Pay for luck no mercado brasileiro: efeitos da pandemia de Covid-19 e o papel da governança

Abstract

The Covid-19 pandemic affected economic sectors heterogeneously, creating a unique context for investigating executive compensation in Brazil. Grounded in agency theory and managerial power theory, this study analyzes how firms’ idiosyncratic performance and external factors were related to the compensation of Brazilian executives and investigates whether pay for luck occurred. The concept of pay for luck refers to compensation disconnected from the firm’s individual performance, being more pronounced in weak governance environments, Where executives may extract wealth through compensation linked to factors beyond their control. The objective of this research is to analyze the relationship between idiosyncratic performance and external factors, measured by accounting returns and stock returns, and executive compensation during the Covid-19 pandemic, while also investigating how governance mechanisms moderate this relationship. The methodology employs the System Generalized Method of Moments (System GMM), which is appropriate for addressing the presumed endogeneity in the relationship between compensation and performance, in an analysis covering the period from 2018 to 2024. In addition, to ensure robustness, models estimated through Ordinary Least Squares (OLS) and Fixed Effects (FE) methods were also employed to identify potential biases associated with static models. The results indicate that performance measured by stock returns had no consistente relationship with executive compensation. In contrast, accounting returns showed more relevant results: external performance, more specifically industry performance, presented a negative association with executive compensation in one specification, while idiosyncratic accounting performance showed weak evidence of a positive association with compensation. Overall, the findings do not indicate consistent evidence of pay for luck in the Brazilian market during the analyzed period. Furthermore, the most consistent results show that executive compensation in Brazil during the period was associated with firm size and dynamic persistence, that is, compensation in previous years. As a technical product, the dissertation includes, in the appendix, a report on a structured database on executive compensation and governance variables.