This paper utilizes a simple two-period model of private syndicated lending to sovereign LDCs where the definition of default is the occurrence of accumulated arrears on principal and interest payments at any time, rather than the standard definition in which a default is identified with the occurrence of a rescheduling of principal payments. The optimal interest spread above LIBOR was estimated using default probabilities obtained from a logistic regression and a Tobit regression. The Tobit scaled probabilities outperformed the unscaled probabilities, indicating that incorporating default as a matter of degree has an important role to play in models of lending to sovereign LDCs. Also the presence of `gaps' in reported interest spreads introduces an important sample selection bias which is corrected using Heckman's method.