Mitigating endogeneity in corporate social responsibility research: An investigation using a neoclassical production function

Abstract

This paper investigates the growing need to control for endogeneity in corporate social responsibility (CSR) research. We modify the theoretical constructs of a neoclassical production function to include CSR and in doing so provide theoretical evidence of endogeneity in the estimation equation. We find that failure to empirically control for endogeneity leads to results that suggest CSR has a negative impact on firm production. However, when controlling for endogeneity through a two‐stage approach, the negative impact disappears. Moreover, we demonstrate the efficacy of using a proven theoretical estimation technique to include CSR and, therefore, mitigate endogeneity concerns.

Board diversity and systematic risk: evidence from emerging markets

“…This method involves identifying an instrument, which is a variable that is correlated with the endogenous regressor but is not correlated with the outcome variable for reasons other than its effect on the endogenous regressor (Angrist and Krueger, 2001). The two necessary criteria for “good instruments” are: (1) influence the endogenous variable, i.e., be relevant, and (2) be exogenous, i.e., not impact the dependent variable except through the endogenous variable (Kruse et al. , 2022).…”