Performance Sensitive Debt and CEOs’ Equity Incentives

In our paper, Negative Hedging: Performance Sensitive Debt and CEOs’ Equity Incentives, which was recently accepted for publication in the Journal of Financial and Quantitative Analysis, we examine whether performance sensitive debt (PSD) contracts enable executives to transfer value to themselves at the expense of shareholders. In particular, our paper tests whether the existence and strength of PSD contract terms are related to managers’ incentives from ownership and compensation.

Performance pricing in commercial debt contracts links the borrower’s interest payments to a measure of financial performance, such as its current credit rating or balance sheet ratios. A typical performance sensitive debt (PSD) contract charges lower interest rates in times of good performance and higher interest during poor performance. As a result, PSD contracts redistribute the gains to equity holders across certain future states, increasing returns to equity when the firm performs very well, while reducing them when performance lies in a middle range.

For our empirical tests, we merge a large sample of commercial bank debt contracts with data about the equity ownership of the borrowing firms’ CEOs. For each CEO in our sample, we calculate the delta, or sensitivity of stock and option values to changes in stock price, as well as the vega, or sensitivity of option values to changes in stock volatility. Using a sample of 4,451 loan contracts (1,236 PSD and 3,215 straight debt) negotiated by 1,359U.S. companies from 1994 to 2002, we find that firms whose CEOs exhibit high deltas from their stock and option holdings tend to have flatter performance pricing schedules; one standard deviation increase from the mean in delta corresponds to a 39% decrease in the slope of the performance pricing schedule. Conversely, we find that CEOs with high vegas from option inventories tend to have steeper performance pricing schedules: after controlling for heterogeneity in borrowers’ characteristics and loan characteristics, a one standard deviation increase from the mean of log (1+vega) corresponds to a 17% increase in the performance pricing schedule’s slope.

We examine the relationship between CEOs’ incentives and the PSD slope more closely in two different ways. We look at the “interest-increasing” and “interest-decreasing” segments of the PSD slope, those that lie at credit ratings just below and just above the firm’s rating at the time of contracting. We find a stronger relationship between CEOs’ delta incentives and the interest-increasing slope, implying that CEOs with high ownership are more concerned with avoiding expected costs of financial distress than with reaping the benefits of high rewards for performance improvements. We also examine the convexity, rather than the slope, of the PSD pricing schedule, and we find that both local and overall convexity are positively associated with CEOs’ vega incentives and negatively related to their delta incentives.

The full paper is available for download here.