How does the Chinese economy react to uncertainty in international crude oil prices?

Highlights

• A structural vector autoregressive model is used to investigate the impact of oil price uncertainty on the Chinese economy.

• Increasing uncertainty of oil prices is found to reduce real GDP growth and investment.

• Chinese government uses monetary and fiscal instruments to offset the impacts of oil price uncertainty.

• No asymmetry is found for the oil price uncertainty effects.

• The impacts of oil price uncertainty are significantly greater in eastern China.

Abstract

This paper investigates the dynamic impacts of uncertainty in international crude oil prices on the Chinese economy. We use two measures, sample standard deviation and conditional standard deviation estimated from a GARCH (1,1) model, to calculate uncertainty in oil prices. We find that an increase in volatility in oil prices tends to reduce the real gross domestic product (GDP) and investment, which in turn encourages the Chinese government to stabilize the economy through expansionary fiscal and monetary policy. Furthermore, when we differentiate the impacts of increases and decreases in oil price uncertainty, we obtain a symmetric result. An increase in oil price uncentainty reduces real GDP and investment, while a decrease boosts the macroeconomy. We attribute the effect of decreasing uncertainty to the combined factor of falling uncertainty and an expansionary monetary and fiscal policy. A cross-sectional check related to economic geography indicates that uncertainty shocks to oil prices has a significantly greater negative impact on real GDP and investment in eastern China, where the economy is more industrialized and depends more heavily on oil than other regions.

Introduction

Since the first oil crisis in the 1970s, oil shocks and their macroeconomic impacts have drawn worldwide attention. A nexus between oil shocks and economic growth was first proposed by Hamilton (1983) for the US market, and it was later extended to other countries around the world (e.g., Zhang, 2008). It has become one of the most intensively investigated issues. In recent years, the impacts of oil shocks on other aspects of the economy have also been widely discussed–for example, on the stock market (e.g., Broadstock, Cao, & Zhang, 2012; Zhang, 2017); on consumption (e.g., Zhang, Broadstock, & Cao, 2014); and on monetary policy (e.g., Bernanke, Gertler, & Watson, 1997). Some studies further investigate the role of the transmission mechanisms on the oil price–macroeconomic nexus (e.g., Shi & Sun, 2017). Other studies examine how macroeconomic uncertainty affects oil prices (for a recent review, see Shen et al., 2018a, Shen et al., 2018b). Understanding the complex relationship between oil shocks and economic factors has broader implications for both scholars and policy makers. Moreover, an economy is an integrated system, and all these factors tend to interact with each other, so it is necessary to expand the bivariate analysis to multivariate setups (e.g., Bernanke et al., 1997).

Oil prices are creating greater concern for two widely recognized reasons. First, oil is a key input for modern production, thus the rise and fall in oil prices are directly relevant to a firm's production cost (e.g., Broadstock, Wang, & Zhang, 2014). Second and even more important, changes in the second-moment measure of oil prices affect a firm's expectations, thus helping to shape a firm's current production and investment decisions (e.g., Federer, 1996). Differentiating these two types of changes in oil prices seems to be crucial. In reality, the first- and second-moment changes in oil prices are not necessarily synchronized. Volatility can increase in both price-increasing and -decreasing periods (see Fig. 1). Fig. 1 shows that volatility in oil prices increased dramatically during the past two oil price-decreasing periods: 2008Q2–2010Q1 and 2014Q1–2015Q3 (see Figure A1 for more discussion). Although price changes in oil prices and their variants (e.g., Mork, 1989) have been the key variables in the majority of existing studies, the uncertainty that arises from the second-moment changes in oil prices has not been explored.

This paper investigates the dynamic impacts of uncertainty in international crude oil prices on the Chinese economy. We use a structural vector autoregressive (SVAR) model to investigate the responsiveness of the Chinese economy to uncertainty in international oil prices. Our paper is closely linked with Kim, Hammoudeh, Hyun, and Gupta (2017), who also use an SVAR model to study oil shocks to the Chinese economy. Their paper, however, concentrates on the first-moment price changes and examines the monetary policy reactions by the Chinese government to international crude oil markets. We believe that Chinese officials are reacting not only to oil price changes but also to uncertainty in international crude oil markets, and these reactions can be quite different. Since the seminal work of Bloom (2009), the causes and impacts of uncertainty have attracted greater attention in the economic literature (e.g., Baker, Bloom, & Davis, 2016; Carriero, Clark, & Marcellino, 2018; Jurado, Ludvigson, & Ng, 2015). We also include many other economic factors in our model to better reflect the reaction of the Chinese economy to oil price uncertainties.

In 2018 China became the world's largest consumer of oil, which is part of the motivation for our focus on China in particular. For this reason, uncertainty in crude oil prices is highly relevant for policy makers in China with respect to the first-moment price changes. China relies heavily on international crude oil markets, and its dependence is projected to rise from 68% in 2017 to 80% in 2035 (e.g., Ji, Zhang, & Zhang, 2019a; Ji & Zhang, 2019). Although increasing demand in China is considered one of the main driving forces for oil price changes, it remains a price taker on international crude oil markets and tends to pay a high premium (e.g., Zhang, Shi, & Shi, 2018a), which is attributed to the dominant pricing power of developed countries for crude oil. In 2018, China launched its first crude oil futures, a reflection of Chinese policy makers' attempt to gain pricing power. However, challenging the position of international benchmark prices such as Brent and West Texas Intermediate (WTI) remains a long-term goal (e.g., Ji & Zhang, 2019; Zhang, Ji, & Kutan, 2019). Since the 2008 global financial crisis, economic growth in China has slowed down and is considered to have achieved a “new normal” stage (e.g., Liu, Shi, & Laurenceson, 2018). Zhang, Lei, Ji, and Kutan (2018b) use the Economic Policy Uncertainty index constructed by Baker et al. (2016) to study the interactions of global markets including oil. In general, external shocks such as uncertainty in international oil markets can have strong impacts on the Chinese economy and inevitably trigger many government reactions through monetary and fiscal policies.

When international crude oil markets became more volatile, cyclical components of Chinese real GDP and investment both decreased, as seen in Fig. 1. This indicates that increases in the oil price volatility tend to inhibit growth in real GDP and investment even during periods when prices are declining and suggests the necessity of investigating the macroeconomic impacts of uncertainty shocks of oil prices despite declines in oil prices. Our paper therefore concentrates on the second-moment change in oil prices and explores its implications for the Chinese macroeconomy.

Our study contributes to the existing literature by adopting a systemic approach and using uncertainty in oil prices in our model. Methodologically, our work differs from the existing literature in three respects. First, the paper employs both sample standard deviation and conditional standard deviation estimated from a GARCH (1,1) model to characterize uncertainty in oil prices and investigates the impacts of oil price volatility shocks on China's real GDP and investment. Some economists have studied other countries using one or two of these measures, such as Ahmed and Wadud (2011) examining Malaysia and Bashar, Wadud, and Ahmed (2013) looking at Canada. We build on their work by studying China, a major oil consumer and importer. Second, following Mork (1989), we use samples comprising periods during which uncertainty was increasing and decreasing, to see whether the impacts of uncertainty shocks to oil prices are symmetric in those periods. Since the work of Mork (1989), the asymmetric impacts of rises and falls in oil prices have been confirmed in most economies. But a similar analysis has not been conducted to identify symmetry, or its opposite, in the volatility in oil prices. Our study aims to fill this gap. Third, we identify the cross-sectional difference in the dynamic impacts of uncertainty shocks to oil prices by examining China in terms of its economic geography to reveal the geographically heterogeneous impacts on oil consumption dependence.

The rest of the paper is organized as follows. Section 2 reviews and summarizes related strands of the literature in greater detail. In section 3, we discuss our data and empirical methodology. Section 4 presents the benchmark empirical results and relevant explanations. In section 5, we discuss extended results with respect to asymmetry and geographical heterogeneity. We conclude and discuss potential policy implications in section 6.