The Debt-Contracting Value of Accounting Information and Loan Syndicate Structure
Speaker
Abstract
| In this paper, we investigate how the fraction of the loan retained by the lead arranger of a syndicated loan deal is affected by the informativeness of the borrower’s accounting information relative to credit quality, after controlling for the direct use of accounting variables in the formal loan contract. We hypothesize that as the ability of publicly reported accounting numbers to capture changes in credit quality on a timely basis increases (i.e., debt-contracting value of accounting increases), lead arrangers will hold a smaller proportion of the syndicated loan deal. The idea is that when debt-contracting value of accounting increases, information asymmetry between the lead arranger and other syndicate participants is reduced, mitigating ex-ante adverse selection problems and reducing the incentive pressure necessary to induce lead arrangers to exert unobservable monitoring effort ex-post. Further, we hypothesize that accounting information with high debt-contracting value is relatively more important in reducing the proportion of the loan retained when the borrower is not rated, the lead arranger’s reputation is low, and when the lead arranger has not previously served as a lead arranger for the same borrower. We estimate a direct proxy for the debt-contracting value of accounting and document evidence consistent with both of our hypotheses. Finally, we exploit the existence of performance pricing provisions, a unique feature of syndicated loan contracts, to investigate how the debt-contracting value of accounting influences the choice of the performance measure used in these provisions. We predict and find that for loans that include performance pricing provisions, the likelihood that the single performance measure used in the provision will be an accounting ratio rather than a credit rating increases as the debt-contracting value of accounting information increases. |
Contact information: |
| Anna Nöteberg |
