NOMINAL SOVEREIGN DEBT

• Author: Toan Phan
• Published date: 01 November 2017
• Date: 01 November 2017
• DOI: http://doi.org/10.1111/iere.12252

INTERNATIONAL ECONOMIC REVIEW

Vol. 58, No. 4, November 2017

NOMINAL SOVEREIGN DEBT∗

BYTOAN PHAN1

University of North Carolina at Chapel Hill, U.S.A., and Federal Reserve Bank of Richmond,

U.S.A .

I show that reputation alone can sustain nominal sovereign debt, which is subject to both the risks of default

and opportunistic devaluations. Nominal debt combined with a countercyclical exchange rate policy allows

more hedging against shocks than real savings if markets are incomplete. Thus, the loss of either repayment or

monetary reputation severely affects the government’s ability to smooth consumption. The model offers a simple

explanation for the Bulow and Rogoff critique, while simultaneously helping explain the issuance of nominal

sovereign bonds by emerging economies. The model also helps explain why many governments borrow and save

at the same time.

1. INTRODUCTION

What sustains sovereign debt is a classic question in the theory of international macroeco-

nomics. Unlike private debt, sovereign debt is not secured by collateral and the sovereign debtor

government is not subject to bankruptcy procedures or seizure of assets. An influential theory,

formalized by Eaton and Gersovitz (1981) and many subsequent papers, is that governments

want to keep a reputation for repayment to avoid the punishment of exclusion from future

credit and to avoid economic losses from sanctions. However, in an important critique, Bulow

and Rogoff (1989) show that in the absence of sanctions, the threat of credit exclusion cannot

sustain sovereign debt, as long as defaulting governments can still save with a sufficiently rich

set of instruments. Intuitively, when debt obligations are large, the government should default

and invest the resources saved from not repaying creditors in savings instruments, which offer

as much hedging against shocks as debt contracts while giving the government a higher level

of consumption. This critique highlights the difficulty in sustaining unsecured credit in general

equilibrium models, and a large literature has emerged to address this problem.2

However, there is an additional layer to this question: What sustains nominal sovereign

debt, that is, sovereign debt denominated in the debtor country’s currency? Nominal sovereign

debt is subject not only to the risk of default, but also to the risk of opportunistic currency

devaluations. In the absence of international sanctions, if it is already difficult for the loss of

reputation to be a credible threat against default, then what could deter a government from using

surprise devaluations to reduce the real value of nominal debt? In advanced countries with well-

established central bank independence, there exist strong institutional constraints that deter the

governments from devaluing the currency opportunistically. However, in developing countries

with weaker institutions, what could be a deterrent against opportunistic devaluations? This

∗Manuscript received May 2015; revised June 2016.

1I am grateful for the support and suggestions from Martin Eichenbaum, Larry Christiano, Matthias Doepke,

Alessandro Pavan, Eddie Dekel, and Mirko Wiederholt. I also thank Arvind Krishnamurthy, Kiminori Matsuyama,

Sergio Rebelo, and various seminar participants for helpful comments. Last but not least, I thank the editor Harold

Cole and two anonymous referees for useful comments and suggestions. The views expressed herein are those of

the author and not those of the Federal Reserve Bank of Richmond or the Federal Reserve System. Please address

correspondence to: Toan Phan, Department of Economics, University of North Carolina at Chapel Hill, NC 27599, or

Federal Reserve Bank of Richmond, Richmond, VA 23219. E-mail: toanvphan@gmail.com.

2For surveys on the theoretical literature of sovereign debt, see Wright (2011) and Aguiar and Amador (2014). For

surveys of the related empirical literature, see Panizza et al. (2009) and Das et al. (2012). Also see two related papers

by Kehoe and Levine (1993) and Chari and Kehoe (1993).

1303

C

(2017) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association