High–low volatility spillover network between economic policy uncertainty and commodity futures markets

• Published date: 01 August 2024
• Author: Youtao Xiang,Sumuya Borjigin
• Date: 01 August 2024
• DOI: http://doi.org/10.1002/fut.22511

Received: 8 December 2023

|

Accepted: 15 April 2024

DOI: 10.1002/fut.22511

RESEARCH ARTICLE

High–low volatility spillover network between economic

policy uncertainty and commodity futures markets

Youtao Xiang |Sumuya Borjigin

School of Economics and Management,

Inner Mongolia University, Hohhot,

China

Correspondence

Sumuya Borjigin, School of Economics

and Management, Inner Mongolia

University, West Rd College, No. 235,

Hohhot 010021, China.

Email: sumuya@imu.edu.cn

Funding information

National Natural Science Foundation of

China, Grant/Award Number: 71861029;

Natural Science Foundation of Inner

Mongolia, Grant/Award Number:

2022MS07022; Project of “Grassland

Talents”in Inner Mongolia Autonomous

Region, Grant/Award Number: 12000‐

12102821; Basic Research Funds for

Universities Directly under Inner

Mongolia Autonomous Region,

Grant/Award Number: 20700‐54230329;

The Project of Innovation Research in

Postgraduate at Inner Mongolia

University, Grant/Award Number: 11200‐

54220396

Abstract

Based on the formation and evolution of systemic risk, we study the high‐low

volatility spillovers between economic policy uncertainty (EPU) and commod-

ity futures and identify the source of risk accumulation and risk outbreak, as

well as the corresponding contagion mechanisms. Upon comparing topologi-

cal characteristics on each volatility layer, our results demonstrate that high

and low volatility spillover networks have different network characteristics

and evolution behaviors. At the system level, high volatility spillovers are

relatively stronger than spillovers in in low volatility network, while the risk

propagation efficiency in the low volatility network is higher. At the market

level, EPU is not only an important risk‐emitter but also a risk‐recipient most

of the time. Additionally, compared with high volatility network, low volatility

network characteristics have greater predictive ability for risk spillover among

commodity futures, which means that it contains additional information and

provides early warning signals for financial stress.

KEYWORDS

commodity futures, economic policy uncertainty, high and low volatility spillovers,

multilayer networks, TVP‐VAR extended joint connectedness

JEL CLASSIFICATION

C50, G12, G20

1|INTRODUCTION

In recent years, due to the frequent occurrence of crisis events, the global economy has been severely impacted, for

example, the COVID‐19 pandemic, the sharp decline in international oil prices in March 2020, and the Russia–Ukraine

conflict, which intensified the downward risks in national economies and increased exacerbated economic policy

uncertainty (EPU) in financial market (Qiao et al., 2022). In the context of frequent “black swan”events in financial

market, EPU triggered market volatility, with fluctuations observed in the commodity futures markets (Fang

et al., 2018; Jiang et al., 2023). Moreover, the impact of EPU

1

is an important risk factor that drives the commodity price

fluctuation and turbulence of the commodity market (Li et al., 2023). EPU can impact the commodity market through

multiple channels (Ren, Tan, et al., 2022), for example, economic policy changes can cause rational economic agents to

J Futures Markets. 2024;44:1295–1319. wileyonlinelibrary.com/journal/fut © 2024 Wiley Periodicals LLC.

|

1295

1

Baker et al. (2016) proposed EPU.

be unable to accurately observe, analyze, and predict in their investment activities. Consequently, an increase in EPU

leads to a reallocation of investment, thereby altering the spillover effects among commodities.

In this context, many researchers and investors have focused on the association between EPU and commodity

markets, but there are also several notable limitations. A substantial body of related studies mainly focus on bivariate

pairwise relationships between EPU and commodity futures markets. For example, Lyu et al. (2021) explored the

impact of EPU shocks on Chinese commodity futures market using the time‐varying parameter vector autoregressive

(TVP‐VAR) method, and found that EPU shocks have a negative effect on commodity market. Huang et al. (2021) also

found similar results, that is, EPU has a significant negative impact on commodity markets. Zhu et al. (2020) studied

the effect of EPU on agricultural and metal commodity futures returns from a quantile perspective. Meanwhile,

scholars predominantly concentrate on a few typical commodity futures markets, neglecting multiple commodities. As

a result, the results may be distorted by incomplete commodity markets, as shocks in one market could be mistakenly

attributed to another if a market is not included in the volatility spillover network (Huang & Liu, 2023). For instance,

Guo et al. (2023) considered only agricultural markets; Jain et al. (2023) considered solely the precious metal markets.

Chen and Mo (2023) considered only the gold markets, but they neglect other commodity markets. Jiang et al. (2023)

studied that the spillover effects between Chinese economic policy uncertainty (CEPU) and commodity markets, but

ignored the very important oil markets

2

(Ding et al., 2021). Thus, we attempt to contribute to growing yet limited

literature that pay attention to the risk spillover effects between EPU and commodities by including more commodity

assets in empirical analysis, which offers the complete picture of risk spillovers among commodity markets.

Our work is also motivated by literature that studies the effect of high and low volatility of stock market (Danielsson

et al., 2018). They argue that systemic financial risk has two dimensions: time and space. Risk contagion is the core element

of systemic financial risk in time and space dimension. In terms of the time dimension, systemic financial risk emphasizes

its formation and evolution process, which could be divided into two stages: risk accumulation and risk realization.

Specifically, the accumulation stage of systemic financial risk belongs to the stage of low volatility in which asset prices

fluctuate less. Simultaneously, the period of systemic financial risk outbreak is marked by an overall increase in asset price

volatility. Furthermore, following Brunnermeier and Sannikov (2014), we note that stable economic and financial

environment is a hotbed for risk accumulation, and low volatility environment induces risk‐taking behavior of economic

entities, which leads to excess lending and leverage and increases the probability of financial crisis. As a consequence, the

low‐volatility market environment actually carries a high level of potential risk (Danielsson et al., 2018), which is of great

significance for the prevention and early warning of financial risk. Simultaneously, with the globalization of the economy

and the acceleration of the global financial process of commodity trading (Jiang et al., 2023), as concerns about commodity

markets connectedness have increased and economic activities have become more complex (Gong & Xu, 2022), the

literature on the spillover effects between commodity markets has shifted to a more systematic social network analysis

approach. Thus, this paper aims to explore the high volatility and low volatility spillover effects between EPU and

commodities by utilizing network analysis approach. As far as we know, little attention has been paid to risk spillover

networks between EPU and commodities from high and low volatility perspective.

Motivated by the abovementioned analysis, we aim to explore the two following issues: (1) What are the low‐

volatility and low‐volatility spillover networks? (2) From the perspective of risk accumulation and risk outbreak, are

significant differences in the spillover effects between EPU and commodity markets in the high and low volatility

spillover networks? (3) Who are the risk emitters or receivers in the commodity markets?

To shed light on the first issue, this study uses a volatility decomposition method (Danielsson et al., 2018)andTVP‐VAR

connectedness approach (Balcilar et al., 2021) to estimate the high‐volatility and low‐volatility spillover effects of EPU and

commodities. Compared with the traditional mean, volatility and tail spillover networks (Geng et al., 2021; Mokni

et al., 2020;Qiao&Han,2023). The main advantage of high‐low volatility spillover networks is that it offers an effective

method to capture the heterogeneous volatility spillovers among commodity markets. In view of the second question,

following Gong, Wang, et al. (2023)andDaietal.(2023), we compare the differences between high and low volatility

spillover networks from system‐level and market‐level. Regarding the third question, we calculate the net spillover effect of

each commodity market from static and dynamic perspectives, which can distinguish risk emitters and receivers.

Our paper provides several contributions and innovations. First, our paper contributes to the comprehensive

consideration of risk spillovers between EPU and commodity markets. Based on two dimensions of risk accumulation

2

According to Ahmed and Sarkodie (2021), energy commodities, such as crude oil and natural gas, play a pivotal role in driving global social and

economic development, serving as essential natural resources utilized across various economic sectors, for example, industry and transportation.

1296

|

XIANG and BORJIGIN