Fresh evidence on the oil-stock interactions under heterogeneous market conditions

https://doi.org/10.1016/j.frl.2023.103726Get rights and content

Highlights

  • We propose a TVAR-AGARCH-M model to examine oil-stock interactions under heterogeneous market conditions.

  • We show that the TVAR-AGARCH-M model is superior to the benchmark VAR-GARCH-M model.

  • Oil price volatility affects stock returns in high volatile states but not in low volatile states.

  • We find evidence of strong asymmetry wherein lagged negative shocks strongly influence volatility in oil and stock markets.

Abstract

This paper ameliorates the existing empirical literature on the oil-stock nexus in three ways. First, we expand the literature on return-volatility interactions by adding non-linear dimensions to it. Second, we propose a threshold VAR asymmetric GARCH-in-mean (TVAR-AGARCH-M) model to examine the mean-spillover and return-volatility associations under heterogeneous market conditions of bull and bear markets. Third, we show that the TVAR-AGARCH-M model is superior to the benchmark VAR-GARCH-M model. Our analysis unfolds that oil price volatility affects stock returns in highly volatile market conditions but not in low-volatile states. We also find evidence of strong asymmetry wherein lagged negative shocks strongly influence volatility in oil and stock markets. Specifically, we show that countries such as Japan can benefit from hedging in international markets by diversifying their portfolio in positively correlated markets.

Introduction

How does oil price fluctuations affect stock returns? One would assume that the question is settled long ago. But, in fact, it is not. While the economic literature provides competing theories regarding the sign of the impact (Fisher, 1930, Williams, 1938, Jones and Kaul, 1996, Kollias et al., 2013), the empirical observations post global financial crisis (GFC) 2008 have added several important dimensions to the oil-stock relationships. Either with standard models or modern incarnations, the findings suggest that stock returns respond to oil price fluctuations based on whether the oil shock is emanating from demand-side factors or supply-side factors (Kilian, 2009, Kilian and Park, 2009); whether the country is an oil importing or oil-exporting (Bjørnland, 2009, Bouri, 2015, Filis et al., 2011, Ramos and Veiga, 2013, Güntner, 2014, Silvapulle et al., 2017); types of firms and their sizes (Narayan and Sharma, 2011, Narayan and Narayan, 2014, Narayan and Sharma, 2014, Phan et al., 2015). Since the GFC, a growing body suggests that the effects of oil price on stock return and its volatility are not linear, rather depend on the existing market conditions (Aloui and Jammazi, 2009, Chen, 2010, Choi and Hammoudeh, 2010, Reboredo, 2010, Chang and Yu, 2013, Caporale et al., 2015, Zhu et al., 2016).

Given this background, our study is motivated by the recommendations made by Smyth and Narayan (2018) that there is a pressing need to combine diverse strands of oil-stock literature. Our first contribution is to fill this gap by marrying the literature on return-volatility interactions to the literature on nonlinearity in the relationships. Second, we contribute by examining the joint behavior of oil-stock markets by adopting a simultaneous modeling framework that is more appropriate than working with separate univariate models as the former allows us to capture ‘mean spillover’ and ‘mean-volatility’ interactions together and thus several hypotheses can be tested directly. However, our main contribution lies in improvising the benchmark model by adding nonlinear dimension to it.1 Third, we contribute by hypothesizing how return spillover and mean-volatility interactions vary under different heterogeneous market conditions, i.e. bear and bull markets.2 Further, stock returns are a financial variable and thus exhibit high and low volatility states, often due to the occurrence of exogenous events such as economic and financial crisis, war, pandemic, etc. Considering our sample span includes a few such events, therefore a regime switching model is best suited to identify and examine these switching dynamics occurred in the oil-stock relationships. The analysis will not only assist forecasters about oil-stock market interlinkages under heterogeneous market conditions but also benefit investors in designing appropriate hedging strategies and managing portfolios in face of any market failure. Our final contribution is implementing a threshold VAR asymmetric GARCH in mean (TVAR-AGARCH-M) model to verify the non-linear linkages, which has not yet been implemented in the literature and is superior to the benchmark VAR-GARCH-M models used recently in the literature. To the best of our knowledge, ours is the first to propose a nonlinear threshold model in VAR framework to study the asymmetric relationships between stock and oil markets.3

Our empirical approach is as follows. First, our paper verifies a set of hypotheses in heterogeneous market conditions. Specifically, we intend to examine how each of the three relationships viz.,(a) impact of oil price on stock return,(b) effect of oil price volatility on stock return, and(c) impact of stock return volatility on stock return, vary with ‘bull’ and ‘bear’ market conditions? Further, we shed some light on the volatility-spillover effect and how oil-stock market volatility is affected by its own past shocks? Second, we test these hypothesis using weekly time series data on three large net oil-importing countries of Asia, namely, India, Japan and Korea.4 Third, we rely on a two-stage method wherein a univariate Markov-switching heteroscedasticity (MSH) model on stock return has been adopted in the first stage to identify the high and low volatility (‘bear’ and ‘bull’, respectively) periods. Then, in the second stage, a TVAR-AGARCH-M model is implemented to identify the non-linear linkages. We confirm consistency of our results with robustness checks.

Our analysis unfolds that oil price volatility affects stock returns in highly volatile market conditions but not in low volatile market states. Further, we find evidence of strong asymmetry wherein lagged negative shocks strongly influences volatility in oil and stock markets. Specifically, we show that investors can hedge risky assets in one market with holding uncorrelated and less risky assets in their international portfolio, as in Batten et al. (2021). Finally, we show the appropriateness of TVAR-AGARCH-M model over the benchmark VAR-GARCH-M model, indicating robustness of our empirical findings.

The rest of the paper proceeds as follows. Methodology and results are discussed in Sections 2 Methodology, 3 Data and results, respectively. Section 4 gives concluding remarks.

Section snippets

Methodology

We first report the MSH model that distinguishes ‘bull’ and ‘bear’ market situations. Thereafter, we briefly mention the proposed TVAR-AGARCH-M model.

Data and results

We have considered three of the four largest net oil-importing countries of Asia, namely, India, Japan, and Korea. Our data is weekly time-series. The sample starts from January 2, 1995 for India and Japan whereas it starts from July 1, 1996 for Korea due to data availability. The sample ends on September 27, 2021 for all three countries. The aggregate stock indices are NIFTY50 for India, NIKKEI225 for Japan, and the KOSDAQ Composite Index for Korea. All financial data are collected from the

Conclusions

This paper is motivated by the unsettled debate on how oil price oscillations affect stock returns. The fact that unfavorable oil price fluctuations are risky for stock markets in large oil-importing countries, certain economic benefits can be derived from portfolio diversification and hedging these risks across international stock markets. We contribute to this ongoing debate by marrying diverse strands of literature on stock-oil integration, as documented in Smyth and Narayan (2018) but in

CRediT authorship contribution statement

Kushal Banik Chowdhury: Conceptualization, Methodology, Investigation, Formal analysis, Software, Validation, Writing – review & editing. Bhavesh Garg: Ideation, Literature review, Visualization, Writing – original draft, Writing – review & editing.

Acknowledgments

The authors are grateful to the Editor-in-Chief, Samuel Vigne, and an anonymous referee for providing valuable suggestions, which improved this paper significantly. The usual disclaimer applies.

References (45)

  • AbhyankarA. et al.

    Oil price shocks and the stock market: evidence from Japan

    Energy J.

    (2013)
  • AhmadW. et al.

    On the intraday dynamics of oil price and exchange rate: What can we learn from China and India?

    Energy Econ.

    (2020)
  • AlouiC. et al.

    The effects of crude oil shocks on stock market shifts behaviour: A regime switching approach

    Energy Econ.

    (2009)
  • AngA. et al.

    International asset allocation with regime shifts

    Rev. Financ. Stud.

    (2002)
  • BattenJ.A. et al.

    Hedging stocks with oil

    Energy Econ.

    (2021)
  • BjørnlandH.C.

    Oil price shocks and stock market booms in an oil exporting country

    Scott. J. Political Econ.

    (2009)
  • BollerslevT. et al.

    Quasi-maximum likelihood estimation and inference in dynamic models with time-varying covariances

    Econometric Rev.

    (1992)
  • BouriE.

    Oil volatility shocks and the stock markets of oil-importing MENA economies: A tale from the financial crisis

    Energy Econ.

    (2015)
  • CaporaleG.M. et al.

    Oil price uncertainty and sectoral stock returns in China: A time-varying approach

    China Econ. Rev.

    (2015)
  • ChangK.-L. et al.

    Does crude oil price play an important role in explaining stock return behavior?

    Energy Econ.

    (2013)
  • ChenS.-S.

    Do higher oil prices push the stock market into bear territory?

    Energy Econ.

    (2010)
  • ChoiK. et al.

    Volatility behavior of oil, industrial commodity and stock markets in a regime-switching environment

    Energy Policy

    (2010)
  • ChowdhuryK.B. et al.

    Regime-dependent effects of uncertainty on inflation and output growth: evidence from the United Kingdom and the United States

    Scott. J. Political Econ.

    (2018)
  • ChowdhuryK.B. et al.

    Nonlinear relationships between inflation, output growth and uncertainty in India: New evidence from a bivariate threshold model

    Bull. Econ. Res.

    (2021)
  • DiazE.M. et al.

    Oil price volatility and stock returns in the G7 economies

    Energy Econ.

    (2016)
  • FilisG. et al.

    Dynamic correlation between stock market and oil prices: The case of oil-importing and oil-exporting countries

    Int. Rev. Financ. Anal.

    (2011)
  • FisherI.

    Theory of Interest: As Determined by Impatience to Spend Income and Opportunity to Invest It

    (1930)
  • FuJ. et al.

    The time-varying connectedness between china’s crude oil futures and international oil markets: a return and volatility spillover analysis

    Lett. Spatial Resour. Sci.

    (2022)
  • GrierK.B. et al.

    The asymmetric effects of uncertainty on inflation and output growth

    J. Appl. Econometrics

    (2004)
  • GüntnerJ.H.

    How do international stock markets respond to oil demand and supply shocks?

    Macroecon. Dyn.

    (2014)
  • HamiltonJ.D.

    A new approach to the economic analysis of nonstationary time series and the business cycle

    Econometrica

    (1989)
  • JiaX. et al.

    How do correlations of crude oil prices co-move? A grey correlation-based wavelet perspective

    Energy Econ.

    (2015)
  • Cited by (1)

    View full text