Essays on Corporate Accountability: Information, Governance, and Incentives

Abstract

Motivated by the growing importance of corporate accountability for environmental and governance outcomes, this dissertation examines three mechanisms through which stakeholders, including citizens, institutional investors, and employees, can discipline or redirect firm behavior. Each essay studies a distinct accountability channel. Together, they clarify how the informational environment and organizational context shape whether external pressure translates into real changes within the firm. In the first essay, I study whether citizen-generated information (CGI), a new form of public data enabled by low-cost sensor technology, can enhance corporate accountability by influencing corporate pollution behavior. Using the staggered rollout of the PurpleAir sensor network and a stacked difference-in-differences design, I find that the installation of a new sensor within five miles of an industrial facility is associated with a 6% to 8% reduction in that facility’s fugitive air emissions. This reduction is driven by increased regulatory scrutiny. Inspection rates rise for both federal (EPA) and state agencies following sensor installation. I further show that these regulators use CGI for different purposes. The EPA uses it as an information substitute to identify previously unmonitored areas, while state agencies use it as a monitoring complement to focus on known high-emitting firms. These findings indicate that decentralized, publicly available information can function as a governance mechanism for corporate externalities, operating primarily through the regulatory channel. In the second essay, I examine how the global adoption of stewardship codes affects investor voting in contested ballot measures, defined as cases in which Institutional Shareholder Services recommends against management. Using a difference-in-differences design across nine adopting jurisdictions from 2013 to 2019, I find that stewardship code adoption increases shareholder votes against management in firms with limited prior U.S. institutional investor presence. This effect is not driven by changes in the voting behavior of existing investors. Instead, it reflects U.S. investors expanding their equity holdings in adopting jurisdictions, an extensive-margin response consistent with the “halo signaling” hypothesis. Under this interpretation, stewardship codes attract foreign investors by signaling a jurisdiction’s alignment with international governance standards. In the third essay, I examine whether diffusing decarbonization-related incentives across more functional units is associated with real emissions reductions, particularly in settings where employee beliefs are misaligned with the firm's sustainability strategy. Using firm disclosures to construct a measure of incentive diffusion, I decompose changes in Scopes 1 and 2 emissions to isolate reductions arising from genuine operational improvements, such as efficiency gains and energy source substitution, from those driven by scale changes or business mix shifts. I find no significant relation between incentive diffusion and total emissions change, but document a consistent negative relation with real emissions changes. This association is strongest in firms operating in environments where employees are likely to hold climate opinions misaligned with the firm's decarbonization strategy, highlighting the value of broad-based incentive design when internal commitment cannot be assumed.