The use of accounting numbers in debt and preferred share covenants: Some Canadian evidence

Utilizing a random sample of 152 contracts that relate to public issues of debt and preferred shares made by Canadian firms over the period of 1978 to 1987, this thesis provides some empirical evidence on the use of accounting numbers in the contracting process. More specifically, it addresses three main research questions: (1) Does financial accounting play a contractual role? (2) Is it reasonable to use the debt/equity ratio as a proxy for "closeness" to breach a debt covenant? (3) Can we predict/explain the use of accounting numbers in the contracts under study?;It appears that accounting numbers are used in the contracting process but to a lesser extent than expected. In fact, a sharp decrease in the use of such numbers is observed in the more recently written contracts. The study also provides evidence which indicates that the use of accounting numbers in the contracting process is costly.;The definitions of the accounting numbers/ratios used in the sampled contracts are investigated in order to provide some evidence on the reasonability of using disclosed accounting numbers/ratios to proxy for "closeness" to breach a covenant. A special emphasis is placed on the debt/equity ratio because of the importance of this ratio in Positive Accounting Theory's research. The evidence presented in this thesis shows that no particular accounting number/ratio is extensively used in practice. In addition, the contractual terms used in actual covenants to define these accounting numbers/ratios vary greatly from one contract to another.;Only two variables appear to be significantly related to the use of accounting-based covenants in all the tests performed in the thesis: "time" and "conversion privilege" variables. The "time" variable represents the year in which a new contractual agreement was written. The "conversion privilege" variable indicates whether an issue under study is convertible into a participating issue or not. These two variables explain about one-third of the variance in the use of accounting-based covenants. They can also correctly predict the presence of such covenants in 77% of the cases.;As a whole, the evidence presented in this thesis shows that the use of accounting numbers in the contracting process is a much more complex phenomenon than expected, and, therefore, that a number of simple assumptions that were made in the past about the contractual role of accounting may not be justified. It also raises a number of important research questions that could be addressed in future studies.