key: cord-0807875-2re39hpf
authors: Hashmi, Shabir Mohsin; Ahmed, Farhan; Alhayki, Zainab; Syed, Aamir Aijaz
title: The impact of crude oil prices on Chinese stock markets and selected sectors: evidence from the VAR-DCC-GARCH model
date: 2022-03-09
journal: Environ Sci Pollut Res Int
DOI: 10.1007/s11356-022-19573-5
sha: 8890a8cc4012baf95d27b98fe8c6185413bd06fa
doc_id: 807875
cord_uid: 2re39hpf

The interaction between oil and stock market returns is one of the most important relationships that have a significant influence on the economy of any country all over the world. Therefore, this paper investigates the impact of crude oil prices on the Chinese stock market and selected industries by using the VAR-DCC-GARCH model over the period from December 26, 2001, to April 30, 2019. The empirical results show that the impact of Brent crude oil prices on the Shanghai Composite Index and selected industries is significant. However, there are some variations in these relationships and the degree of influence on each differs during different sample periods. Brent crude oil prices exert substantial influence on some specific industries, like mining, chemical, nonferrous metals, and steel. Whereas, the volatility spillover effect of Brent crude oil prices is stronger within the mining, chemical, steel, nonferrous metal, building materials, building decoration, electrical equipment, electrical equipment, textile and garment, light manufacturing, public utility, and transportation industries than within other industries. When oil prices change abruptly, the risk of spillover impacts of oil prices on stock markets will also increase. In conclusion, the impact of Brent crude oil prices on the Chinese stock market is generally positive. Furthermore, the subsequent volatility of Chinese stock market prices will, in turn, influence the volatility spillover of Brent crude oil prices on the indexes. The result is an ongoing back and forth of changes in price volatilities.

Crude oil, as crucial energy in all industries, is regarded as the blood of modern industry; therefore, it has a significant impact on major macroeconomic factors of the worldwide economy. In recent decades, the price of crude oil fluctuates significantly, and this influences the companies' operational costs, and therefore, their sales. In fact, Brent crude oil price, one of the benchmarks of the global crude oil pricing system, has demonstrated instability since the twenty-first century. From 2000 to 2008, crude oil prices rose from $73 to $132.3 at most and experienced a steep decline after the subprime loan crisis to $39.95. When the world economy recovered from the crisis, crude oil prices also rose to over $100 in 2011. Nevertheless, in 2016, it fell to 26.68, which is the lowest ever. Until now, the crude oil price has risen to over $70. Consequently, this variation in the oil price affects the asset price and increases the volatility, which has a direct impact on the financial market. Nowadays, more and more individuals and institutions regard crude oil as a financial tool to hedge and increase values, which increase Responsible Editor: Ilhan Ozturk the volatility of oil price, thus increasing the volatility of the economy of a country, especially the stock market.

Being one of the largest economies in the world, China is not immune to such variation impact. In the last 4 decades, China has witnessed major changes in its economy, and this was accompanied by a revolution in all sectors. One of these sectors is its industrial sector that depends heavily on oil as its sole energy and its dependence on foreign oil reach 70.9%, with year-on-year growth of 2.5%. In fact, China is considered the largest oil import country in the world in 2018. According to China's customs, China imported 462 million tons of oil in 2018, with year-on-year growth of 10.1%, which indicates 9.24 million barrels of oil each day. Hence, the volatility of oil prices requires deep and comprehensive study since it will lead to great shocks on China's economy. Sahu et al. (2021) revealed that the rise in crude oil price, GDP, and population density will increase renewable energy use in the short run and in the long run as well. E.-C Hani (2019) examined the asymmetric influence of oil price shocks on the stock markets of GCC countries. The authors concluded that the negative influence of oil prices on the stock markets of Bahrain and Kuwait is more severe than the positive influence. On the other hand, Mohamed and Julien (2009) provided a contradictory view. The study concludes that oil prices share a positive relationship with the Oman, UAE, and Qatar stock markets. However, the stock movement is independent for Kuwait and Bahrain. Other studies, which have explored the asymmetric relationship are Reboredo (2010) and Wen et al. (2019) . Moreover, there is another piece of literature that explores the influence of oil price shocks on the stock prices of different industries. For example, Elyasiani et al. (2011) examined the impact of oil price shock on the stock returns of thirteen US industries. The result supports that oil price variations create a systematic asset price risk at the industry level.

The rest of the study is organized as follows: In the next section, we have provided the literature review. The "Data and methodology" section covers the methodology. The "Data selection and summary statistics" section offers data selection and summary statistics, while the "Conclusion and policy implications" section concludes the study with some policy implications, whereas references are given at the end of the study.

Since the 1980s, researchers and scholars have conducted many studies that observe the association between oil price and stock returns. Some went further and tried to see the causal and dynamic relationship between the two variables. The studies are focused on both oil-importing and oil-exporting countries. (Alhayki 2014; Bouoiyour et al. 2017; Mokni and Youssef 2019) . The subsequent section provides a comprehensive review of the nexus between crude oil prices and stock markets returns.

One of the pioneering papers by Jones and Kaul (1996) examined the dynamic nexus between oil prices shocks and stock markets performances in four developed economies using the cash flow dividend model. The study concludes that oil price fluctuations influence firms' cash flows, which reduce their profitability and thus create a direct influence on the stock market of the USA and Canada. However, oil price volatility has no significant influence on the stock market performances of Japan and the UK. Similarly, Park and Ratti (2008) investigated the causal relationship between oil prices and stock market performances in oil-importing and oil-exporting countries. The author concludes that oil prices directly impact the stock market performances of the USA and 12 European countries. Inversely, the study also concludes that oil prices positively influence the stock markets of oil-exporting countries. Jiang and Kong (2021) found the positive one-way spillover effect of international crude oil returns on China's energy stock returns. Yousaf et al. (2021) revealed a positive causal effect from Brazil and Mexico's stock price changes to the oil market during the global financial crisis. Using the VECM method, Alzyoud (2018) supported the above findings and concluded a positive association between oil prices and Canadian stock market performances. Roberto et al. (2017) further strengthen the above findings and confirm a direct association between oil prices and stock market performances of oil-exporting countries. However, the authors further added that the positive association remains the same even for the oil-importing countries. They reiterated that there is no distinction between oil-exporting and importing economies. On the other hand, a study conducted by Filis et al. (2011) on the stock markets performances of oil-importing and exporting countries negated the above findings. The authors concluded that oil prices have an adverse impact on the stock markets of both the oil importing and exporting countries.

In continuation to the above studies, which have only assessed the symmetric impact of oil prices on the stock markets, there are other studies that have explored both the symmetric and asymmetric relationships. Narayan and Sharma (2011) examined the impact of oil prices on 560 US firms. The results highlight that oil prices impact firms' share performance differently based on their locations. The authors also conclude that out of the 14 sectors, only 5 sector share returns are affected by sectorial regimes. Likewise, Mohanty et al. (2014) studied the influence of oil price shock on six major US industries. They concluded that oil price shocks have a significant negative influence on the allied industries. The oil price shock is more severe during the global economic crisis specifically during the COVID-19 pandemic (Corbet et al. 2021) . Recently, Yun and Yoon (2019) explored the impact of crude oil prices on the stock prices of the airline industries by taking the sample of four airlines (Korean, Asiana, Air China, and China Eastern Airlines). The study concludes that the volatility spillover effect is more severe than the return spillover. Besides, the study also concludes that the effect of oil price shock is more significant in small airlines compared to big airlines.

Crude oil import in Turkey is more responsive to changes in income level (Ozturk and Arisoy 2016) . Subsequently, based on the above review, we can infer that, broadly, the relationship between oil price shocks and stock returns is classified into three segments. For instance, oil-exporting countries, oil-importing countries, and the influence of oil price shocks on several industries. The enriched literature on the nexus between oil price shock and stock volatility shows mixed and inconclusive results. The results of most of the studies are classified as the negative effect of the oil prices on stock markets (Driesprong et al. 2008; Papapetrou 2001; Basher and Sadorsky 2006) , the positive effect of oil prices on the stock markets (Zarour 2006; Wang et al. 2013; Arouri and Rault 2012) , and trivial or no influence of oil prices on the stock market performances (Al Janabi et al. 2010; Apergis and Miller 2009) .

The literature review also highlights that there are still scant studies focused on exploring the sectoral and country-specific analysis of oil price shocks in a developing and emerging economy like China. Since the advent of Chinese oil reforms, many scholars have visited the relationship of oil price variations on different Chinese economic factors (Peng and Martin 1994; Chen et al. 2009; Du et al. 2010; Tang et al. 2010; Faria et al. 2009 ). However, as already discussed, the studies on sectoral analysis and oil prices shocks are limited. Against this backdrop, we have tried to revisit the above relationship using the novel the VAR-DCC-GARCH model.

In order to analyze how stock markets, react confronted with the volatility of oil prices we use the DCC-GARCH model to explore the dynamic correlation between oil prices and stock markets in China. VAR is a useful approach to link the macroeconomic variables such as exchange rate, inflation, global oil prices, and output. The VAR has a number of advantages, and it helps to identify the structural shocks through a Cholesky decomposition of innovations. The effects of global oil price shocks to other macroeconomic variables on domestic inflation would also be investigated under a VAR framework. As is well known, time-varying correlations are often estimated with multivariate GARCH models that are linear in squares and cross products of returns. A new model-the dynamic conditional correlation (DCC) is proposed by Engle (2002) . The DCC is a standard multivariate volatility model. Primarily, it is based on closing prices. In this study, we have jointly used the VAR-DCC-GARCH model. The main advantage of this model is that it is more robust in terms of non-normal errors. It also incorporates the presence of outliers, heterogeneity, and skewness in the dependent variable. Numerous studies have confirmed that when the VAR-DCC-GARCH model is applied, it gives the best results when analyzing the dynamic correlation between oil prices and stock markets.

The DCC-GARCH model assumes that residual term obeys the t-distribution for the sake of describing the peak and fat tails of financial assets' time series. To estimate the dynamic conditional coefficient of association, we need first to estimate the conditional variance of each market's return, which is then divided by residual and to obtain a standard residual series. Eventually, we use the series to estimate the dynamic conditional coefficient of association. The equations are as follows:

where r t is the return, Ω t−1 is the information set known, H t is conditional covariance matrix; D t conditional standard deviation diagonal matrix, that is

, h i,t is the conditional variance computed by the GARCH model, i is residual squared coefficient lagged p order, and j is the conditional variance coefficient lagged q order, among which i ≥ 0 , j ≥ 0 , and

where R t is dynamic correlation coefficient matrix, q 11,t is the conditional covariance standardized by residual, Q t is covariance matrix, t−m is the standardized residual lagged m orders, m is the coefficient of residual squared, n is the conditional variance coefficient, m ≥ 0 , n ≥ 0 , and m + n < 1 . In DCC-GARCH (1,1), the dynamic conditional coefficient between two financial variables is

Conditional dynamic coefficient denotes the degree of risk spillover, the larger 12,t is, the large the degree of risk spillover.

For the sake of examining the impact of crude oil prices on China's stock market, this paper uses Brent crude oil spot prices and Shanghai Composite Index to establish the model. As for the selection criteria, we used all industries listed in the Shenwan primary industry as it represents the stock performance of each industry in China's stock market.

The industry indexes include mining, chemical, steel, nonferrous metal (NM), building materials (BM), building decoration (BD), electrical equipment (EE), mechanical equipment (ME), national defense and military industry (NDMI), automobile, household appliances (HA), textile and garment (TG), light manufacturing (LM), commercial trade (CT), AFAF (agriculture, forestry, animal husbandry, and fishery), food and beverage (FB), leisure service (LS), medical biology (MB), public utility (PU), transportation, real estate (RE), electronic engineering (EEng), computer, media, communication, bank, non-bank financial institutions (NBFI), and comprehensive firms (CF).

The data is obtained from China's data provider company Wind Information for the period from December 26, 2001, to April 30, 2019, on a daily basis. Figure 1 below shows the historical trend of the Shanghai Composite Index and Brent crude oil. It is obvious that the two variables are moving almost in the same direction. Figure 2 shows the daily return of the Shanghai Index, Brent crude oil, and the industry indexes. Figure 1 demonstrates the volatility trends of oil prices and stock markets, from which we may discover that there exist phenomena of volatility clustering in all of these markets, and for the sake of accurately describing volatility, we are about to utilize the GARCH model, and the graphical presentation in Fig. 2 shows the volatility in Shanghai Composite Index, Brent crude oil and SW industry index.

With the data we obtain, to calculate the return of oil and stock indexes, we use the following formula:

is the return of either stock index or oil prices, P t is the price of oil or stock index in the period of t, and P t−1 is the price of oil or stock index in the period of t − 1.

The result of descriptive statistics is as follows.

Tables 1 and 2 present the descriptive statistics for all variables over the selected sample period. The results show that Shanghai Composite Index has a higher mean than Brent oil price. Among industries, MB has the highest mean while TG

has the lowest. With the exception of Brent crude oil, bank, and NBFI, all variables have negative skewness. Besides, the Kurtosis for all variables exceeds 3, so this rejects the hypothesis of normality for all variables.

Before establishing a VAR model, a unit root test must be carried out. After processing the data, the result is can be seen in Table 3 and Fig. 3 .

Our data variables are stationary as the results of both Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) are significant. In other words, all variables pass the unit root test and the VAR model can be established in the following procedure.

To test the asymmetric influence of oil price on Shanghai Composite indexes, we used a vector error autoregressive model (VAR). Table 4 presents the results of this model. Oil prices proved to have a significant impact on approximately one-third of the sample industries. Their .

From Table 5 , we can see that Brent (− 1) has a significant impact on the Shanghai Composite Index and most industry indexes, especially on mining, chemical, nonferrous metal, and steel, with a significant level of 99%.

The Granger causality is shown as follows.

From Table 6 , we can see that it is safe to conclude that all factors in this model are interlinked to each other, and they influence each other, we may continue constructing our model. Figure 4 shows the impulse responses of the Asian stock market to oil prices, from which we can come to some significant conclusions. As there are 30 impulse responses, we only present some typical graphs and summarize the common characteristics. Other responses will be shown in the Appendix.

To begin with, according to all the responses, we can see that when the oil prices offer stock indexes a positive impact, the stock indexes always go up, only that the degree of the impact to each stock index is different from that of others. Besides, after going up in the first 3 periods, the impulse response heads down to zero eventually at the 4th period. Last but not least, the degrees of responses are different, and they can be shown in the following In accordance with Table 7 , the impulse responses of the indexes can be divided into two different levels, 0-0.1 and 0-0.004, from which we may conclude that the impact of Brent crude oil on different industries is different. Additionally, the impulse response of Brent crude oil on SH is relatively minor, while on the other hand, the impulse response of Brent crude oil on Mining is significantly bigger, which demonstrates that compared to the market index, Brent crude oil exerts more influence on some certain industries and the trend of the conditional dynamic coefficients can be seen in Fig. 5 . The average and variance of the conditional dynamic coefficients are as follows:

To conclude from Table 8 , firstly, the average of the conditional dynamic coefficients fall into (0.38, 0.11), and the variance of it are almost of the same level (in (0.005, 0.009) and are mostly approximately 0.007), from which we may see that the volatility spillover of Brent crude oil on Chinese stock market is stable in every industry. Secondly, from the numbers, the volatility spillover of Brent crude oil on every industry is different and the spillovers are larger on mining, chemical, steel, NM, BM, BD, EE, ME, TG, LM, PU, and transportation (bigger than 0.0641). Thirdly, for most of the time, the coefficients are positive, and from 2008 to 2007 and 2008 to 2012, oil prices fell dramatically, leading to the coefficient to drop sharply, and when oil prices started to rise, the coefficient began to increase as well, which demonstrates that when oil prices change fiercely, the risk spillover of oil prices to Chinese stock market will also increase. Last but not least, the impact of Brent crude oil on the Chinese stock market is mostly positive, and the volatility of the Chinese stock market will influence the volatility spillover of Brent crude oil on the indexes. With regard to the details, interestingly, most of the coefficients fluctuate around 0.3 and seldomly fall under 0, meaning that there is the same limit of influence of Brent crude oil exerting to every index.

To begin with, the impact of Brent crude oil on the Shanghai Composite Index and every industry is significant, especially on mining, chemical, nonferrous metal, and steel, with a significant level of 99%. To be exact, confronted with the impulse of Brent oil price, the response of Shanghai Index and SW industry indexes increase first, and the degree of response decreases slowly in the next period, and eventually head down to zero fast in the third period. Apparently, facing the volatility of oil price, Shanghai Index and SW Industry indexes react almost the same; however, the degree of the influence on each index is different. That is, compared to the market index, Brent crude oil exerts more influence on certain industries. This can also be seen from the result of the DCC-GARCH model, which shows the dynamic correlation coefficients of oil to the Chinese stock market and SW industry indexes. According to the result, the volatility spillover of Brent crude oil on the Chinese stock market is stable in every industry, and the spillovers are larger on mining, chemical, steel, NM, BM, BD, EE, ME, TG, LM, PU, and transportation. When oil prices change fiercely, the risk spillover of oil prices to stock markets will also increase. Last but not least, the impact of Brent crude oil on the Chinese stock market is mostly positive and the volatility of the Chinese stock market will influence the volatility spillover of Brent crude oil on the indexes. This paper first uses Brent crude oil prices, Shanghai Composite Index, and SW industry indexes to establish a VAR • The impact of Brent crude oil on the Shanghai Composite Index and every industry is significant. • Confronted with the impulse of Brent oil price, the response of Shanghai Index and SW industry indexes increase first, and the degree of response decreases slowly in the next period, and eventually head down to zero fast in the third period. • Facing the volatility of oil price, Shanghai Index and SW industry indexes responded identically; however, the degree of influence on each index varies. Hence, compared to the market index, Brent crude oil exerts more influence on certain industries, like mining, chemical, nonferrous metal, and steel. • The volatility spillover of Brent crude oil on the Chinese stock market is stable in every industry, while the spillovers are larger on mining, chemical, steel, nonferrous metal, building materials, building decoration, electrical equipment, electrical equipment, textile and garment, light manufacturing, public utility, and transportation. When oil prices change fiercely, the risk spillover of oil prices to stock markets will also increase. • The impact of Brent crude oil on the Chinese stock market is mostly positive, and the volatility of the Chinese stock market will influence the volatility spillover of Brent crude oil on the indexes.

Several policy implications can be drawn from this analysis. Owing to the fact that oil prices have a significant impact (1) Notes: the null hypothesis of statistical tests are rejected at 10% ( * ), 5% ( ** ), and 1% ( *** ) levels on the Chinese stock market and all industries, correspondent policies should be drawn up to cope with the upcoming impact of the recent fluctuation of Brent crude oil prices. According to our findings, when oil prices fluctuate significantly, the stock markets will fluctuate correspondingly, and the Shanghai stock markets' reaction will be great as well.

Therefore, actions should be taken to cope with the risk spillover to these markets. For the first part, financial supervision bureaus of all regions should monitor stock markets closely and prevent systematic risks when confronted with comparatively dramatic risks, stabilizing domestic financial orders. For the Notes: the null hypothesis of statistical tests are rejected at 10% ( * ), 5% ( ** ) and 1% ( *** ) levels Response of STEEL to BRENT second, investors of all markets should take into account the risk of spillover of oil to stock markets to prevent potential loss. When the volatility of oil becomes exceedingly high, those who invest in relevant assets or relevant industries should change tactics to avoid losses. What is more, when oil price fluctuates drastically, it is suggested that investors The trend of the conditional dynamic coefficients might seek opportunities in those sectors that are not so influenced by the oil price. When the oil price is experiencing steady growth, one can invest in some sectors highly related to the oil price. Last but not least, investments should be considered to promote the development of new energies to reduce our dependence on crude oil so that we will suffer from less impact from the volatility of oil.

Author contribution Sabir Mohsin Hashmi: conceptualization, literature review, and methodology (25%); Farhan Ahmed: literature review, data analysis, and drafting (25%); Zaynab Albayki: data, methodology, reviewing, and policy recommendations (25%); Amir Aijaz Syed: analysis, conclusion, and reviewing (25%).

The data used in this study can be provided by the corresponding author on reasonable request.

Ethics approval All the sources have been cited appropriately, and there is no such issue during this study.

All the authors will participate in the review and publication process.

All the authors have given their consent to publish this study.

The authors declare no competing interests. 

An empirical investigation of the informational efficiency of the GCC equity markets: evidence from bootstrap simulation

The dynamic co-movements between oil and stock market returns in: the case of GCC countries

Dynamics of Canadian oil price and its impact on exchange rate and stock market

Do structural oil-market shocks affect stock prices?

Oil prices and stock markets in GCC countries: empirical evidence from panel analysis

Oil price risk and emerging stock markets

Response of stock returns to oil price shocks: evidence from oil importing and exporting countries

Price causal relations between China and the world oil markets

Volatility spillovers during market supply shocks: the case of negative oil prices

One belt and one road: the grand geostrategy of China's rise

Striking oil: another puzzle?

The relationship between oil price shocks and China's macro-economy: an empirical analysis

Oil price shocks and industry stock returns

Dynamic conditional correlation: A simple class of multivariate generalized autoregressive conditional heteroskedasticity models

The effect of oil price on China's exports

Dynamic correlation between stock market and oil prices: the case of oil-importing and oilexporting countries

The impact of oil prices on stocks markets: new evidence during and after the arab spring in gulf cooperation council economies

The impact of international crude oil prices on energy stock prices: evidence from China

Oil and the stock markets

Oil price fluctuation, volatility spillover and the Ghanaian equity market: implication for portfolio management and hedging effectiveness

On the short-term influence of oil price changes on stock markets in GCC countries: linear and nonlinear analyses

Oil price risk exposure: the case of the US travel and leisure industry

Measuring persistence of dependence between crude oil prices and GCC stock markets: a copula approach

New evidence on oil price and firm returns

An estimation of crude oil import demand in Turkey: evidence from time-varying parameters approach

Oil price shocks, stock market, economic activity and employment in Greece

Oil price shocks and stock markets in the US and 13 European countries

Oil price shocks and policy responses in the post-reform Chinese economy

Nonlinear effects of oil shocks on stock returns: a Markov-switching approach

Internationalization of state multilatinas: a multi-case study in the oil sector

Investigating the asymmetry effects of crude oil price on renewable energy consumption in the United States

Oil price shocks and their short-and long-term effects on the Chinese economy

Oil price shocks and stock market activities: evidence from oil-importing and oil-exporting countries

Oil prices and Chinese stock market: nonlinear causality and volatility persistence

Adeel I (2021) Risk and Return transmissions from crude oil to Latin American stock markets during the crisis: portfolio implications

Impact of oil price change on airline's stock price and volatility: evidence from China and South Korea

Wild oil prices, but brave stock markets! The case of GCC stock markets