Two Essays on the Influence of Performance-Based Environment, Social, and Governance (ESG) Vesting Provisions in Executive Compensation
This dissertation comprises of two essays that examine the use of performance-based environmental, social, and governance (ESG) options, stocks, and cash grants to chief executive officers (CEOs) in large U.S. firms from 1998 to 2020 and its influence on earnings attributes. Though investors are increasingly considering firms with better ESG performance, there is substantial disagreement on whether firms incorporate ESG measures in their strategic decisions, compensation practices, and corporate reporting, as mere window dressing, or as a legitimate tool to improve corporate social performance and increase firm value (Haque 2017; Gillan, Koch, and Starks 2021). Even with the increasing prominence in performance-based ESG grants, there is limited prior empirical evidence on this topic. This study focusses on filling the gap in compensation literature by classifying non-financial metrics in CEO compensation into environmental, social, and governance measures, estimating the fair value of the grants in a systematic and complete manner, and manually collecting metric weights from annual proxy statements (DEF 14-A fillings). Equipped with such values, I examine the influence of ESG-based grants on earnings attributes (Francis, LaFond, Olsson, and Schipper 2004). In the first essay of my dissertation, "Performance-Based Environmental, Social, and Governance (ESG) in Executive Compensation: A Descriptive Analysis", I document the emergence of performance-contingent (p-c) grants including the increase in use of ESG related metrics in executive compensation. I examine certain characteristics of p-c ESG grants, such as grants that are based on an absolute or relative metrics, have a cliff or ratable vesting schedule, and grants that are paid out as equity or cash. I also document the grant-date fair values (GDFVs) of ESG, accounting, stock-price-based grants and calculate the proportion of these performance-based grants relative to all performance-based executive pay in a year. I examine potential determinants of p-c ESG grants, including firm-specific characteristics, executive-specific characteristics, and corporate governance factors that can influence the frequency and magnitude of ESG-based compensation. I find that the use of p-c ESG grants is positively associated with the post-2005 period, market cap, earnings volatility, stock return volatility, and firm sales. While the use of p-c ESG grants is negatively associated with market-to-book and if CEO is also the chairman of the board. On average, firms that make p-c ESG grants have larger boards, more independent boards, and boards with financial expertise. In my second essay, "CEO Performance-based Environmental, Social, and Governance (ESG) Vesting Provisions and Earnings Attributes", I examine the influence of CEO's equity and cash p-c grants with ESG vesting provisions on seven attributes of earnings: accrual quality, predictability, smoothness, persistence, value relevance, timeliness, and conservatism (Francis et al. 2004). I find that firms that give p-c ESG grants to their CEOs are associated with poorer earnings quality. This finding is robust to controls for innate determinants of earnings attributes (such as cash flow variability, sales variability, operating cycle, incidence of loss, intangible intensity, intangible use, capital intensity, and ROA), CEO characteristics and incentives (such as age, tenure, gender, delta, and vega of a CEO) and governance related measures (such as board size, board independence, board financial expertise). When examining other attributes of earnings, I find some evidence that firms that award p-c ESG grants have less predictable earnings, less smooth earning, less persistent earnings, and less relevant earnings. However, these results seem to be sensitive to model specifications. When examining sub-groups of ESG, I find mixed results: p-c ESG grants with social metrics are mostly associated with poorer earnings attributes such as lower earnings quality and less persistent earnings. Additional analyses show that the association between p-c ESG grants and earnings attributes are primarily attributable to p-c ESG grants with cliff vesting schedule and ESG cash grants. Supplementary analyses suggest that these findings are robust to alternate model specifications using a Heckman two-step approach to control for self-selection issues and fixed effects regressions which control for time-invariant omitted factors. I believe that evidence from this study provides important insights regarding these innovative compensation practices, which may be of interest to regulators and market participants.