skip to content

Cambridge Endowment for Research in Finance (CERF)

 

Research Interests: sovereign debt crises, financial contagion, and heterogeneous agent models.

Abstract:

Throughout the decade preceding the COVID-19 pandemic, low-, and middle-income countries' sovereign debt levels have become increasingly large, and increasingly painful to restructure. When crisis strikes, this can severely undermine debt sustainability and impair the sovereign's fiscal capacity to respond to the crisis, as was the case with recent economic shocks hitting some of the world's poorest countries the hardest. To explain why countries end up at such a disadvantageous point of high debt and high economic costs of default, I propose a potentially detrimental mechanism: The Risk-Debt spiral. Here, a shift toward more painful debt portfolios allows governments to sustain elevated debt levels. Using high-frequency data of credit default swaps around court rulings on sovereign debt litigation cases, I can empirically show that as default exogenously becomes more painful, sovereigns' perceived default probabilities drop, allowing for cheaper and higher borrowing. I then develop a debt crisis framework à la Cole and Kehoe (2000), in which a government can issue two bonds that differ in how painful they are to restructure. I find that under certain conditions it is optimal for a country to use the issuance of more painful bonds as a commitment device to eliminate the possibility of a self-fulfilling debt crisis instead of deleveraging. For reasonable parameter choices, this shift towards more painful debt produces increases in sustainable debt levels similar to the observed increase in LMIC's aggregate debt-to-GDP ratios over the past decade. Global frameworks that strengthen the sovereign's bargaining power in debt restructurings, and tighter international regulatory standards for domestic banks' holdings of sovereign debt could help dampen this risk-debt spiral.

Up to Date Project Reports

For Lennart Niermann's webpage, click here.