Behind the Scenes of Regulation: Measuring Informal Business Influence in U.S. Federal Rulemaking

It is a well established fact that firms spend substantial resources to influence regulatory policy in their industries. Largely due to data availability, most empirical evidence in this field has been produced in the context of the United States. Over the last few decades, extensive knowledge has been generated about how companies seek to influence members of both chambers of Congress. However, given ongoing trends, including substantial delegation of regulatory authority from Congress to executive agencies and a decline in congressional capacity, federal rulemaking in the United States has become one of the major venues for regulatory policymaking and, consequently, one of the key foci of regulated industries’ influence efforts. The existing literature on business influence over federal rulemaking in the United States has produced substantial evidence, but also it has important limitations. Empirical research has mostly focused on the public commenting procedure, measuring influence as text similarity between comments and final regulations. However, it is reasonable to expect that companies are much more inclined to channel their influence through less public and less transparent mechanisms, such as direct lobbying contacts before a regulation reaches the public comment stage. In addition, the literature often assumes that regulated industries are inherently anti-regulatory and therefore always prefer less regulation. This is a strong assumption, as it is widely recognized that some companies, depending on their market position and other characteristics, may favor new regulations. This paper addresses both limitations. Firstly, I focus on informal influence of business on the regulatory policy. While data on lobbying and campaign contributions allow us to assess the magnitude of influence attempts, they do not reveal firms’ attitudes in their informal influence efforts. I therefore use firms’ existing public comments to derive their general regulatory stances. Building on prior work, I apply a machine learning model to classify comments as in favor of or opposed to regulation and, based on this classification, categorize companies as pro- or anti-regulation, assuming that companies’ general regulatory orientations, as expressed publicly on non-politically salient issues, are consistent with those employed in informal influence efforts and are sufficiently stable over time. Secondly, rather than measuring changes in regulatory text, I use the Unified Agenda, a standardized reporting system for agencies’ rulemaking timelines, to examine how corporate influence alters or sustains agencies’ regulatory plans. Overall, building on the existing literature, this paper provides original empirical evidence on informal business influence over regulatory policy beyond public commenting, as well as a novel methodological approach that employs machine learning-based classification of public comments to infer firms’ general regulatory orientations.