Richard DiSalvo

I am a PhD candidate at the University of Rochester Department of Economics. I study education, health, and environmental policy using administrative and survey data. I am currently on the job market and will be available for interviews at the upcoming APPAM and ASSA conferences.

I am currently engaged in research on drinking water policy, maternal health, school discipline, grade configurations, high school athletics, and oil and gas development. I have broad policy interests which I support through collaborations with researchers in other fields, and policymakers and practitioners in the field.

I am a member of Professor Elaine Hill's Lab at the University of Rochester School of Medicine and Dentistry. I have presented my work at policy and industry conferences. Feel free to get in touch to discuss research or opportunities to collaborate (email:

My research is presented below. For more detail on my experience, including non-research work, click here for my Curriculum Vitae.

Working Papers

Drinking Water Contamination and Infant Health: Evidence from Pennsylvania. (Contact for working paper.) With Elaine Hill (University of Rochester).
We study the relationship between public drinking water contamination and birth outcomes, by linking births to detailed water sample results for the state of Pennsylvania over the years 2003-2014. Previous research in the US context has found negative health effects of contamination when it triggers regulatory violations. However, US water quality standards are often weaker than public health goals, and an important policy debate is whether to tighten the standards. In this paper, we focus on the effects of water contamination for births not exposed to regulatory violations, thus providing evidence on the possible public health gains from tighter standards. To separate the effects of water contamination from likely confounders, we employ econometric models that make comparisons only within water systems over time, or only across siblings within families. We find that greater overall public drinking water contamination during gestation negatively affects birth outcomes, even when regulatory violations are not triggered. Our most rigorous specifications suggest that moving from the 10th to the 90th percentile of water contamination among births not exposed to regulatory violations increases the chance of low birth weight by about 7.2% and the chance of pre-term birth by 9.8%.
Grade Configurations and Student Test Performance: Evidence from Recent National Data (Available on SSRN)
What grade configuration is best for student performance? Research to date typically suggests that school districts should minimize structural school switching to maximize achievement. This calls into question the common K-5/6-8 school grade structure, which serves the majority of American students in these grades. I use recent national data to bring new evidence to bear on this question. First, I identify the by-grade effects of school termination on test scores using within-district variation in student performance and school terminal grades. Consistent with the literature, I find that test performance falls just after a school terminal grade. However, I also find that students in districts with structural switching consistently outperform their non-switching peers in the grades prior to the switch. This suggests that the policy-relevant ultimate effects of terminal grade use are ambiguous. I use regression adjustment and matching models to attempt to estimate the total effect of terminal grade use on achievement growth and eighth-grade test performance. These models suggest that school transitions reduce both of these performance measures.
The Effects of Suspension Policy on Learning: Evidence from Gender Gaps in Exceptional Districts (Available on SSRN)
I study the effects of suspension policy on student learning in grades three through eight using variation across America's school districts. While the effects of district-level suspension rates on student performance are naturally confounded with student behavior, districts with very atypical rates of in-school or out-of-school suspension may be likely to have adopted exceptional policies, rather than having exceptional students. Following this intuition, I identify exceptional districts using the rate of suspensions for females and ask whether in such apparent policy-adopting districts, males learn relatively faster compared to their female counterparts. Differencing rates of learning across genders is motivated by the fact that males are disciplined at about twice the rate as females in American schools; thus, the direct effects (though possibly not the peer effects) of suspension policy should be larger for males than for females. Among the mid-to-large size districts I study, males learn more slowly on average, but these gaps are smaller for districts that lean toward in-school suspension and districts that use suspension little. The evidence for benefits from policies favoring in-school suspension use is the most robust. I conclude with a discussion of the potential costs involved for districts that heavily use out-of-school suspension to switch to in-school suspension, the policy reform most justified by my findings.

Research in Progress

The Effects of Property Allocations, Generations Later: Railroad Grants, Mineral Rights Severance, and Oil and Gas Leasing in Colorado With Andrew Boslett and Elaine Hill (University of Rochester)
We study how initial property rights allocations influence property ownership patterns centuries later, by relating 19th century railroad land grants to contemporary unconventional oil and gas leasing in northern Colorado. The initial settlement of this region was subject to two alternative policies resulting in widespread privatization of federal land: the Homestead Act of 1862 and the Union Pacific Railroad land grant. Although these land grant allocation rules appear complicated, as a rule, the federal government reserved land for railroad companies in a "checkerboard" pattern. We use this plausibly exogenous pattern as an instrumental variable to estimate the effects of the initial land allocation on modern-day mineral rights severance rates. Since mineral rights severance is not observed for all properties, we use contemporary oil and gas leases to construct two proxy measures for analysis. We find evidence that mineral rights severance rates today are higher on squares originally granted to the Union Pacific Railroad company. In fact, these contrasts are almost entirely driven by mineral rights still held either by Union Pacific itself or a small set of companies to whom Union Pacific has known to have sold mineral rights. These results are robust to alternative specifications and subsets of the data. We conclude that historical resource allocations influence present-day mineral ownership patterns.
Shale Development and Oil and Gas Money in State Legislatures With Zhao Li (Stanford University), Andrew Boslett (University of Rochester), and Elaine Hill (University of Rochester).
Beginning in the mid-2000s, shale oil and gas resources that were previously thought to be inaccessible suddenly became some of the most productive, due to innovations in so-called "unconventional" drilling. This led to a national boom in oil and gas production, and this boom greatly affected a small set of communities that are endowed with this resource. A wave of oil and gas interest entered these communities, and alongside the unconventional wells came political donations. In this paper we study the effects of this innovation on state-level campaign finance in the local communities and states with unconventional development. Our focus on state campaign finance is motivated by the fact that oil and gas policy is primarily determined at the state-level. Using an instrumental variables strategy, we find significant flows of oil and gas money to legislative districts that directly contain unconventional drilling. We also find that legislative districts that do not themselves contain drilling, but which are in a state with drilling elsewhere, also receive oil and gas money, but these effects are about five times smaller than those for districts with drilling. In addition, we use the case study of the New York moratorium to examine the response of campaign finance to policy change. We find that this significant policy is associated with an equally significant departure of New York's oil and gas campaign finance from comparator states.
Patterns in Drinking Water Contamination in Selected US States. With Alexis Zavez (University of Rochester) and Elaine Hill (University of Rochester).
Under the United States' Safe Drinking Water Act (SDWA), public water systems are required to regularly sample drinking water contaminants and report exceedances to both regulatory authorities and local communities. We utilize SDWA drinking water sampling results to better understand water contamination in New York, Ohio, and Pennsylvania. We analyze four key water contaminants: arsenic, nitrate, lead, and coliform. We first calculate trends in water contamination over time and compare current levels of each contaminant to their corresponding maximum contaminant level (MCL). Contaminant MCLs are often higher than maximum contaminant level goals (MCLGs), which represent the level of a contaminant below which there is no known or expected risk to health. Since reducing MCLs towards MCLGs could improve public health, we propose hypothetical reductions for current MCLs and estimate the necessary response required by policymakers to achieve these levels. We determine that both arsenic and nitrate MCLs could be reduced by half or more without requiring a large response from public water systems. However, reductions in the MCLs for lead and coliform could require greater action from system management or may not be feasible if contamination is caused by environmental factors.
Property Value Winners and Losers from the Appalachian Shale Gas Boom. With Andrew Boslett and Elaine Hill (University of Rochester).
We evaluate the heterogeneous effects of the Appalachian shale gas boom on property values in the region. Using a simple difference-in-differences design, it appears that the boom is associated with decreased property values, on average. However, the boom is unlikely to have affected all properties equally. In particular, under plausible assumptions on the elasticity of demand for land, we anticipate properties with more lease potential will experience greater relative price increases. In addition, under weak assumptions, properties with non-severed mineral rights (i.e. mineral rights bundled with surface rights) are also likely to experience greater relative price increases. We use the available data to study these dimensions of effect heterogeneity. Using a triple-differences strategy, we find that properties with greater lease potential experienced more positive price effects from the shale gas boom. In addition, properties with non-severed mineral rights, and properties that are themselves likely to have been leased, also experienced more positive price effects. These results confirm the importance of the shale gas innovation for regional real estate market dynamics, and, by identifying which landowners stand to gain or lose from shale gas development, they also inform research on the political economy of hydrofracking.

Preliminary work or work under extensive revision

Suspension Reduction and Student Performance: Evidence from Apparent Suspension-Limiting Policies. With Josh Kinsler (University of Georgia) and Hao Teng (San Diego State University).
Pass-to-Play, Sports Participation, and Student Engagement: Evidence from Sports Seasons