Work in progress

Abstract

In recent years, research has highlighted the significant health risks associated with living near active oil and gas development sites, with nearly 18 million people in the United States residing within close proximity to these environmental hazards. The pollutants emitted from these sites, such as carbon monoxide, nitrogen dioxide, and fine particulate matter, pose substantial threats to human health, particularly for vulnerable populations like unborn infants and children. This analysis examines the impact of ambient air pollution from oil production on birth outcomes, specifically low birth weight, in California, using comprehensive data on over 242,000 oil and gas wells and county-level birth records from the CDC Natality Files spanning from 1995 to 2021. A central challenge to identification is the potential bias introduced by the simultaneous effects of oil production on local economies and health outcomes. The potential bias arises because oil production can boost local economic conditions, such as employment and healthcare access, while also increasing pollution, making it challenging to separate air pollution's direct negative health impacts from the positive effects of economic growth. To address this, I employ a novel instrumental variable (IV) approach that leverages the interaction between historical county-level oil and gas employment shares from the 1990s and lagged national oil production trends. This method creates plausibly exogenous variation in local oil production, allowing a more accurate estimation of the causal impact on birth outcomes. By isolating this variation from local economic conditions, the IV strategy mitigates endogeneity concerns, providing a clearer understanding of how increased oil production influences low birth weight. The analysis shows that the marginal effect of one million barrels of oil equivalent (MMBOE) increase in oil production above the county's average results in approximately 11.73 additional low-birth-weight births per 1,000 live births, translating to approximately 132 additional low-birth-weight births annually across California. The impact is particularly pronounced among younger and Hispanic mothers, with Hispanic mothers experiencing 2-3 more low-birth-weight births per year for each MMBOE increase in oil production above the county average compared to white mothers. Using back-of-the-envelope calculations, the associated healthcare costs of this increase in low-birth-weight births are significant, with an estimated annual cost of $12.7 million in direct hospitalization expenses. These estimates underscore the significant financial implications of the environmental health risks tied to industrial activities like oil production.

Work in progress

Abstract

Performance-Based Funding (PBF) policies, which allocate state funding to higher education institutions based on specific student outcomes like graduation rates and degree completions, have become a focal point in educational policy. However, the criteria set by these policies can sometimes skew institutional priorities in unexpected ways. For example, if PBF formulas heavily emphasize graduation rates, institutions might concentrate resources on boosting these rates, potentially neglecting other crucial educational objectives such as critical thinking and workforce preparedness. This could lead to an uneven allocation of resources, focusing narrowly on meeting specific targets at the expense of broader educational goals. This paper investigates the mechanisms through which IHEs respond to PBF policies, focusing on how these policies influence institutional spending, enrollment composition, and student outcomes. However, several challenges arise when analyzing how Institutions of Higher Education (IHEs) respond to these financial incentives. Decision-making within IHEs typically rests with central administrations that allocate resources across departments, which can complicate efforts to directly align spending with student metrics tied to state funding. Additionally, the construction of PBF funding formulas, sometimes influenced by internal stakeholders, can shape how IHEs allocate their resources. These formulas may push institutions to focus on specific metrics or maintain the status quo, leading to strategic financial decisions that may not fully align with the policy's original intent. Using a comprehensive analysis of data from all public four-year IHEs between 2007 and 2019, I employ a difference-in-differences (DiD) approach to estimate the effects of PBF policies by leveraging the variation in their adoption across states and over time. The results of my analysis reveal that PBF policies lead to a 1.9% to 3.6% increase in total revenues, driven primarily by increases in state appropriations (5.2% to 8%) and endowment revenues (7.5% to 27.6%). On the expenditure side, IHEs significantly increased their spending on instruction (1.1% to 3.4%) and student services (2.3% to 4.5%), suggesting a strategic reallocation of resources toward areas directly linked to performance metrics. However, there were declines in spending on research and scholarships, reflecting potential trade-offs as institutions prioritize areas tied to state funding. These findings underscore the complex financial dynamics at play as IHEs adapt to PBF policies.

Work in progress

Abstract

This paper examines the impact of Kentucky's bourbon barrel tax on education spending and student achievement. The bourbon industry, a significant contributor to Kentucky's economy, faces a unique state and local ad valorem tax on aging bourbon barrels, with a record $40 million paid in taxes in 2022, of which $26 million was allocated to local school districts. This study utilizes detailed district-level data from the Kentucky Department of Education and the Department of Revenue to investigate how increased bourbon barrel tax revenue influences educational outcomes. Leveraging the plausibly exogenous variation in funding created by this tax, the empirical strategy employs a control function approach to address potential endogeneity concerns and estimate the causal effect of per-pupil spending on student test scores. The findings indicate clear evidence that the additional funding significantly increases instructional support spending. However, the results on student test scores are noisy and inconclusive, suggesting that while targeted fiscal policies can enhance educational quality, their impact on student achievement may be more complex and requires further investigation.

Abstract

Gender peer effects on the educational production function are one of the most important determinants in choosing the most optimal sorting of males and female students into a school, grade, and classroom. The Burkinablé Response to Improve Girls’ Chances to Succeed (BRIGHT) project created high-quality “girl-friendly” schools with the goal of increasing the number of female student enrollment and academic achievements for all students across the 10 lowest performing provinces. The BRIGHT project creates a unique opportunity to measure the impact of an increased proportion of female students on male students’ academic performance. I exploit the discontinuity that is created from the selection process of BRIGHT schools by employing a regression discontinuity design 1 year after implementation of the program. I find variation in math and French achievement for male students who are exposed to female students. The empirical results presented in this paper suggest that a higher female cohort composition has a positive effect on male students’ test scores.

Download