential benefits from being assigned to same-gender groups, nor between between groups of different sizes.

Selecting Good Caretakers: The Impact of Family Networks in Tanzania on Orphans’ Education. October, 2010 [paper]

This paper analyzes the consequences of parental mortality in Sub-Saharan Africa.  Family networks are shown to be important institutions to mitigate the negative consequences of adult mortality on the education outcomes of orphan children. In particular, this research models orphan caretaking as an allocation problem within the family where better caretakers are selected from the available family members. Using a purposely collected data set on caretakers and non-caretakers we show that this selection process increases orphan education by about one year, highlighting the value of functioning family.


Mobiles and mobility: The Effect of Mobile Phones on Migration in Niger. (with Jenny Aker, Tufts and Michael Clemens, Center for Global Development) [Draft]

This paper studies why large spatial wage gaps across West Africa persist in the face of substantial migration. Indeed, in the areas we study, around 40% of households engage in seasonal migration. The persistence of wage gaps can be due to differences in average returns to labor, credit and insurance constraints, as well as information asymmetries. We provide evidence that information asymmetries are large, by analyzing the impact of two separate randomized interventions providing mobile phones in rural villages in Niger. We find that the provision of mobile phones which reduced search costs increased migration substantially – from an already high level of 40% of households to 47%, and that the impact is strongest for wealthier households. The latter result suggests that mobile phones did not relax credit constraints.


The Causal Impact of Schooling on the Distribution of Achievement: Evidence from Ghana. (with Kim Lehrer) June, 2012  [Draft]

We estimate the impact of schooling on achievement, using a policy change that extended high school by one year in Ghana. On average, the additional year of schooling leads to an improvement of about 2/3 of a standard deviation for the aggregate grade. The strongest impacts are found on the middle part of the distribution. Unusually, we also observe a control group which was not affected by this change, Nigeria. The high school leaving exams in both Ghana and Nigeria in any given year are identical for a subset of subjects. The change led to a 16 percentage points higher passing rate for students who received an extra year of education in Ghana, relative to previous years and relative to Nigeria.


Research in Progress


“Education and Rationality” (joint with Annemie Maertens, Sussex University).

We investigate the impacts of an adult literacy program on rationality in decision-making. In particular, we find that illiterate women in one of the poorest districts in Uttar Pradesh display low rationality (as measured by consistency with the Generalized Axiom of Revealed Preferences). We find some evidence that participating in an adult literacy program leads to a higher degree of consistency in decision-making.  


“Caught in the Middle of a Revolution: The impact of the Arab Spring in Libya on Migrants from Niger” (joint with Jenny Aker, Tufts).

We study the impact of the recent revolution in Libya on migrants from Niger, their return and their families at home. A first round of data was collected in December 2010, before the revolution started, a second round was collected in May 2011, and more detail on the return of migrants will be collected in December 2011.