Debt burden and perceived sovereign default risk: evidence from credit default swaps.

Author:Kriz, Kenneth A.

    Default risk has been one of the main issues studied in the area of public debt management and policy. Commonly asked questions include: How much debt is too much? What economic and financial factors could potentially influence the level of debt? How does debt capacity affect prospective financing of projects? How does the level of public debt reflect the likelihood of default?

    While there are different ways to determine debt level, studies have primarily focused on three approaches. The first method is benchmarking or analyzing comparative debt burdens among jurisdictions (Miranda & Picur, 2000). The second method is regression analysis of debt burdens on various predictor variables (Ramsey, Gritz & Hackbart, 1988). The bulk of the literature in this area relies on these two methods. A third emerging stream of literature uses forecasting and simulation methods to assess the relative risk of default for a given jurisdiction.

    The present research, however, employs a relatively new type of market data--credit default swaps (CDSs)--to study market perceptions of default from an international perspective and in so doing, to provide a new angle to assess public debt capacity. In essence, CDSs reflect market views on the expected loss and probabilities of default faced by bondholders (Chan-Lau, 2003, p. 18). With rapid growth of the credit derivatives market, default swap contracts have become increasingly popular and have gradually come to dominate the market in recent years. They currently account for more than 70 % of the credit derivatives market and are its most liquid instruments (Zhang, 2003; Chan-Lau, 2003). While existing literature focuses on either the valuation of CDSs or the debt capacity, very few works have investigated the response by markets to changes in debt burden.

    The main purpose of this study is to assess the relationship between debt burden and perceived sovereign default risk. By using the implied cumulative probability of default (CPD) calculated from the pricing of sovereign CDSs, we will examine how debt burden and other factors affect market perceptions of risk. In other words, we will assess the role of debt burden on the market-driven CPD while controlling for economic growth, credit ratings, and other important control variables.

    This study contributes to the public debt literature in three significant ways. First, it analyzes factors that determine the variation of CDSs with a focus on the role of external debt. Second, it utilizes a large and underexplored dataset (i.e., data on CPD from several countries from 2009 to 2013) to assess the impact of debt burden on perceived risk. Third, this research suggests a new approach to evaluating debt capacity at the level of sovereign countries. The results from this study help clarify the relative role of debt burden in measuring debt capacity and provide a framework for looking at the concept of market risk.

    The remainder of the paper is organized as follows. Section 2 presents background information on CDSs. Section 3 reviews literature on debt capacity, CDSs, and market perceptions of default. Section 4 describes the data and model specifications. Section 5 presents empirical findings and a discussion of the results. The last section offers conclusions and notes limitations of our study.


    CDSs are "contracts where a buyer makes a payment to a seller in return for a promise that the seller will compensate the buyer if a specified credit event occurs" (SIFMA, n.d.). They are one of the many types of over-the-counter (OTC) derivatives that are traded on the basis of privately negotiated terms between two parties rather than on standardized terms. CDSs are considered the most liquid instruments in the credit derivatives markets (Chan-Lau, 2003, p. 4).

    CDSs were first introduced in the mid-1990s, and their use has become increasingly popular over time. The British Bankers' Association (2002) estimates that, from a total notional amount of $180 billion in 1997, the credit derivatives market grew more than tenfold to $2.0 trillion by the end of 2002. Between 2002 and 2007 the gross notional amount of outstanding CDSs grew from about $2 trillion to almost $60 trillion (ISDA,[I], n.d.). It then fell below $40 trillion in 2009 due to the financial crisis and negative publicity over the perceived role of CDSs in hastening that crisis. Still, as Figure 1 shows, the notional value of the CDS market by the middle of 2013 was just under $25 trillion, or 135 times what it had been 16 years earlier (Bank for International Settlements, 2013).

    Most CDSs are documented using standard forms promulgated by the International Swaps and Derivatives Association (ISDA). CDSs have many variations that are tailored to meet specific needs. In addition to the basic single-name swaps made on the basis of one underlying security, there are basket default swaps (BDSs), index CDSs, funded CDSs, and loan-only credit default swaps (LCDSs) (Weistroffer, n.d.). CDSs are not traded on an exchange and there is little or no required reporting of transactions to a government agency, but there has been some interest in increasing the transparency of the CDS market and in giving regulators greater access to transaction information since the 2008 financial crisis (Kiff & Jennifer Elliott, n.d.).

    In essence, a CDS is a credit derivative contract between two parties. The buyer, often referred to as the "protection buyer," makes periodic (usually quarterly) payments (called the premium or "spread") to the seller (or "protection seller") and receives in return the promise of a payoff if an underlying financial instrument defaults or experiences a similar credit event (ISDA, [P], n.d.; CFA Institute, 2008). CDSs are typically linked to a specified loan or bond of a "reference entity" or "reference obligor," usually a private corporation or government (Federal Reserve Bank of Atlanta, n.d.).

    The reference entity may or may not be a party to the contract. In the former case, a holder of a bond buys "protection" to hedge against its risk of de fault. Thus, a CDS resembles credit insurance. In the latter case, investors can also buy and sell protection without actually owning the debt of the reference entity. These non-debt-backed CDSs, called "naked credit default swaps," allow traders to speculate on the creditworthiness of reference entities. It is like buying insurance against someone else's house burning down. The use of naked CDSs to obtain risk exposure is similar to any other form of investment and trading activity such as stocks, bonds, options or futures. In fact, naked CDSs constitute a large part of the market in CDSs (Kopecki & Harrington, n.d.).

    CDSs resemble a traditional insurance policy to some extent, but the two differ in at least two important ways. First, CDSs are trading products bought and sold by financial professionals every day, much like stocks and bonds, whereas insurance contracts are generally held for their stated terms. Second, CDSs do not require an insurable interest. For example, CDS protection can be bought against the default of a company, even if the buyer doesn't own the company's bonds.

    If the reference entity defaults on its debt, either a physical or a cash settlement takes place. In a physical settlement, the bond holder delivers the defaulted bonds to the intermediary bank that sold the CDSs to protection buyers for payment of the face value. In a cash settlement, the intermediary bank pays the bond holder the difference between the face value and the market price of a specified debt obligation (salvage value). Figure 2 provides a description of the rights and obligations between CDS buyers and sellers. Note that, once a reference entity has sold the debt assets to investors, it will play no role in the exchange of CDSs if it doesn't buy or sell CDSs; similarly, the investor who buys a bond is not a part of a CDS transaction unless it also buys CDS for hedging.

    The "spread" of a CDS is the annual amount (the premium) that the protection buyer must pay the protection seller over the length of the CDS contract, usually quoted as a percentage of the notional amount. For instance, if the CDS spread of an AAA municipal government is 30 basis points, or 0.3%, then an investor buying $1 million worth of AAA municipal bonds must pay the seller $3,000 per year. These payments are usually made on a quarterly basis and continue until either the CDS contract expires or a credit event occurs.

    A CDS associated with a higher spread is considered more likely to default by the market, since a higher fee is charged to protect against such an event. Although factors such as liquidity and estimated salvage value can make the comparison difficult, market participants consider CDS spreads, along with credit ratings of the underlying issuer, to be the best indicators of the likelihood of default or the riskiness of the underlying debt asset (Weistroffer, n.d.).


    In the last decade a large body of empirical work has been conducted on credit-risky bonds, with an emphasis on U.S. corporate bonds (Duffee, 1999; Collins-Dufresne, Goldstein, & Martin, 2001), but relatively little research has been done on the default swap market. Das (1995), Hull, Predescu, and White (2004), and Das and Sundaram (1998) have laid some theoretical ground in this regard. Berndt et al. (2008) studied default risk premiums using default swap data. Houweling and Vorst (2005) implemented a set of simple reduced-form models with regard to market swap quotes and corporate bond quotes in order to investigate the pricing performance of the model. Hull et al. (2004) analyzed the impact of rating announcements on the pricing of swaps. Longstaff et al. (2005) found evidence for a difference between default swap spreads and corporate bond yield spreads, using various risk-free benchmarks. Ericsson, Jacobs, and Oviedo (2009) examined the...

To continue reading