Sovereign defaults, banks and financial institutions

Nicola Gennaioli, Alberto Martin, Stefano Rossi, 17 November 2010

a

A

Recent sovereign defaults highlight a close link between government default and financial sector turmoil, where banks often take centre stage. In the Russian default of 1998 the government's suspension of debt payments triggered large losses on the balance sheets of Russian banks, which had heavily invested in public bonds. These events, further exacerbated by the devaluation of the rouble, allegedly contributed to cause a financial sector meltdown and a credit crunch. Although particularly severe, the Russian episode is by no means exceptional. During the years 1998-2002, the same link between government default, bank bond-holdings and banks' balance sheets appears to have played a key role in Ecuador, Pakistan, Ukraine and Argentina (IMF 2002).

The current debt crisis in Europe also illustrates the link between government default and financial fragility. The downgrading of Greek public bonds in April 2010 raised concerns about the solvency of Greek and other European banks precisely because of their exposure to Greek bonds. Similar concerns have arisen over banks exposed to other European states facing distressed public finances such as Portugal, Spain and most recently Ireland. In this context, market participants view the €750 billion package committed by the EU to avoid public defaults as a way to sustain the continent's banking sector, whose exposure to the bonds of the financially distressed states is estimated to be in the order of €1 trillion.

These events suggest that it may be very hard for governments to discriminate and repay only domestic lenders, or to engineer post-default bailouts of domestic banks in order to prevent defaults from hurting the local economy. There are two implications. First, governments may choose to repay their debts precisely to protect domestic banks, especially when the latter hold many public bonds. Second, if the cost of public defaults arises because they disrupt the domestic financial system, financial development may systematically affect the incentives of governments to repay their debts. Some evidence indeed suggests that public default risk is lower in more financially developed systems (Reinhart et al. 2003, Kraay and Nehru 2006), but the mechanism for why this is the case remains to be understood.

Public borrowing and financial development

In Gennaioli et al. (2010), we study the link between public borrowing and financial development both theoretically and empirically. We build a model where public default is non-discriminatory and where banks demand public bonds as a store of liquidity (Holmström and Tirole 1993). In this model, the government’s default decision involves the following trade-off. On the one hand, a government default beneficially increases domestic wealth because some bonds are held by foreigners. On the other hand, since banks hold some public bonds, a default hurts their balance sheets and hinders intermediation, investment and output. In this setup, financial development increases the government’s cost of default by boosting the leverage of banks. In fact, although higher leverage allows banks to finance a higher level of real investment, it also amplifies the negative impact of an adverse shock to their balance sheets. As a result, the model predicts that government defaults should be more disruptive to credit and output, and thus more costly, in more financially developed markets where banks are more leveraged. These effects are particularly strong under financial openness, for capital inflows allow banks to further enhance their leverage thereby exacerbating their vulnerability to a public default. If financial markets are sufficiently developed, the cost of default is so large that the government can commit to repay its debt.

The evidence

Does the available evidence back up the predictions of our model? To address this question, we build a large panel of emerging and developed countries over the years from 1980 to 2005. We measure financial development by using the “creditor rights” score of La Porta et al. (1998), which is the leading measure of the quality of financial institutions around the world (Djankov et al. 2007). Among other things, our data allows us to control for country fixed effects – that is, for all time invariant differences among countries that may be spuriously associated with financial institutions – as well as for the major domestic and external economic shocks.

We first document that, in line with anecdotal evidence, public defaults are followed by large and systematic drops of aggregate financial activity in the defaulting country. We also find strong and robust evidence supporting the subtler predictions of our model. That is, the post-default credit crunch is stronger in countries that are financially more developed and where banks hold more public debt. We also document that in these same countries the probability of public default is lower.

These effects are economically large. We find that in the year following a sovereign default, private credit falls by 2.4 points as a fraction of GDP and by 8.6% in absolute terms. A one-point increase in the creditor rights score of a defaulting country (e.g. a move from a score of one as in Argentina, to a score of two as in Chile) is associated with a larger contraction of private credit by 5.7% in absolute terms, which amounts to 1.7% of GDP. Finally, every one-standard-deviation increase in the bond holdings of banks in a defaulting country is associated with a more severe contraction of private credit, which falls by an additional 69% of a standard deviation. Similarly, a one-point improvement in creditor rights is associated with a 3.5% reduced likelihood of a sovereign default.

These results indicate that, by boosting the costs of public defaults, the presence of large and developed financial markets greatly enhances the sustainability of public debt. Besides shedding light on the causes and consequences of serial defaults in emerging markets, this principle suggests that the ability of developed-country governments to raise foreign financing and provide liquidity to banks during the recent crisis could be due precisely to the importance of the banking sector in these economies. This perspective adds one critical caveat to traditional accounts of the government as a liquidity provider, namely that the government’s ability to perform this role will be limited by the size of the domestic banking sector. Otherwise, public liquidity provision might just boost public default risk, transforming a merely private debt crisis into a twin crisis involving both the public and the private financial sectors.

References

Djankov, Simeon, Caralee McLiesh, and Andrei Shleifer (2007), “Private Credit in 129 Countries”, Journal of Financial Economics, 84: 299-329.

Gennaioli, Nicola & Martin, Alberto & Rossi, Stefano (2010) "Sovereign Default, Domestic Banks and Financial Institutions," mimeo, Universitat Pompeu Fabra. Earlier version available as CEPR Discussion Paper 7955.

Holmström, Bengt, and Jean Tirole (1993), “Market Liquidity and Performance Monitoring”, Journal of Political Economy, 101:678-709.

IMF (2002), “Sovereign Debt Restructurings and the Domestic Economy Experience in Four Recent Cases”.

La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert W Vishny (1998), “Law and Finance”, Journal of Political Economy, 101:678-709

Topics: Global crisis, International finance
Tags: Debt crisis, default, financial development, Fiscal crisis

Nicola Gennaioli

Researcher, CREI; and Adjunct Professor, Economics Department, Universitat Pompeu Fabra

Alberto Martin

Researcher, CREI; Adjunct Professor, Universitat Pompeu Fabra, and Research Professor, Barcelona Graduate School of Economics

Stefano Rossi

Assistant Professor of Finance, Imperial College Business School and CEPR Research Affiliate