I'm an economist at the Antitrust Division in the Department of Justice. Outside of my day job working on industrial organization topics, my personal research focuses on a combination of environmental economics, law and economics and local public good provision. You can email me at sam.d.krumholz@gmail.com.
Evolution of Elites in the United States
Changes in Concentration and Composition of Elite High Schools: In this paper, I use novel data on two sets of elite high school students, National Presidential Scholars (1964-2024) and Gatorade State Players of the Year for several major high school sports (1985-2024) to examine how the concentration and composition of high schools that have produced elite scholars and athletes have changed over the past half century.
Changes in the Geographic Diversity of the United States Political, Business, Scientific and Cultural Elite Over Time: In this paper, I assemble a unique dataset of the birthplace and high school of members of the United States political, cultural, scientific and business elites over the past fifty years. I test how the concentration and composition of locations in the United States that produce members of the elite change over time as well as how the concentration and composition varies among different types of elites (e.g., Circuit Court Judges relative to Medal of Science winners).
Determinants of Public Good Provision
Kinship Ties, Public Sector Employment and Public Good Provision: In the modern era, civil service protections are meant to protect against nepotism in civil service employment. However, anecdotal evidence suggests that kinship ties may be important determinants of In this paper, I use administrative data from a large number of major municipalities, states and school districts to test for the existence of using the distribution of rare surnames within an agency. I then employ several tests to uncover the mechanisms of the observed associations and test whether a stronger kinship network within an agency is associated with agency performance.
Descriptive Representation and Public Sector Outcomes: Since 2015, major social movements in the United States have pushed for greater descriptive representation of women and certain racial/ethnic groups in leadership positions of public agencies. This paper uses a novel dataset of police chiefs combined with police agency survey data and administrative data on school districts in two states to examine the extent to which descriptive representation of these groups has changed over the past decade and how these changes have impacted both the distribution and level of policing and educational services in affected jurisdictions.
Manufacturing Closures, Lost Property Tax Revenue and Public Service Provision: One important potential impact of manufacturing closures is a reduction of the factory's jurisdictions tax base, which can lead to reduced public goods provision and higher property tax rates. Such an impact may magnify the effect of any closure, creating a positive feedback loop and reducing localities' abilities to bounce back from the closure. However, identifying any tax base impacts separate from the impacts of job losses is challenging because such impacts typically happen simultaneously. In this paper, I attempt to address this challenge by leveraging a 1979 Illinois law that kept local business personal property tax revenues (the source of most manufacturing property tax) constant even if manufacturing plants closed (or if new manufacturing plants opened) and compare tax, revenue and expenditure outcomes to nearby localities in other states in localities that did or did not experience large plant closures.
The Effect of Non-Violent Urban Crime on Voting and Residential Decisions: In this paper, I plan to use the unprecedented wave of thefts of Kia and Hyundai vehicles in the early 2020s to test the impact of crime exposure on urban residents' voting patterns and residential decisions. To control for any endogeneity I use vehicle registration from two states to compare zip codes that had high proportion of Kias and Hyundais to similar zip codes that did not (e.g. zip codes that had high proportions of Toyotas and Nissans). I will then test the impact of the increase in thefts on local voting decisions and migration decisions.
Property Taxation as Compensation for Local Externalities: Evidence from Large Plants (joint with Rebecca Cannon Fraenkel, accepted by Journal of Public Economics)
The Coal Transition and its Implications for Health and Housing Values (joint with Rebecca Cannon Fraenkel and Josh Graff Zivin)
Early Childhood Health and Development in India: A Review of the Evidence and Recommendations for the Future (joint with Achyuta Adhvaryu and Prashant Bharadwaj)
Enforcing Compliance: The Case of Automatic License Suspensions
Non-incarcerative punishments such as fines are widely-used in the United States criminal justice system. This class of punishments involves a compliance choice---an offender can choose to pay the fine or to face the consequences of nonpayment. In this paper, I first show that this compliance choice has two major implications: i) deterrence can be equivalently increased through increases in fines or increases in punishments for noncompliance and ii) the choice of instrument will have large distributional consequences. I then test these theoretical predictions empirically using a natural experiment in Washington state that mandated automatic driver's license suspensions for noncompliance with traffic offenses, a policy already in effect in more than 40 states and leading to millions of license suspensions annually. Using a regression-discontinuity design, I show that the automatic license suspension policies have large effects on punishment compliance, long-run fine repayment and license suspension rates. Using a combination of regression discontinuity and difference-in-difference designs, I further show evidence that this policy decreases traffic violations and traffic accidents among lower-income drivers suggesting that the policy is an effective, but highly regressive way to increase traffic safety.
Enforcing regulations through litigation is an important tool for policymakers, but in many settings we lack strong empirical evidence on litigation's effectiveness and incidence. I study this question using a major environmental enforcement initiative, which led to more than $25 billion in settlement compliance costs across more than 1/3 of US coal-fired power plants. I show that this initiative did indeed lead to large (~20%) decreases in emissions of NOX and sulfur dioxide. These decreased emissions further led to meaningful improvements in local air quality and decreases in local cardiovascular mortality rates. I conclude by examining who bore the costs for these improvements. I find that in regulated electricity markets, the average electricity retail price increased by 10% following a settlement, suggesting that the cost of improvements was largely paid for by utility ratepayers.
The Effect of District Attorneys on Local Criminal Justice Outcomes:
In the United States, elected district attorneys' offices prosecute over 85% of all felony cases, but we know little about their effect on local criminal justice outcomes. Using a newly-collected dataset of district attorney elections, I show that the a Republican district attorney leads to a 6-8% increase in new prison admissions and new sentenced months per capita in the four years following their election. Effects are disproportionately driven by changes in incarceration for drug offenses. These results show that the identity of local district attorneys is an important determinant of incarceration rates.