Network connectedness dynamics of the yield curve of G7 countries

https://doi.org/10.1016/j.iref.2022.02.052Get rights and content

Highlights

  • We study the connectedness between the yield curve components of the G7 countries.

  • The static analysis shows a higher connectedness in the long-term.

  • The dynamic analysis demonstrates three distinct phases of connectedness over time.

  • Germany and France are the net transmitters of shocks.

  • UK and Japan are the net receivers of shocks.

Abstract

Our study examines the connectedness between the sovereign bond yield curve components (Slope, Curvature and Level) of the Group of Seven (G7) countries. Using the framework of Nelson and Siegel we are able to track their connectedness over the long, medium and short-term horizons. The results of the static analysis show an increased connectedness in the long-term as compared to medium and short-term factors, whereas the results of dynamic analysis demonstrate three distinct phases of connectedness over time. Specifically, these phases are characterized by stable, decreasing and then increasing level of connectedness of the G7 system before, during and after the 2008 crisis, respectively. We also find France as well as Germany function as the net transmitters of shocks whereas UK and Japan function as the net receivers of the shocks for all the components of their yield curves. The findings may shed light on the dynamics and interaction of yield curve shocks between these major economies and may be essential for financial market participants such as investors, fund managers, and policy makers, which debt consists a part of their assets allocation.

Introduction

One of the well-known phenomena in the recent decades is the evolution of the world economy into a single global village, which makes cooperation between these countries inevitably, but particularly challenging. This increased globalization involved creation of formal and informal trade blocs, currency unions and economic alliances. The G7 (or Group of Seven) is one of the most significant informal block of world's seven most powerful, advanced economies (namely, Germany, France, Japan, Canada, Italy, US, and the UK). According to the World Bank, these economies have a pivotal role in the global economy, given that they account for nearly 45% of the world GDP.1 They are also at the forefront of the financial globalization due to advanced technological and industrial capabilities.

Globalization led gradually to increased interlinkages, interdependence and connectedness across the world economies, markets and the asset classes, which may lead to positive as well as negative effects for the economies (Demirgüç-Kunt & Detragiache, 1998; Kaminsky & Reinhart, 1999). The recent COVID-19 pandemic, as well as former market crises have demonstrated the dark side of increased connectedness between financial markets. While the connectedness between G7 countries in this respect was mainly focused on the equity market (e.g., Abbas, Hammoudeh, Shahzad, Wang, & Wei, 2019; Jiang, Tian, & Mo, 2020; Ma, Wang, & He, 2020) or their currencies (e.g., Wan & He, 2021), much less common are works related to their dynamics in the sovereign debt market. Such study is essential, given that interest rates are directly and indirectly related to both the equity and currency markets, but more importantly, two of the major recent crises have been strongly related to the fixed income market (namely, the 2008 sub-prime and the 2011 sovereign debt crises). Moreover, G7 bond markets are the most developed, largest, liquid and important bond markets of the World and have significant influences on the other country bond markets (Ahmad, Mishra, & Daly, 2018; Sowmya, Prasanna, & Bhaduri, 2016; Stona & Caldeira, 2019). From this viewpoint, we aim to complete this gap in the literature and investigate the connectedness in the sovereign bond markets of G7 countries. Specifically, we analyze the static and dynamic connectedness of the government bond yield curves of the G7 countries.

The main question that we address in this paper explores the underlying dynamics of the sovereign yield curve of the most advanced economies in the world over different time-horizons encompassing short, medium and long maturities. We postulate that given the intricate nature of the yield curve, it is important to decompose it into its components to uncover the underlying patterns of connectedness at various time horizons. We use the modification of Diebold and Li (2006) for Nelson and Siegel (1987) model, for the estimation of the yield curve slope, level and curvature. This method is preferred because it provide parsimonious estimates with a 99% accuracy of retrieving the yield curve movement and is also synchronous to economic theory (Diebold & Rudebusch, 2013; Litterman & Scheinkman, 1991; Umar, Riaz, & Zaremba, 2021). In order to analyze the connectedness relationship of the yield curve components of the G7 economies, we make use of Diebold and Yılmaz (2014) connectedness matrix as well as their dynamic connectedness index. This approach permits us explore the bivariate and joint progress of the yield curve components in both static and time-varying setting. In particular, the time-varying analysis allows us to account for the evolution of the yield curve over various stages of the economic cycle, potential outliers and nonlinear dynamics. This is particularly important because our sample period captures important events such as the subprime crisis, the subsequent era of negative real interest rate, the European sovereign debt crisis and the corona pandemic crisis encompassing the first two decades of the 21st century.

Our results can be summarized as follows. The static examination shows the highest level of connectedness in the level (long-term factor) followed by curvature (medium) and slope (short-term factors) of the yield curve. In addition, a higher connectedness is found between the European countries as compared to non-European G7 countries. Through dynamic analysis, we show that Germany and France are the net transmitters while UK is the net receiver within the G7 countries. Furthermore, our examination reveals three distinct trends in the connectedness of system that can be divided into before, during, and after the global financial crisis. We discover increased heights of connectedness before the 2008 financial crisis that touches the lowest during the crisis, and eventually, rebounds (Umar, Riaz, et al., 2020).

The remainder of this work is organized in following manner: Section 2 presents literature review, Section 3 describes the methodology, followed by data and empirical analysis in Section 4. Section 5 concludes the study.

Section snippets

Literature review

We segregate the existing literature in two broad themes. First, theme is related to the literature documenting the yield curve connectedness. The second theme is related to the literature documents the connectedness of the G7 countries. Thus, we identify the gap in literature related to the connectedness of the yield curve of G7 countries.

Importantly, there are different methods to measure the connectedness between system variables, Inter alia, the Granger causality (Billio, Getmansky, Lo, &

Methodology

Our methodology involves two procedures. First, we disentangle the yield curve into its latent factors by applying the Diebold and Li (2006) approach. Thereafter, we use the Diebold and Yılmaz (2014) network connectedness method of to quantify connectedness among the various components of the yield curve. We describe both these approaches in the following subsections.

Data and empirical results

We start this section by a brief discussion for the data employed and the descriptive statistics, followed by analysis of the empirical findings.

Summary, conclusions, and implications

This paper investigates risk transmission and spillover across the sovereign yield curves of the G7 countries. Initially, we employ the Nelson and Siegel (1987) model for the estimation of the level, slope, and curvature of the yield curve, and then introduce these estimated latent factors in Diebold and Yılmaz (2014) model to estimate the connectedness between the three-yield curve latent factors (i.e., long-term, short-term, and medium-term factors). Our data covers the 2008 crisis, the 2011

Declaration of competing interest

None.

Acknowledgements

none.

References (55)

  • N. Giannellis et al.

    Gold price and exchange rates: A panel smooth transition regression model for the G7 countries

    The North American Journal of Economics and Finance

    (2019)
  • M. Hanisch et al.

    The international transmission channels of US supply and demand shocks: Evidence from a non-stationary dynamic factor model for the G7 countries

    The North American Journal of Economics and Finance

    (2017)
  • C. Ioannidis et al.

    The impact of oil price shocks on the term structure of interest rates

    Energy Economics

    (2018)
  • S. Kim et al.

    On the relationship between changes in stock prices and bond yields in the G7 countries: Wavelet analysis

    Journal of International Financial Markets, Institutions and Money

    (2007)
  • K.H. Liow

    Volatility spillover dynamics and relationship across G7 financial markets

    The North American Journal of Economics and Finance

    (2015)
  • G. Magkonis et al.

    Exploring the effects of financial and fiscal vulnerabilities on G7 economies: Evidence from SVAR analysis

    Journal of International Financial Markets, Institutions and Money

    (2014)
  • D.K. Patro et al.

    A simple indicator of systemic risk

    Journal of Financial Stability

    (2013)
  • H.H. Pesaran et al.

    Generalized impulse response analysis in linear multivariate models

    Economics Letters

    (1998)
  • S.J.H. Shahzad et al.

    Asymmetric risk spillovers between oil and agricultural commodities

    Energy Policy

    (2018)
  • S. Sowmya et al.

    Linkages in the term structure of interest rates across sovereign bond markets

    Emerging Markets Review

    (2016)
  • F. Stona et al.

    Do U.S. Factors impact the Brazilian yield curve? Evidence from a dynamic factor model

    The North American Journal of Economics and Finance

    (2019)
  • Z. Umar et al.

    The static and dynamic connectedness of environmental, social, and governance investments: International evidence

    Economic Modelling

    (2020)
  • J. Westerlund et al.

    Panel evidence on the ability of oil returns to predict stock returns in the G7 area

    Energy Economics

    (2019)
  • G. Abbas et al.

    Return and volatility connectedness between stock markets and macroeconomic factors in the G-7 countries

    Journal of Systems Science and Systems Engineering

    (2019)
  • V. Acharya et al.

    Capital shortfall: A new approach to ranking and regulating systemic risks

    American Economic Review

    (2012)
  • V.V. Acharya et al.

    Measuring systemic risk

    The Review of FInancial Studies

    (2017)
  • M. Akhtaruzzaman et al.

    COVID–19 media coverage and ESG leader indices

    Finance Research Letters

    (2021)
  • Cited by (21)

    • The term structure of yield curve and connectedness among ESG investments

      2024, Research in International Business and Finance
    • Are investment grade Sukuks decoupled from the conventional yield curve?

      2024, International Review of Financial Analysis
    • Time-varying bond market integration and the impact of financial crises

      2023, International Review of Financial Analysis
    View all citing articles on Scopus
    View full text