This paper develops a sovereign debt model in which governments are privately informed about their likelihood of default but can themselves have a biased perception of this likelihood. I show that in this setup government borrowing acts as a signal that is only partially informative about fundamental default probabilities, and bond prices do not necessarily reflect true credit risk. I also show that in a two country version of the model correlations between the two countries induces yardstick comparisons, and the borrowing decision of one government affects the bond price received by the other. Whether the information spillover increases or decreases the distortion created by the bias depends on the extent to which borrowing signals reinforce each other.
I thank Andreas Knabe, Samreen Malik, Max Mihm, Joachim Weimann, and three anonymous referees for their helpful comments and suggestions.
- Sovereign debt
- Behavioural bias
- Yardstick comparisons