Managerial ability and stock price crash risk – the role of managerial overconfidence

Abstract

Purpose This paper aims to examine the relation between managerial ability and stock price crash risk, conditional on managerial overconfidence. In addition, conditional on managerial overconfidence, the authors investigate the effect of managerial ability on firms’ choice of bad news hoarding channels, which result in a stock price crash. Design/methodology/approach Using a sample of 24,289 firm-years from companies listed on Compustat and CRSP from 1994 to 2018, the authors conduct panel regression analysis. Findings The authors find that managerial ability is positively associated with stock price crash risk only when managerial overconfidence is high. Furthermore, the authors find that managerial ability seems to exacerbate (attenuate) the bad news withholding by the overconfident managers using the earnings guidance (earnings management) channel. The authors find limited evidence that high-ability managers are likely to withhold bad news through the overinvestment channel and “other channels” when managers are overconfident. Finally, the authors find that the joint effect of managerial overconfidence and managerial ability on firms’ crash risk is more pronounced when there is a material weakness in firms’ internal controls, high investor belief heterogeneity and high information asymmetry. However, this effect appears to dissipate during the recent financial crisis in 2008. Originality/value This research reveals that managerial ability is costly to firms by engendering bad news hoardings and stock price crash risk when managers are overconfident. It also sheds light on how managerial overconfidence and managerial ability affect managers’ choice of bad news withholding channels and stock price crash risk. Finally, the paper is of practical value to the board of directors in selecting the prospective executives.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 25 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.