Sovereign Defaults and Debt Sustainability: The Debt Recovery Channel, with Michel Guillard and Hubert Kempf, Submitted

Abstract: This paper focuses on the debt recovery channel linking the dynamics of public debt to partial sovereign defaults. We build a simple model which incorporates sovereign default and a debt recovery rule. It depends on a parameter that allows for partial debt recovery. We show that the maximum debt-to-GDP ratio that a country can sustain without defaulting is increasing, nonlinear, and sensitive to the debt-recovery parameter. A higher debt recovery parameter increases the fiscal space but worsens the financial position of a borrowing country after a default episode. We show the empirical relevance of this channel for estimating country-specific fiscal spaces.

Sovereign Default Risk and Climate Change: Is it Hot Enough ?, with Adham Jaber

Abstract: We estimate the effects of temperature anomalies – temperature’s deviation from its longrun mean – on sovereign default risk and explore the transmission channels. We use crosscountry panel data covering 80 countries over the period 1999-2017. Our results suggest that an increase of temperature leads to an increase of the sovereign credit default swap (CDS) premium. Building on an equilibrium bond pricing equation, we document the existence of a “debt limit channel” of temperature: higher temperature, relatively to the long-run mean, has a negative impact on future growth rate, which lowers the country’s debt limit– the maximum debt-to-GDP ratio it can sustain without defaulting. As a result, the probability of default increases, leading to a higher CDS spread.

Sovereign Defaults in a World of Climatic Disasters: The Expectations Channel

Abstract: This paper analyzes the expectations channel linking the risk of climatic disasters and the prospects of sovereign defaults. I build a stochastic model of sovereign default that allows for time-varying probability of climatic disasters and analyze the role of creditors' expectations about disaster risk. First, I show that the maximum debt-to-GDP ratio that a country can sustain without defaulting is decreasing and nonlinear in the probability of disasters. Second, I compare three types of expectations about disaster risk: the cases of constant, naive, and forward-looking expectations of disaster risk. I show that constant expectations of disaster risk lead to a constant maximum debt ratio. On the other hand, the case with naive expectations of disaster risk – creditors revising the disaster probability in each period while disregarding any future changes – leads to a time dependent maximum debt ratio, but it relatively underestimates default risk compared to the case with forward looking, rational expectations of disaster risk.