International swap market contagion and volatility
Introduction
The recent global financial crisis (GFC) has highlighted the connectedness and interdependence of international financial markets. One implication of this new world order is that it is critical to determine the extent that interest rates in different countries are correlated and to identify the volatility components that may dominate the transmission process. In this paper we use interest rate swap yield4 and spread data to investigate the link and volatility transmission process between three major international swap markets where there is also a high degree of economic integration: Japan, the UK and the US. A key novelty is that we decompose the volatilities of the swap yield and spreads into long- and short-term components and investigate almost the entire history of the swap market from 1990.5
This study builds upon an existing literature that links market integration, contagion and volatility transmission between financial assets that trade both within and across countries. Initially, we show that an appropriate volatility specification is needed to describe the patterns of market links and volatility transmission. In fact, the use of aggregate volatility shocks makes it difficult to measure the degree of integration and to identify which volatility component is dominant in the transmission process. Consequently, many earlier studies find inconclusive evidence of information transmission between the markets investigated. Second, the previous literature typically examines the correlation, contagion and causality among the markets in isolation. However, in this paper we are able to provide a comprehensive study of the market linkages and volatility transmission between swap markets, although we stop short of drawing conclusions on regulatory policy. This we leave to others such as Georgoutsos and Migiakis (2013).
This approach enables us to make the following contribution: First, this is the first study that examines the linkages and volatility transmission across the three major swap markets by decomposing the aggregate volatility into short- and long-term components.6 Second, we identify the strength (correlations) and direction (causality) of volatility transmission from both contemporaneous and Granger causality perspectives. Third, we examine the contagion effect of crises on the linkages between swap markets. Finally, this study considers different approaches to modelling the volatility links among the three major swap markets (Japan, the UK and the US).
While market integration theory suggests that the swap yield curve should be parallel across different markets, short-term deviations from the no-arbitrage condition may cause non-parallel shifts of the term structure of yield, or spread. Lekkos et al. (2007) argue that these links exist among derivative markets due to common variations in the business cycles across economies and the effect of coordinated arbitrage and hedging activities (see also Lekkos and Milas (2001); Eom et al. (2002)), while the recent work of Jotikasthira et al. (2015) find that world inflation and US yield level explain over two-thirds of the covariance of yields of the US, the UK and Germany at all maturities. However, the existing swap literature does not explicitly disentangle the common variation that may be due to business cycle risks. Dornbusch et al. (2000) classify transmission into three categories: (i) the transmission or spillover that takes place during both ‘good’ times and ‘bad’ times, (ii) the excess co-movement of shocks that have less relevance to fundamental links (financial links, economic/real links and political links) and (iii) an increased correlation, defined as ‘pure contagion’ by Forbes and Rigobon (2002), during the ‘turmoil period’ relative to ‘tranquil period’. Hence, this study addresses three key research questions (RQs):
- RQ1:
Does long-term volatility co-vary across the three major international swap markets investigated?
- RQ2:
Does short-term volatility co-vary across the three major international swap markets?
- RQ3:
Is there any contagion effect in the above three major international swap markets?
The first research question asks whether the long-term volatilities across the swap markets are correlated and, if they are, are the correlation times varying? This is linked to the Dornbusch et al. (2000)'s classification of the fundamental link and correlations of business cycle risk. Note that Litzenberger (1992) and Lang et al. (1998) show that default risk of swap counterparties co-varies with the business cycle risk. The second research question is related to Dornbusch et al. (2000)'s classification of excess co-movement, also known as the correlation of the noise component, or skewness risk, across financial markets. Adrian and Rosenberg (2008) define market skewness risk as a measure of tightness of financial constraints. The credit spread differentials across the domestic and international markets can be regarded as skewness risk in swaps (Nishioka and Baba, 2004). With respect to the third research question, Eom et al. (2002) argue that the trading behaviour associated with crisis events is likely to accentuate the integration of swap markets. Hence, it is important to evaluate the influence of key events – such as crisis – on swap market integration. The specific events investigated in this study are explained in the methodology section, but include key periods of financial crisis arising during the recent Lehman default and the Asian financial crisis of 1998.
To investigate these questions, the empirical analysis is based on swaps with a 5-year maturity trading in Japan, the UK and the USA, which are three of the most liquid markets. For brevity we limit the analysis to these maturities and markets. However, this approach can be duplicated by others for other markets and maturities. The analysis is conducted as follows: First, the study decomposes the aggregate volatility shocks into short- and long-term components using the Factor-Spline-GARCH (hereafter FSG-Spline GARCH) of Rangel and Engle (2012). Second, the strength of integration is measured through the short- and long-term volatility correlations using Engle's (2002) Dynamic Conditional Correlation (DCC). These correlations are then utilized to examine the contagion effect. Finally, we model contemporaneous and Granger causality of the volatility components to determine the direction of transmission.
Our findings are as follows. Relating to the first and second research questions, the time-varying correlations between Japan and the UK and between Japan and the US are very low for both swap rates and spreads. These results imply that the level of swap market integration between these countries is statistically weak, and suggests that when swap rates and spreads change, the underlying yield curve shifts in a non-parallel manner. If these circumstances persist international investors could take advantage of this opportunity by going long (or buying) Japanese yen interest rate swaps and going short (or selling) US dollar, or UK pound, swaps to take advantage of the differential between the lower long-term yields of Japanese government bonds and the higher long-term yields of US, or UK, bonds. The low correlations between Japan and the US, or the UK, may also cause an increase in the yen swap rate. Our findings of weak integration between Japan and the UK, and between Japan and the US, suggest that the market linkages detected by prior studies are mainly due to volatility misspecification. Finally, our analysis demonstrates that most of the crisis events influenced the correlations of long- and short-term volatilities across the markets investigated. However, the contagion effect was more evident on the swap spreads than on the underlying yield curve. That is, credit risk components appear to be more affected than swap market risk.
The remainder of this paper is organized as follows. The importance of studying swap market linkages and volatility transmission is discussed briefly in Section 2. To motivate our hypotheses, in Section 3, we review the prior literature on volatility spillovers in swap markets and in two-factor (short- and long-term) volatility models. Section 4 describes the data while Section 5 explains the estimation techniques. Section 6 reports the empirical findings and analysis thereof, while Section 7 concludes.
Section snippets
Importance of studying swap market linkages
An interest rate swap is a highly liquid over-the-counter (OTC) derivative instrument comprising two legs, one paying fixed rate and the other paying a floating rate, typically the London interbank offered rate (LIBOR). This study focuses on the plain vanilla swap, i.e., fixed-for-floating swap rate. The swap spread causes the swap yield (fixed-for-floating rate) curve to be above the Treasury (government bond) yield curve. In theory, this spread, at any given maturity, reflects the additional
Prior literature, market links in two-factor volatility models and hypotheses
The finance literature typically uses the ‘volatility transmission hypothesis’ to measure the extent of financial integration, information spillover and transmission between markets. This approach interprets volatility as a “global factor” [Eom et al. (2002), p. 6]. In this context, news originating from one market may simultaneously affect other markets. Engle et al. (2012) provide a summary of the empirical literature on volatility transmission in different financial markets. They argue that
Data
We examine the linkages between interest rate swap markets via a two-factor volatility model and use daily 5-year data from the three major swap markets, namely Japan, the UK and the USA for the period from September 1989 to January 2010. Daily data is used to accommodate the fact that recent technology has enhanced the transmission of shocks from one to the other markets. Our analysis is confined to these three markets as they comprise a large share of the total swap markets. The euro is
Modelling short- and long-term volatility components and dynamic correlations
To test the hypotheses (H1, H2) developed in Section 2, it is necessary to measure the correlation dynamics. This task is typically done using Engle's (2002) Dynamic Conditional Correlation (DCC) model. The DCC has two stages. In the first stage, a univariate GARCH model is estimated. In our case we use the Factor-Spline-GARCH (FS-GARCH, hereafter) model of Rangel and Engle (2012). To conserve space, the FS-GARCH approach is detailed in Appendix B. The methodology is applied to decompose
Preliminary analysis
This sub-section presents the first set of empirical results. For comparison of the volatility patterns, we begin our analysis by considering the decomposed volatilities obtained from the FS-GARCH model of Rangel and Engle (2012). The estimation results for the three markets are reported in Table 2. For each country, we use the daily swap rate and spread changes data for the relevant 5-year swap. In Table 2, α and β indicate the ARCH and GARCH effects, while c indicates the asymmetric effect in
Concluding remarks
This paper examines the market links and volatility transmission across three major international swap markets (Japan, UK and USA) for the period from 1989 to 2010. Unlike the previous literature, this study uses a step by step procedure for measuring the linkages present in these markets. We have three main results: First, when examining the correlation of long-term volatility (proxied by business cycle risk) and short-term volatility (proxied by skewness risk), we find that long-term
Acknowledgements
For valuable comments and suggestions on earlier versions of this paper, the authors would like to thank Tarun Chordia and the seminar/conference participants at the Asia Pacific Derivatives Association (APAD) Conference (Busan), Midwest Finance Association Conference (Chicago), Australasian Banking and Finance Conference (Sydney) and the Accounting and Finance PhD Symposium (Prato, Italy). The usual disclaimer applies.
A.S.M. Sohel Azad is a lecturer in finance and financial planning at Deakin University, Australia. His research interests include: volatility and risk modelling in financial markets, macroeconomic risk and financial markets, and Islamic finance.
References (81)
- et al.
Financial linkages between US sector credit default swaps markets
J. Int. Financ. Mark. Inst. Money
(2014) - et al.
The multiscale causal dynamics of foreign exchange markets
J. Int. Money Financ.
(2013) - et al.
Estimating stochastic volatility diffusion using conditional moments of integrated volatility
J. Econ.
(2002) - et al.
Modeling contagion in the Eurozone crisis via dynamical systems
J. Bank. Financ.
(2015) - et al.
Alternative models for stock price dynamics
J. Econ.
(2003) - et al.
Option valuation with long-run and short-run volatility components
J. Financ. Econ.
(2008) Emerging market sovereign bond spreads and shifts in global market sentiment
Emerg. Mark. Rev.
(2014)- et al.
Information and volatility linkages in the stock, bond, and money markets
J. Financ. Econ.
(1998) - et al.
Volatility linkage among currency futures markets during us trading and non-trading periods
J. Multinatl. Financ. Manag.
(1999) The effect of the interbank network structure on contagion and common shocks
J. Bank. Financ.
(2013)
Heterogeneity of the determinants of euro-area sovereign bond spreads; what does it tell us about financial stability?
J. Bank. Financ.
The evolution of risk premium as a measure for intra-regional equity market integration
Int. Rev. Financ. Anal.
An empirical examination of the convexity bias in the pricing of interest rate swaps
J. Financ. Econ.
Timeline of a financial crisis: introduction to the special issue
J. Bus. Res.
Market risk and the concept of fundamental volatility: measuring volatility across asset and derivative markets and testing for the impact of derivatives markets on financial markets
J. Bank. Financ.
Volatility spillovers across international swap markets: the US, Japan, and the UK
J. Int. Money Financ.
Why do term structures in different currencies co-move?
J. Financ. Econ.
On financial contagion and implied market volatility
Int. Rev. Financ. Anal.
Contagion of the Global Financial Crisis and the real economy: a regional analysis
Econ. Model.
Dynamics of bond market integration between established and accession european union countries
J. Int. Financ. Mark. Inst. Money
Determinants of interest rate swap spreads
J. Bank. Financ.
The use of financial derivatives and risks of U.S. bank holding companies
Int. Rev. Financ. Anal.
Exchange rate contagion in Latin America
Res. Int. Bus. Financ.
Statistical inference in vector autoregressions with possibly integrated processes
J. Econ.
One crisis, two crises…the subprime crisis and the European sovereign debt problems
Econ. Model.
Bayesian dynamic linear modeling for exploring the impact of recent financial crisis on Japan Credit Default Swap market
Expert Syst. Appl.
Identifying risks in emerging market sovereign and corporate bond spread
Emerg. Mark. Rev.
Statistical Release. OTC Derivatives Statistics at End-June 2014
Volatility transmission across the term structure of swap markets: international evidence
Appl. Financ. Econ.
Stock returns and volatility: pricing the short-run and long-run components of market risk
J. Financ.
Range-based estimation of stochastic volatility models
J. Financ.
An overview of the crisis: causes, consequences, and solutions
Int. Rev. Finan.
What drives interest rate swap spreads
Low-frequency volatility of yen interest rate swap market in relation to macroeconomic risk
Int. Rev. Finan.
Time-varying world market integration
J. Financ.
Market integration and contagion
J. Bus.
An economic analysis of interest rate swaps
J. Financ.
An empirical analysis of the dynamic relation between investment-grade bonds and credit default swaps
J. Financ.
Modelling the coherence in short-run nominal exchange rates: a multivariate generalized arch model
Rev. Econ. Stat.
Introductory Econometrics for Finance
Cited by (8)
Day-of-the-week effects in financial contagion
2019, Finance Research LettersVolatility spillovers between foreign exchange and stock markets in industrialized countries
2018, Quarterly Review of Economics and FinanceCitation Excerpt :The long-run (permanent) component provides a measure of volatility generated by fundamental factors [see, for example, Blake and McMillan (2004) and Byrne and Davis (2005)], while the short-run (transitory) component represents mostly transitory volatility conditioned by financial market considerations, such as the arrival of new information, speculation and hedging positions. 6 We consider the variance causality among the estimated volatility components in a structural Vector Auto-Regression (SVAR) framework (Azad, Batten, Fang, & Wickramanayake, 2015). 7 Following Bollerslev (1990) under this multivariate regression framework, the models can be thought of as an extension of Seemingly Unrelated Regression (SUR) and thus, the models are estimated in a SUR framework.
Causes and hazards of the euro area sovereign debt crisis: Pure and fundamentals-based contagion
2016, Economic ModellingLow-frequency volatility and macroeconomic dynamics: Conventional versus islamic stock markets
2022, Singapore Economic ReviewFinancial contagion: review of empirical literature
2018, Qualitative Research in Financial Markets
A.S.M. Sohel Azad is a lecturer in finance and financial planning at Deakin University, Australia. His research interests include: volatility and risk modelling in financial markets, macroeconomic risk and financial markets, and Islamic finance.
Jonathan A. Batten is a Professor of Banking and Finance at Monash University, Australia. Prior to this position he worked at the Hong Kong University of Science & Technology. He is the editor of Emerging Markets Review and associate editor of the Journal of Banking & Finance, Journal of the Asia Pacific Economy, Research in International Business and Finance, and International Review of Financial Analysis.
Victor Fang is an Associate Professor of Finance at Deakin University, Australia. His research interests include: fixed income and interest rate swaps and related issues, financial risk management, modelling the term structure of interest rates, and volatility modelling.
Jayasinghe Wickramanayake is a senior lecturer in the Department of Accounting and Finance. He is a Fellow of the Financial Services Institute of Australasia. His research interests are in the fields of superannuation and pension finance, funds management, central banking, mergers and acquisitions and small and medium industry finance.