Abstract: Private-market innovations may be a cost-effective solution to healthcare provision challenges. I exploit the expansion of retail clinics at private pharmacies in Mexico to analyze changes in provider choices and prescription quality. I find that the first entry in the local market leads to a 6-9% decline in public healthcare use. I also find large shifts toward stronger antibiotics in private-market sales. Hence, retail clinics shuffled patients away from existing providers toward doctors with stronger incentives to over-prescribe. These results illustrate how market solutions to congested public services may affect consumers due to agents' incentives that trade-off quality for access.
(joint with Emilio Gutierrez) - resubmitted, World Bank Economic Review
Abstract: We ask whether taxing sugar-sweetened beverages in areas where clean water is unavailable may increase diarrheal disease. Exploiting a tax in Mexico, we find a significant 6.6% increase in gastrointestinal disease rates throughout the first year of the tax, with evidence of a declining impact in the second year. We document avoidance behavior by affected households through differential consumption of bottled beverages two years post-tax. The costs implied by our results are small. However, our findings inform the need for accompanying soda taxes with interventions that guarantee safe drinking water for vulnerable populations.
(joint with Mariano Bosch) - R&R, Journal of Human Resources
Abstract: We study how reducing transaction costs for deposits affects savings, and whether non-branded advertising matters for these effects. Focusing on retirement savings in the privatized pension system in Mexico, we analyze two policies: an expansion of access channels for additional contributions at 7-Eleven stores, and the subsequent media campaign providing information on this expansion and persuasive messages to save. We find that access alone has small and weak effects, but including the campaign leads to a 10-12% increase in savings behavior. However, due to declines in the average contribution size, we find null effects on the total amount saved.
(joint with Emilio Gutierrez) - resubmitted, Management Science
Abstract: Existing literature suggests that hospital occupancy matters for quality of care as measured by various patient outcomes. However, estimating the causal effect of increased hospital busyness on in-hospital mortality remains an elusive task due to statistical power challenges and the difficulty in separating shocks to occupancy from changes in patient composition. Using data from a large public hospital system in Mexico, we estimate the impact of congestion on in-hospital mortality by exploiting the shock in hospitalizations induced by the 2009 H1N1 pandemic, instrumenting hospital admissions due to acute respiratory infections (ARIs) with measures of ARI cases at nearby healthcare facilities as a proxy for the size of the local outbreak. Our instrumental variables estimates show that a one percent increase in ARI admissions in 2009 led to a 0.25% increase in non-ARI in-hospital mortality. We show that these effects are non-linear in the size of the local outbreak, consistent with the existence of tipping points. We further show that effects are concentrated at hospitals with limited infrastructure, suggesting that supply-side policies that improve patient assignment across hospitals and strategically increase hospital capacity could mitigate some of the negative impacts. We discuss managerial implications, suggesting that up to 25-30% of our estimated deaths at small and non-ICU hospitals could have been averted by reallocating patients to reduce congestion.
(joint with Emilio Gutierrez and Jaakko Meriläinen) - resubmitted, PNAS
Abstract: In the aftermath of a large negative shock, such as an epidemic, retrospective voters evaluate the policy-makers' response and either punish or reward them in elections. A prominent concern during the on-going coronavirus pandemic has been whether politicians who are worried about re-election have incentives to impose stringent mitigation measures that may carry high economic costs. On the other hand, voters might also react to the shock itself, no matter what the government does. To understand the effect of epidemic outbreaks on voting, we revisit the 2009 H1N1 outbreak in Mexico. Leveraging detailed administrative data and a difference-in-differences approach, we document a strong, negative relationship between the magnitude of the local epidemic outbreak and the governing party vote share in the 2009 congressional election. Although the average effect is small, the magnitudes may be decisive in close elections. The electoral punishment depended on the magnitude of the peak of the epidemic curve but not on its timing, and it occurred independently of the remedial actions taken by the government. Furthermore, we find persistent effects in the 2012 election. Our results indicate that voters are aware of epidemics, responding to them at the ballot box. The findings indirectly suggest that mitigation policies may be politically valuable even if they are costly for the economy.
(joint with Emilio Gutierrez and Tiago Tavares; Covid Economics Working Paper) - under review
Abstract: Information may be an important policy tool for managing epidemics. Focusing on Covid-19 in Mexico, we randomize an informational treatment in an online survey showing a higher or lower death toll by leveraging delays in death reports, and estimate its effects on individuals' beliefs and behavior. We find that reporting a lower death count -- when delays are not accounted for -- leads to a lower perceived risk of contagion and a lower intention to comply with social distancing. An equilibrium model incorporating the endogenous behavioral response documented by our intervention illustrates the implied differences for the evolution of the epidemic.
(joint with Emilio Gutierrez and Tiago Tavares; Covid Economics Working Paper)
Abstract: Understanding the determinants and implications of delays in reporting COVID-19 deaths is important for managing the epidemic. Contrasting England and Mexico, we document that reporting delays in Mexico are larger on average, exhibit higher geographic heterogeneity, and are more responsive to the total number of occurred deaths in a given location-date. We then estimate simple SIR models for each country to illustrate the implications of not accounting for reporting delays. Our results highlight the fact that low and middle-income countries are likely to face additional challenges during the pandemic due to lower quality of real-time information.
Abstract: Regulations restricting over-the-counter (OTC) access to drugs are important policy tools for mitigating self-medication and encouraging doctor visits. However, they may have unintended consequences when access to affordable healthcare is lacking. This paper estimates the effect of a law restricting OTC sales of antibiotics in Mexico on public hospital admission rates. I find a temporary but significant decline of around 40% in admission rates due to infections. This effect is driven by viral infections, adult patients, and patients from higher education municipalities. OTC regulations may inadvertently increase the socioeconomic health gradient unless accompanied by policies that expand affordable healthcare.
The Effect of a Cash Transfer Program for the Elderly in Mexico City on Co-Residing Children’s School Enrollment. World Bank Economic Review, 2017. (joint with Emilio Gutierrez and Laura Juarez) link
Select work in progress
Gasoline Pipeline Leaks, Health and Crime (joint with Emilio Gutierrez and Fernanda Marquez-Padilla)
Allocation of Government Resources and Service Provision: Evidence from the Quality of Healthcare
Climate Change and Tourism: Evidence from Algae Outbreaks
Utility Disconnection Policies and Health Outcomes (joint with William Violette)