An Empirical Investigation of the Effect of Accounting Based Debt Covenants on the Magnitude of the Long Window Earnings Response Coefficient

This study empirically investigates the marginal security price responses to unexpected earnings and examines the effect of accounting based debt covenants on the magnitude of the long-window marginal security price responses. This study has two primary research objectives. The first objective is to improve measurement of the determinants of discount factor or demand rate of return (kit). The second objective is to provide additional evidence that return on equity is a positive function of the default risk of debt. The extension and improvement proposed by this study relies heavily on the Earnings Response Coefficient (ERC) literature that describes the price-informativeness of earnings to include appropriate control variables, firm size, reported accounting earnings, level of accruals. This research used three different matching techniques for control groups of non-violating firms in order to provide more powerful tests. This study examined the actual covenant violation, as a proxy for the default risk of debt. Three research hypotheses are generated based on the theoretical propositions of recent accounting based debt covenants, accounting choice, firm size, ERC, and differential information hypothesis literature. The major results of the study are as follows. The results support the notion that ERC is negatively related with the default risk of firm as measured by the actual violation of debt covenants after controlling for the effects of firm size and level of reported net income. However, test results fail to support the notion that ERC is negatively related with the default risk of firm when the firms are matched on the level of their accruals