Research

Job Market Paper

“A Novel Measure of the Policy Content of Federal Regulations in Ideological Space” (with Joseph Essig)

(Job Market Paper) Submitted

The lack of a systematic measure of the ideological content of regulations has greatly limited research on the federal bureaucracy. Using the comments of Federal legislators on agencies’ proposed rules, we estimate the ideological location of 257 important and politically contested rules in DW-NOMINATE space, spanning 51 agencies over 20 years. This measure of rule content, included in a newly compiled dataset of comment text and hand-coded rule-level variables, reveals significant variation in the content of Federal regulations beyond what is captured by prior agency-level measures. Applying our measure to theoretical models of rulemaking, we find that the preferences of agencies, Congress, and especially the President, impact the ideological location of federal regulatory policy. Broadly, our measure enables more precise empirical study of the Federal bureaucracy than has been possible before, with implications for scholarship in several key areas like separation of powers, bureaucratic procedure, distributive politics, and agency discretion.

Other Research

“Procedural Politics in Rulemaking”

In Progress

Rulemaking by administrative agencies has become a crucial mechanism in the U.S. governance system, facilitating regulatory action beyond Congressional capacities. This paper investigates Congressional oversight within this process through the lens of procedural comments made by legislators on proposed rules. By expanding upon the initial work of Lowande and Potter (2021), this study employs a larger and more accurate dataset to assess the relationship between ideological disagreement and the propensity to engage in procedural commenting. Incorporating both member-fixed and rule-fixed effects, our analysis reveals a significant link between ideological opposition and the frequency of procedural comments. Further, a Regression Discontinuity Design (RDD) applied to close elections supports the hypothesis that Republican victories increase comments on liberal rules and decrease on conservative ones. This investigation enhances our understanding of the strategic use of procedural comments as a form of Congressional oversight, offering insights into how ideological battles are waged within the rulemaking arena.

“Rulemaking Under Trump: An Empirical Analysis”

In Progress

By many accounts, rulemaking during the Trump administration was highly unusual. Trump’s impetuous, adversarial, and incompetent leadership led to mass confusion within agencies and ineffective regulatory and deregulatory actions. Bullheaded insistence on extreme and foolish deregulation spawned an unparalleled level of resistance within agencies, protecting society from some of Trump’s worst notions. While there is consensus among many political observers regarding this narrative, researchers have barely begun to take stock of how rulemaking changed during the Trump administration. In this study, using a novel dataset on rules between 1995 to 2021, I analyze rulemaking under Trump and compare rulemaking under Trump to rulemaking in previous administrations. I find that many rulemaking patterns were not much different in the Trump administration, but there are important exceptions.

“Strategic Timing of Rulemaking During Presidential Transitions”

In Progress

Rulemaking is an important process through which agencies gain significant power in policymaking. However, agencies cannot wield this power free from the constraints of other political actors. I argue that agencies strategically time rulemaking to avoid presidential oversight. Agencies that believe friendlier overseers will soon be in power have an incentive to delay rulemaking. Conversely, agencies that believe more hostile overseers will soon be in power have an incentive to finalize rulemaking while the friendlier overseers are still in office. To provide evidence of the strategic timing of rulemaking, I investigate the timing of final rule submissions during midnight periods. Midnight periods offer the best opportunity to evidence strategic timing because agencies can anticipate future opposition since overseers are elected but not yet sworn in. Reviewing all proposed rules between 1994-2014, I find strong evidence that agencies time final rule submissions in accordance with their motivation to avoid hostile oversight.

“A Machine Learning Approach to Classifying Rules as Deregulatory”

In Progress

This study introduces a machine learning framework to classify federal regulations as either deregulatory or regulatory, addressing a critical gap in the empirical literature on rulemaking. By leveraging Executive Order 13771, which mandated agencies under the Trump administration to categorize rules accordingly, we develop and train a supervised learning model on this labeled dataset. The model is then applied to a broader corpus of rules without prior classifications, significantly expanding the scope of analysis. This novel classification enables researchers to investigate regulatory patterns across administrations, shedding light on how political and institutional actors influence the regulatory process. The findings contribute to the understanding of the dynamics between executive oversight and administrative agencies in shaping regulatory policy.