Government Debt and Risk Premia

I document that government debt is related to risk premia in various asset markets: (i) the debt-to-GDP ratio positively predicts excess stock returns with out-of-sample R2 up to 30% at a five-year horizon, outperforming many popular predictors; (ii) the debt-to-GDP ratio is positively correlated with credit risk premia in both corporate bond excess returns and yield spreads; (iii) higher debt-to-GDP ratio is associated with lower real risk-free rates, (iv) higher debt-to-GDP ratio corresponds to lower average expected returns on government debt; (v) debtto-GDP ratio positively comoves with fiscal policy uncertainty. I rationalize these empirical findings in a general equilibrium model featuring recursive preferences, endogenous growth, distortionary taxation, and time-varying fiscal uncertainty. In the model, the tax risk premium is sizable and its time variation is driven by fiscal uncertainty. Furthermore, the model generates an endogenous relationship between the debt-to-GDP ratio and fiscal uncertainty. Fiscal uncertainty increases debt valuation through lower government discount rate. This mechanism is reinforced as higher debt conversely raises uncertainty in future fiscal consolidations. (JEL E62, G12, G17, H63) ∗Yang Liu ([email protected]) is at the Department of Economics, University of Pennsylvania, 3718 Locust Walk, Philadelphia, PA 19104. I am immensely grateful to my advisors Amir Yaron, Enrique Mendoza, Frank Schorfheide, and Ivan Shaliastovich for their invaluable guidance and advice. I also thank Ravi Bansal, Ric Colacito, Hal Cole, Max Croce, João Gomes, Urban Jermann, Dirk Krueger, Nick Roussanov, Tom Sargent, Lukas Schmid, and seminar participants at Penn, Wharton, Philly Fed, and EconCon for helpful discussions. I thank George Hall for government debt data. All errors are my own.