A SIMPLE METHOD FOR MEASURING SYSTEMIC RISK USING CREDIT DEFAULT SWAP MARKET DATA

Abstract

This paper proposes a simple method that employs credit default swap (CDS) data for analyzing systemic risk. The proposed method overcomes inconsistency problems in existing methods and can produce various indicators of systemic risk in a consistent manner. In addition, this method can measure systemic risk contributions. In particular, the method measures systemic risk contributions in both directions, that is, the overall effect of systemic risk on individual credit risks and vice versa. Using CDS data, we employ the proposed method to measure systemic risk for a group of large financial institutions in the U.S. In addition, we provide empirical results for systemic risk contributions as well as various measures of the overall level of systemic risk and verify the applicability of the proposed method.Keywords: Systemic Risk, Financial Stability, Systemic Risk Contribution, Credit Default SwapJEL classification: C15, E53, G21(ProQuest: ... denotes formulae omitted.)1. INTRODUCTIONSevere financial instability can directly and indirectly entail high costs for the economy (see, for example, Hoggarth et al., 2002). To maintain financial stability, financial regulators use various policy tools not only to prevent individual financial institutions from defaulting but also to control the riskiness of the financial system as a whole (i.e., systemic risk). This system-wide macroprudential perspective has become widely accepted through the 2007-2009 global financial crisis.1Policy efforts to maintain financial stability first require accurate and timely information on systemic risk. However, measuring systemic risk is not a simple task. Employing informative data is indispensable for measuring systemic risk. Historical credit event data are typically provided with a considerable time lag. In addition, defaults by financial institutions are relatively rare events, and thus, it may be difficult to predict such events by using historical data. By contrast, equity return data can convey market participants' expectations of financial institutions in a timely manner. Because of their availability and informativeness, equity return data have been widely used to measure systemic risk.2 By employing structural models, we can infer the default probability from equity prices. However, such models require restrictive assumptions. For example, Merton's (1974) model regards a firm's equity as a call option written on its unobservable asset with some predetermined maturity date and debt (plus interest payments) amount. Because financial institutions are continuously intermediating depositors and borrowers, it may be difficult to determine a precise maturity date, and moreover, the amount of debt at maturity should change as a result of the intermediation.Data on credit default swaps (CDSs) may be used for measuring systemic risk. Under a CDS contract, the protection buyer pays periodic premiums to the protection seller and can be compensated for financial losses from some designated credit event. Such data on CDS premiums may provide high-quality information on credit risk. In particular, the time horizon for credit risk can be considered the CDS contract's maturity date. In addition, market participants determine CDS premiums based mainly on their expectations of well-defined credit events. Indeed, the default probability can be accurately inferred from CDS premiums. CDS products are relative new, but recent years have witnessed the rapid growth of the CDS market. As a result, data on CDS premiums are available for an increasing number of financial institutions.Despite these desirable properties of CDS data, few studies have employed them to measure systemic risk. For example, Huang, Zhou, and Zhu (2009, 2010, 2011) inferred default probabilities from CDS premiums and then combined them with equity return correlations to simulate credit events. They measured systemic risk by the price of insurance against defaults for a hypothetical asset portfolio in the financial system. …