How pervasive is the strategic use of corporate political spending to avoid agency enforcement? In a new working paper, I attempt to gain traction on this question with data on nearly two decades of corporate political spending and investigations by the U.S. Securities and Exchange Commission (SEC), which I obtain in part through an original Freedom of Information Act request.

Despite high-profile scandals periodically thrusting these issues into the spotlight, we have limited systematic empirical evidence on the pervasiveness of corporate political spending as a strategic tool to avoid agency enforcement. Our limited understanding of these issues reflects an unfortunate gap in the literature. Corporate political spending has grown significantly in recent decades (Coates 2012; Drutman 2015), a development that may be linked to increased representational inequality (Hacker and Pierson 2010; Schlozman, Verba, and Brady 2012). These trends have prompted calls from the nascent “law and political economy” movement for researchers to better understand how “economic power translates into political power” (Britton-Purdy et al. 2020). Meanwhile, this growth in corporate political spending appears to have followed a similarly large growth in government investigative activity, raising questions about the connections between these developments.

To make progress on these questions, I present a simple theory of agency enforcement with endogenous political costs and bring the theory to data on nearly two decades of corporate political spending and SEC investigations. In my theory, a firm chooses a level of potentially unlawful conduct, which can trigger an investigation. After learning whether it is under investigation, the firm chooses a level of political spending, which will impose costs on the regulator if the regulator chooses to subsequently bring an enforcement action. My theory generates the intuitive hypothesis that politically active firms will respond to investigations by increasing their political spending. I test this hypothesis empirically with a panel design, leveraging changes in corporate political expenditures and investigation status over time.

Before testing for a causal relationship between investigations and corporate political spending, I document the basic descriptive fact that investigated firms are generally much more politically active than other firms. On the extensive margin, about half of SEC investigations involve firms that actively lobby and about a quarter involve firms with active PACs, more than double the rate for the broader universe of firms. On the intensive margin, average lobbying expenditures and PAC contributions are much higher for firms under investigation than other firms, even limiting the comparison to politically active firms. Political spending thus appears to be relatively common and intense for firms that have run into regulatory trouble.

To test whether and how these politically active firms have used political spending to avoid enforcement, I leverage a difference-in-differences design, estimating the impact of investigations on lobbying expenditures and corporate PAC contributions. My results indicate that investigations prompt firms to substantially increase their lobbying expenditures, with average spending rising by about 32% in my preferred specification. These results hold under two different instrumental variables designs, one that leverages the SEC’s tendency to conduct “industry sweeps” and another that exploits regional variation in investigative backlog. I also find some limited evidence that investigations prompt firms to increase their PAC contributions, though these results are weaker.

To test for more nuanced campaign contribution effects, I estimate saturated fixed effects models at the firm–candidate–electoral cycle level that allow me to disaggregate the effects of investigations on the allocation of contributions across different candidates, given the positions of power held by these candidates. The results from this analysis suggest that firms respond to investigations by reallocating their PAC contributions toward more strategically valuable candidates, increasingly favoring incumbents, party and committee leaders, and majority members.

Given the possibility of influence via corporate directors and executives in addition to firm-level spending, I also explore the impact of investigations on the behavior of these individual corporate elites. Subject to the caveat that personnel changes around investigations could introduce bias into my sample of corporate elites, my results suggest that chief executive officers respond to investigations by engaging in strategic behavior similar to firms, increasing the amount of their campaign contributions and increasingly favoring incumbents.

Overall, these results suggest that firms regularly use political spending as a strategic tool to avoid agency enforcement. The results also shed light on the specific channels that firms may use to exert influence.

Although my findings do not conclusively resolve whether such political spending is normatively desirable, the paper discusses some potential reasons for concern and policy interventions to address these concerns. In particular, I focus on two potential reforms: strengthened disclosure requirements for enforcement-related lobbying and a “participatory governance” approach to enforcement oversight.

These findings contribute to multiple literatures across law, political science, economics, accounting, and finance. First, I contribute to a rich interdisciplinary literature on the influence of business in politics by providing both descriptive and causal evidence on the strategies used by firms to influence policy implementation. Given my regulatory setting, this evidence is especially relevant to literature on the politics of financial regulation (e.g., McCarty, Poole, and Rosenthal 2013; Levitin 2014). Second, I add to the literature on the politics of agency enforcement (e.g., Gordon and Hafer 2005), including securities enforcement in particular (e.g., Correia 2014). The securities enforcement literature has largely taken the political environment as exogenous and analyzed the SEC’s responses to this environment. I provide a different perspective by turning the causal arrow in the other direction and focusing on the behavior of firms. Third, I answer a call from the nascent “law and political economy” movement in legal scholarship for researchers to better understand how “economic power translates into political power” (Britton-Purdy et al. 2020). The quantitative approach taken in this paper contrasts with the qualitative approach typically taken by scholars associated with the law and political economy movement.