Peer-reviewed Publications


Mtenga, E. L., and Pesko, M. F. (2024). The effect of vertical identication card laws on teenage tobacco and alcohol use. Health Economics, 1-33.

Abstract: We study the impact of vertical identification card laws, which changed the orientation of driver’s licenses and state identification cards from horizontal to vertical for those under 21 years, on teenage tobacco and alcohol use. We study this question using fournational datasets (pooled national and state Youth Risk Behavior Surveillance System, National Youth Tobacco Survey, CurrentPopulation Survey to Tobacco Use Supplements, and Behavioral Risk Factor Surveillance System). We improve previous databases of vertical ID law implementation by using original archival research to identify the exact date of the law change. We estimate models using standard two-way fixed effects and stacked difference-in-differences that avoid bias from dynamic and heterogeneous treatment effects. Using data through 2021, we do not find evidence of reductions in teenage tobacco and alcohol use. While these laws reduce retail-based purchasing, they also increase social sourcing, thus leading to no net impact on use.


Meles, T. H., Lokina, R., Mtenga, E. L., and Tibanywana, J. J. (2023). Stated Preferences with Survey Consequentiality and Outcome Uncertainty: A Split Sample Discrete Choice Experiment. Environmental and Resource Economics, 86(4), 717-754.

Abstract: Stated preference studies are often based on the assumptions that respondents perceive their survey responses as consequential and that proposed outcomes would realize with certainty. This paper conducts a randomized trial of stated preferences experiments to examine the effects of survey consequentiality and outcome uncertainty on responses to a discrete choice experiment study on improving the quality of electricity supply among business enterprises in Tanzania. Our results from the mixed logit model and a model in willingness to pay (WTP) space show significant differences in respondents’ preferences and WTP estimates across the three sample groups (control group, survey consequentiality, and outcome uncertainty). Compared to the control group, including uncertainty in attribute levels of choice tasks increases the overall WTP estimates, while survey consequentiality tends to reduce the estimated WTP. The findings support the call for stated preference studies to explicitly account for outcome uncertainty and survey consequentiality in their design to provide credible estimates in the cost-benefit analysis of the goods and services under consideration.


Working Papers

Estimating Treatment Effects in Presence of Overlapping Programs and Misreporting (with Pierre Nguimkeu) - Job Market Paper [Draft available upon request]

Abstract: Non-mutually exclusive interventions are common features of programs implemented in developing and developed countries. Evaluating the impact of individual and joint participation in these programs may be of policy interest in guiding the cost-effective allocation of resources. However, program participation is substantially misreported in survey data, which may result in biased treatment effect estimates. While the literature has focused on misreporting in one program, our study proposes a method to consistently estimate individual and joint treatment effects of potentially misreported overlapping (and exogenous) programs. We focus on false negative cases which are more prevalent in observational studies. We derive the bias in the traditional ordinary least squares (OLS) estimator and show that it is not possible to determine the direction of the bias a priori. The joint treatment effect may also have an opposite sign to the true effect, which may have dramatic consequences if used to inform policy on whether the programs are complements or substitutes.  As in the previous literature, we argue that any instrumental variable (IV) that meets relevant criteria fails to adhere to the exclusion restrictions, resulting in biased IV estimates. We then develop a consistent estimator of treatment effects using misclassification probabilities, available through validation studies and other external sources. When misclassification probabilities are unknown, we provide an approach to estimate and apply them in the proposed method.  Monte Carlo simulations show that the estimator performs well in finite samples.  Finally, we provide an empirical example, estimating the effect of two correlated and documented to be substantially misreported programs, the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), on food security and healthy eating.


Air pollution and Student Achievement: Evidence from Tanzania (with Tejendara Pratap Singh)

Abstract: Using novel data on students' performance on national exams administered during secondary schooling in Tanzania, we study how air pollution exposure on the day of the exam affects student performance on these exams. To uncover causal effects, we leverage plausibly exogenous changes in local wind direction in an Instrumental Variables (IV) setup. Our IV estimates imply that an increase in PM2.5 concentration by 10 mg/m3 on the day a student appears for the exam worsens their performance on the exam by 0.05 standard deviations. Our results are robust to a host of falsification checks. We also document that the effects are more pronounced for younger students, girls, students appearing for exams in government owned examination centers, students in poorer regions, and those at the lower end of the achievement distribution. Further, we find suggestive evidence that adverse effects of air pollution on exams that test fluid intelligence drive our main results.


The impact of tax treaties on air pollution (with Alberto Chong) - [Draft available upon request]



Work in Progress


Policy and Other Publications