Working Paper: NBER ID: w24435
Authors: Isil Erel; La H. Stern; Chenhao Tan; Michael S. Weisbach
Abstract: Can algorithms assist firms in their decisions on nominating corporate directors? We construct algorithms to make out-of-sample predictions of director performance. Tests of the quality of these predictions show that directors predicted to do poorly indeed do poorly compared to a realistic pool of candidates. Predictably poor performing directors are more likely to be male, have more past and current directorships, fewer qualifications, and larger networks than the directors the algorithm would recommend in their place. Machine learning holds promise for understanding the process by which governance structures are chosen, and has potential to help real-world firms improve their governance.
Keywords: Machine Learning; Corporate Governance; Director Selection; Predictive Algorithms
JEL Codes: G34; M12; M51
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| machine learning algorithms (C45) | accurate prediction of shareholder support (G34) |
| accurate prediction of shareholder support (G34) | actual performance outcomes (L25) |
| machine learning algorithms (C45) | identification of alternative choices of potential directors (G34) |
| machine learning algorithms (C45) | increase diversity in board composition (G34) |
| machine learning algorithms (C45) | predict firm profitability (D22) |
| machine learning algorithms (C45) | predict announcement returns of director appointments (G17) |
| shareholder votes (G34) | reflect perceptions of director quality (G34) |
| algorithms (C60) | identify directors who will enhance firm performance (L25) |
| algorithms outperform traditional OLS models (C52) | do not predict performance accurately (C52) |