UiS Working Papers in Economics and Finance,
University of Stavanger

Abstract: We analyze whether executive compensation reflects firm default risk, measured by distance to default of Merton (1974). Using a large panel of firms, we explore several empirical frameworks. In least squares, fixed effects and quantile regression settings, default risk and volatility possess significant explanatory power for compensation. The signs of the estimated coefficients are consistent with higher compensation for higher risk-bearing--managers of firms with higher default risk are paid more. In the benchmark models, estimates for default risk are quite small, indicating effects on average compensation of around $9,000. The corresponding figure for the upper 5-percentile of CEOs is $50,000. Volatility has much larger average effects, in excess of $25 million, with upper 5-percent effects above $150 million. In a partial identification framework, compensation is systematically associated with default risk. Moving to a higher default risk firm raises compensation by at least $.96%, corresponding to around $70,000 for average CEOs, and above $400,000 for the top 5-percentile. The largest effect applies for moving into the highest default risk firms. For such firms, the treatment effect exceeds $180,000 for average CEOs, and for the top 5-percent, the effect is well above $1 million. Conditional means exhibit monotonicity, suggesting that compensation is amenable to a treatment response approach. Surprisingly, when we utilize an instrumental variables approach, an increase in default risk reduces total compensation. Thus our results uncover a complex but powerful link between firm risk and CEO compensation, which may make it difficult to regulate compensation by means such as `Say-on-Pay'.

Keywords: Default; Executive Compensation; Treatment Response; Partial Identification; Volatility

JEL-codes: A10

48 pages, September 11, 2014

Full text files

uis_wps_2014_11_chollete_eckbo_lu.pdf  

Download statistics

• University of Stavanger
• Home page for this series

Questions (including download problems) about the papers in this series should be directed to Bernt Arne Odegaard ()
Report other problems with accessing this service to Sune Karlsson ().

This page generated on 2024-02-05 17:13:43.