NBER WORKING PAPER SERIES
EMPIRICAL RESEARCH ON SOVEREIGN DEBT AND DEFAULT
Michael Tomz
Mark L. J. Wright
Working Paper 18855
http://www.nber.org/papers/w18855
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
February 2013
This article is forthcoming at the Annual Review of Economics, doi: 10.1146/annurev-economics-061109-080443.
We thank Ben Chabot for helpful conversations, Pol Antras, Gadi Barlevy and Alejandro Justiniano
for comments, and Todd Messer for outstanding research assistance. Tomz thanks the National Science
Foundation for support under grant SES-0548285. Wright thanks the National Science Foundation
for support under grant SES-1059829. The views expressed herein are those of the authors and not
necessarily those of the Federal Reserve Bank of Chicago, the Federal Reserve System, or the National
Bureau of Economic Research.
NBER working papers are circulated for discussion and comment purposes. They have not been peer-
reviewed or been subject to the review by the NBER Board of Directors that accompanies official
NBER publications.
© 2013 by Michael Tomz and Mark L. J. Wright. All rights reserved. Short sections of text, not to
exceed two paragraphs, may be quoted without explicit permission provided that full credit, including
© notice, is given to the source.
Empirical Research on Sovereign Debt and Default
Michael Tomz and Mark L. J. Wright
NBER Working Paper No. 18855
February 2013
JEL No. C82,E01,F21,F34,F51,F55,N20
ABSTRACT
In this essay we review the empirical literature about sovereign debt and default. As we survey the
work of economists, historians, and political scientists, we also emphasize parallel developments by
theorists and recommend steps to improve the correspondence between theory and data.
Michael Tomz
Stanford University
tomz@stanford.edu
Mark L. J. Wright
Federal Reserve Bank of Chicago
230 South LaSalle St.
Chicago, IL 60604
and University of California, Los Angeles
and also NBER
mlwright@econ.ucla.edu
1. Introduction
Sovereign states have borrowed money for hundreds of years. Sovereign debt
was one of the first financial assets ever traded, and it continues to comprise a
significant share of global financial assets. In this essay we review the empirical
literature about external sovereign debt, which arises when sovereigns borrow
from foreign investors.
The significance of external sovereign debt is remarkable considering that
sovereign debt is difficult to enforce. For centuries the legal doctrine of sovereign
immunity limited suit against defaulting sovereigns, and even today few
government assets are available for attachment in foreign jurisdictions.
Moreover, although governments have political incentives to serve the interests
of their own citizens, it is not obvious why they would also respect the wishes of
foreign investors. Why, then, do governments ever honor their debts to foreign
investors, and what gives private bondholders and banks the confidence to lend
to foreign sovereigns?
The long history of external sovereign debt and associated problems of
enforcement have attracted researchers in many fields. In this paper, we survey
empirical work by economists, historians, and political scientists. As we review
the empirical literature, we emphasize parallel developments in the theory of
sovereign debt and recommend steps to improve the correspondence between
theory and data.
2. A Model of External Sovereign Borrowing
We organize our review around a now-standard model of external sovereign debt
and default.
1
In this model, adapted from Eaton and Gersovitz (1981), a
sovereign country is represented by an agent that receives an exogenous
random flow of a single consumption good
each period, where is Markov
and indexes the state of nature. To smooth its consumption the sovereign can
borrow internationally by issuing one period zero coupon bonds. The bonds, with
face value , sell at a discount
,
, a function of the level of borrowing and the
state of nature.
The sovereign enters a period with bonds and, after observing the new value of
, decides whether to repay (R) or default (D). Its recursive value function
,
satisfies
,
max
,
,
.
1
Aguiar and Amador 2013 and Wright 2012a review the literature, following early papers by
Aguiar and Gopinath 2006, Arellano 2008, and Hamann 2004.
If the sovereign repays, it retains access to international capital markets and
chooses new borrowing ′ and consumption to maximize its welfare, subject to
a constraint that its consumption cannot exceed income plus the value of new
borrowing, minus repayment of previous debts. Hence
,
max
,
,
|
,
subject to
,
,
where is a concave period-utility function, is the sovereign’s discount factor,
and
.|
is an expectation operator conditioned on today’s state of nature.
If the sovereign defaults it suffers two costs: it is excluded from financial markets
for a period of time, and it loses a fraction of output. The proportion of lost output,
, proxies for all other costs of default, such as domestic financial distress and
disruptions to trade. At the end of each period, with probability a country in
default regains the ability to borrow and all previous debts are forgiven. Hence
1
0,
1
|
.
Bond prices are determined by competition among risk neutral foreign creditors
who discount the future at rate , such that
,
1
,
1
,
where ′, denotes the probability the sovereign will default in the next period.
This model provides a useful framework for organizing our review of the empirical
literature. In section 3 we focus on the volume and price of sovereign debt, which
are captured in variables and . We discuss how researchers have measured
the amount of debt; quantify sovereign debts in absolute terms and relative to
other assets; and review research about the price of sovereign bonds.
We conclude section 3 by discussing three extensions to the standard model.
First, the model assumes that all bonds last for only one period, but in reality
countries undertake both short term and long term borrowing. We describe how
researchers have measured the maturity of sovereign debt and consider the
causes and consequences of various maturity structures. Second, the standard
model expresses all transactions in terms of a single consumption good, but
countries actually borrow and repay in various currencies. We discuss why
countries often issue debt in foreign currencies and how this affects the likelihood
of default. Finally, the standard model overlooks variation in the legal details of
debt contracts. We discuss trends in the design of debt contracts and consider
the consequences of such provisions.
Section 4 focuses on the sovereign’s decision to repay or default. The model
describes two extreme options, R and D, which represent full compliance or
complete abrogation of the debt contract. The actual behavior of sovereigns is
more complex. Instead of renouncing their debts entirely, some sovereigns make
partial or delayed payments, with or without the consent of creditors.
Consequently, we review how empirical researchers have measured default and
summarize patterns across countries and over time. The standard model also
presumes that defaults end stochastically: in each period there is a probability
that the country emerges from default. Following this theme, we review the
empirical literature about how long defaults last.
We conclude section 4 by examining the costs of default. In the standard model,
defaulters lose access to international capital markets and pay an additional
penalty, . Researchers have studied history to infer how default affects a
sovereign’s ability to borrow, and whether default triggers additional costs such
as trade sanctions or military intervention. We present the state of research on
these important questions.
Finally, the standard model depicts the sovereign as a unitary actor who
maximizes the aggregate consumption of the country as a whole. This
simplification overlooks the role of domestic politics. In Section 5 we discuss
recent empirical research about how voters, special interest groups, and
domestic political institutions affect the decision to repay or default. The essay
concludes by suggesting additional avenues for research.
3. Sovereign Debt
In this section we describe research on the quantity and price of external
sovereign debt. We also discuss three features—currency composition, maturity
structure, and contractual clauses—that are missing from the standard model.
Measuring the Stock of External Sovereign Debt
In the introduction, we defined external sovereign debt as obligations that arise
when governments borrow from foreign investors. In practice, researchers may
not know, from moment to moment, the identity and location of investors who
own the debts of a particular sovereign. For this reason, most empirical
researchers measure external debt as debt issued in a foreign legal jurisdiction
or denominated in a foreign currency. We employ these measures when
presenting descriptive statistics about the quantity of external debt, while
