James Aidini, and everyone on the Devil Bears. And most of all, I thank the people who have been ... more James Aidini, and everyone on the Devil Bears. And most of all, I thank the people who have been my bedrock of support during graduate school: my parents, Lynda and Brad Kerwin, my brother, Adrian Kerwin, my lifelong friend, Sam King, and my amazing partner, Audrey Dorélien. I could not have done it without you. This dissertation is the result of extensive empirical research conducted in Malawi and Uganda, which would not have been possible without the hard work and help of a number of people. I thank Christopher Nyirenda, Anderson Moyo, and Synab Njerenga for their excellent management of the fieldwork for the first chapter of this dissertation, and Ndema Longwe for his exemplary fieldwork management for the second chapter. Moffat Kayembe and Carl Bruessow from the Mulanje Mountain Conservation Trust played a crucial role in making the research for my second chapter possible, and Esperanza Martinez Maldonado provided excellent research assistance on that project. The randomized evaluation of the Northern Uganda Literacy Project would not have been possible without the collaboration of iii Victoria Brown, Bernadette Jerome, Benson Ocan, and everyone at Mango Tree Educational Enterprises Uganda; I deeply appreciate their role in making it happen. My research also benefited from the efforts of support staff at the
Data for: Income Timing and Liquidity Constraints: Evidence from a Randomized Field Experiment
Data from Brune and Kerwin (2019), "Income Timing and Liquidity Constraints: Evidence from a... more Data from Brune and Kerwin (2019), "Income Timing and Liquidity Constraints: Evidence from a Randomized Field Experiment". Please cite the manuscript when citing this dataset
In late 2013 we submitted an analysis plan for the one-year pilot RCT that was conducted during t... more In late 2013 we submitted an analysis plan for the one-year pilot RCT that was conducted during the 2013 school year. 1 This document expands and revises that analysis plan to include our plans for the analysis of the full, four-year RCT, which now runs from 2013 to 2016. We are submitting it on January 16, 2015, prior to the entry of any outcome data for the 2014 school year. 1 That document remains available on the AEA trial registry site. The details for it are as follows:
This paper tests a model in which responses to disease risks can be "fatalistic": higher risk bel... more This paper tests a model in which responses to disease risks can be "fatalistic": higher risk beliefs can lead to more risk-taking rather than less. This occurs because high risk beliefs raise the perceived chance that you are already infected, lowering the marginal cost of risk-taking. I test for fatalism by randomly providing information about the true (low) average risk of HIV transmission in Malawi. Just as the model predicts, the treatment causes sexual activity to rise slightly on average but decline sharply for people with high initial risk beliefs-especially those with high baseline levels of sexual activity.
This paper demonstrates the acute sensitivity of education program effectiveness to the choices o... more This paper demonstrates the acute sensitivity of education program effectiveness to the choices of inputs and outcome measures, using a randomized evaluation of a mother-tongue literacy program. The program raises reading scores by 0.64SDs and writing scores by 0.45SDs. A reduced-cost version instead yields statistically-insignificant reading gains and some large negative effects (-0.33SDs) on advanced writing. We combine a conceptual model of education production with detailed classroom observations to examine the mechanisms driving the results; we show they could be driven by the program initially lowering productivity before raising it, and potentially by missing complementary inputs in the reduced-cost version.
This paper studies workplace peer effects by randomly varying work assignments at a tea estate in... more This paper studies workplace peer effects by randomly varying work assignments at a tea estate in Malawi. We find that increasing mean peer ability by 10 percent raises productivity by 0.3 percent. This effect is driven by the responses of women. Neither production nor compensation externalities cause the effect because workers receive piece rates and do not work in teams. Additional analyses provide no support for learning or socialization as mechanisms. Instead, peer effects appear to operate through "motivation": given the choice to be reassigned, most workers prefer working near high-ability co-workers because these peers motivate them to work harder.
Temptation plays a key role in theoretical work on spending and saving in developing countries. T... more Temptation plays a key role in theoretical work on spending and saving in developing countries. The limited empirical evidence on its importance, however, suggests that cash transfers do not induce increases in temptation spending. This paper expands the evidence base by studying the effect of randomized exposure to temptation on spending decisions in rural Malawi. Consistent with the cash transfer literature, a more tempting environment does not induce significant changes in temptation spending. However, the magnitudes of both temptation spending levels and the treatment effects are somewhat sensitive to the definition of temptation spending used. This paper examines the potential factors that may be driving these null results, and suggests that future research may find a limited role for temptation in the economic decisions of the poor.
This paper demonstrates the acute sensitivity of education program effectiveness to the choices o... more This paper demonstrates the acute sensitivity of education program effectiveness to the choices of inputs and outcome measures, using a randomized evaluation of a mother-tongue literacy program. The program raises reading scores by 0.64SDs and writing scores by 0.45SDs. A reduced-cost version instead yields statistically-insignificant reading gains and some large negative effects (-0.33SDs) on advanced writing. We combine a conceptual model of education production with detailed classroom observations to examine the mechanisms driving the results; we show they could be driven by the program initially lowering productivity before raising it, and potentially by missing complementary inputs in the reduced-cost version.
Uploads
Papers by Jason Kerwin