On Debt Service and Renegotiation when Debt-Holders are More Strategic

Abstract

The contingent claims analysis of the firm financing often presents a debt renegotiation game with a passive bank which does not use strategically its capability to force liquidation, contrary to what is observed in practice. The first purpose of this paper is to introduce more strategic bank behaviour into the continuous-time model developed by Mella-Barral and Perraudin (1997) and Hackbarth, Hennessy, and Leland (2007). Its second purpose is to account for variations in the information obtained by the parties during the contract period. We show that with public information and private debt only, the optimal probability of debt renegotiation is high when the firm's anticipated liquidation value is high. When we add public debt and asymmetric information, the good-type firm may be tempted to mimic the bad-type to reduce its debt service. We show that to deter such mimicking, banks may sometimes refuse to renegotiate with strong firms having a low liquidation value. We consider private and public debt. At the date of debt emission (when the bank and the firm agree on the private debt level), the bank and the firm do not know the LGD value, so both use the ex ante expected value that set the anticipated ex post renegotiation possibility. After the signature of the debt contract, the firm learns its private LGD while the bank does not observe it, so there is asymmetric information on this parameter at the renegotiation stage. Two scenarios are considered. When we consider an economy where the tax rate is high and a corresponding high debt capacity, the bank has no incentive to fix a renegotiation probability lower than one on any firm type because this would reduce debt value, and thus the firm's debt capacity. However, when the tax rate is sufficiently low, the bank may find it beneficial to use a self-selection mechanism to separate the firm types and increase its debt capacity. We show that a renegotiation probability strictly between zero and one for the bad-type (lower liquidation value) allows the bank to deter the good-type from mimicking the bad-type in order to obtain better renegotiation conditions (sooner and with a lower debt service). Our results are in line with the empirical observation that recovery rate at emergence of bankruptcy is function of the share of private debt in all the firm's debt and is relatively low (Carey and Gordy, 2007).