key: cord-0709659-uc5efvto
authors: Xu, Weijia; Li, Aihua; Wei, Lu
title: The Impact of COVID-19 on China’s Capital Market and Major Industry Sectors
date: 2022-12-31
journal: Procedia Computer Science
DOI: 10.1016/j.procs.2022.01.011
sha: 979b3dcbb754e68837c98b4bc71b000570e0f681
doc_id: 709659
cord_uid: uc5efvto

This paper studied the impact of COVID-19 on China’s capital market and major industry sectors via an improved ICSS algorithm, a time series model with the exogenous variable and a non-parametric conditional probability estimation. Through the empirical analysis, it is found that the epidemic has no significant impact on the return of the stock and bond markets, but it has increased the market volatility and the impact on the stock market volatility is gradual and more obvious. There are significant differences in the significance, direction and duration of the epidemic on different sectors. In addition, the impact of COVID-19 has been gradual in some industries and rapid in others. Different industries show different sensitivities in their response to COVID-19. Based on the analysis of the impact, this paper put forward the corresponding suggestions for investment strategies and macro-control decisions.

The breakout of COVID-19 in 2020 caused a serious public health emergency in China, which affected people's life and production. Nowadays, the connection between the finance and the real economy in China has become deeper and more complex. The impact of the crisis significantly increases risk spillovers [1] . Therefore, the impact of COVID-19 on production activities can be transmitted to capital markets through various channels and mechanisms. The study of the impact of COVID-19 on the capital market and industry sectors can help to avoid risks and provide suggestions to control countermeasures.

Public emergencies have different continuous, immediate and weakening effects on financial markets [2] . Some emergencies have significant impacts on the stock market and specific sectors, such as aviation, tourism

The breakout of COVID-19 in 2020 caused a serious public health emergency in China, which affected people's life and production. Nowadays, the connection between the finance and the real economy in China has become deeper and more complex. The impact of the crisis significantly increases risk spillovers [1] . Therefore, the impact of COVID-19 on production activities can be transmitted to capital markets through various channels and mechanisms. The study of the impact of COVID-19 on the capital market and industry sectors can help to avoid risks and provide suggestions to control countermeasures.

Public emergencies have different continuous, immediate and weakening effects on financial markets [2] . Some emergencies have significant impacts on the stock market and specific sectors, such as aviation, tourism and accommodation [3] [4] [5] . However, the impact of public health emergencies on the capital market is rarely discussed. Public health emergencies are the results of the interaction of natural and human factors [6] . More attention should be paid to the continuity and variability, rather than the specific day after the incident [7] . Some scholars analysed the influences of SARS with traditional event analysis methods [8] . The researches on the COVID-19 show that it affects the insurance industry [9] and the pork market [10] .

The rest of the paper is as follows. In section 2, we discussed related works. In section 3 we constructed the model and explained data. Section 4 is empirical analysis, and section 5 contains main conclusions and recommendations.

Public health emergencies can disrupt a country's normal production and cause huge impacts on market entities such as enterprises and consumers [11] . The epidemic endangers people's health and increases the pressure on government's public health expenditures, and high risk of infection leads to low production efficiency and breakages of capital chains. Yang et al. (2020) [12] found that the SARS led to a decline in consumer confidence and production price index. There is strong volatility spillover between the stock market and the macro economy [13] , and shocks to the real economy can easily be transmitted to the capital market.

In addition, the influence of COVID-19 also transmits to the capital market through psychological factors. Investors' negative emotions due to the epidemic will affect their decisions and increase their risk aversion [14, 15] . Shan (2011) [16] studied the impact of Wenchuan earthquake and found that negative investor sentiments caused by the earthquake were the main cause of the decline in stock returns. In addition, Chu and Liu (2003) [17] believed that investors would generate information-enhanced herd effect due to panic information. Similar to SARS, COVID-19 will also change investment behaviours through psychological factors and increase the uncertainty of the capital market. Therefore, we proposed hypothesis 1.

H1：COVID-19 will have a shock effect on the capital market and increase the market volatility. Similarly, the impact of COVID-19 on different industry sectors is mainly transmitted through the real industries and the psychology and behaviours of investors. Smith (2006) [18] found that a higher risk of infection reduced people's demand for tourism, transportation, and other entertainment industries. For example, the pandemic of H1N1 has caused significant losses to the tourism, catering, and aviation industries in some countries [19, 20] . During the SARS, China's tourism and hotel sector suffered a strong negative impact [8] . In contrast, the impact on industry was relatively limited [21] . Therefore, various industries have different relationships with the epidemic, and the hypothesis 2 is proposed.

H2：COVID-19 has significant impacts on some industry sectors, and the impacts on different industry sectors are obviously heterogeneous.

We first used the improved ICSS in [22] to find the structural changes of stock and bond volatility after the epidemic. The method determines breakpoints by calculating and comparing it with the threshold.

( 1 )

In this paper, represents the cumulative sum of squares of the stock and bond returns. The parameter is estimated using parameter estimation methods. If is larger than the threshold, the point at that moment is a volatility structural change point.

To analyse the direction and magnitude of the impact, we built EGARCH models and introduced exogenous variables to describe changes in the epidemic. Since the bond index has no ARCH effect in the sample interval, the EGARCH model is only established for stock indexes. In addition, by comparing the fitting results of models, we only introduced the variables of COVID-19 into the variance equation:

is the exogenous epidemic variable. Since the daily notification of COVID-19 is the data of the previous day, a lagging variable is selected for modeling.

is 0 before the epidemic and is real data after that. The impact of COVID-19 on the entire economy and society is more systematic, direct and significant during the sample window, so except for the outbreak and evolution of COVID-19, it can be considered that the main factors affecting the capital market have not changed significantly. Moreover, the sample interval is divided into two stages based on the changes of COVID-19, and the variance equation is as follows: ( 4 ) Furthermore, we studied the impact of COVID-19 on different industry sectors of the capital market. is the abnormal return of a certain industry sector index at t, and is the cumulative abnormal return in the period from to . Using the method in [7] , the non-parametric conditional probability distribution of abnormal returns is obtained through local polynomial regression.

To study the impact of COVID-19 on the capital market, we selected the Shanghai Composite Index (SSEC), Shenzhen Component Index (SZI), and CSI300 to represent the stock market, and the government bond index (GBI) and corporate bond index to represent the bond market. To study the impact on different industry sectors, ten SSE industry indexes were selected. We used the daily closing prices to calculate logarithmic returns.

In terms of COVID-19 data, the number of diagnosed people in Chinese mainland was selected. The data sources are the official website of the Wuhan Municipal Health Commission and Netease News. We selected the sample interval of the epidemic data from December 31, 2019 to May 13, 2020. Since the number of confirmed diagnoses reached the maximum on February 17, 2020, we used this date as the boundary and divided the epidemic into the rapid increase stage and the slow decrease stage. For studying the impact on the capital market, we selected data from August 30, 2019 to May 13, 2020, and for studying the impact on different industry sectors, we set the estimation window from January 2nd, 2019 to December 30th, 2019. All capital market data comes from Flush Software.

The results of ICSS showed that only the SZI has a volatility change point at a significance level of 5%, and at the significance level of 10%, all indexes except the corporate bond index have change points. The break points of stock volatility are all located at Jan. 22 nd , 2020, while they are on Jan. 23 rd , March 6 th and 13 th of the government bond. Figure 1 shows the change in the number of daily diagnosed people and the positions of the breakpoints of indexes. It shows that the structural breakpoints of volatility in stock indexes are at the early stage of the epidemic. Dividing the volatility of the government bond index based on its breakpoints, we found that the fluctuations of the government bond changed later and returned to a small level. We further used EGARCH models with exogenous variables to model index returns and epidemic changes, and quantitatively analyzed the impact of the epidemic on market returns and volatility. We normalized the number of people to [0,1], and recorded the processed data as .Since in the sample interval the bond indexes do not have the ARCH effect, we established EGARCH models only for stock indexes, and built autoregressive moving average models for bond indexes. We established the whole stage and the separated stages models for the Shanghai Composite Index, Shenzhen Component Index and CSI 300 Index. Based on the parameter significance and model fitting effect indicators, adding the COVID-19 variable to the mean equation failed to improve the model fitting effect and the coefficient of the exogenous variable was not significant, so the COVID-19 variable was only introduced in the variance equation. The parameter estimation results of COVID-19 variables and goodness-of-fit of models for three indexes are shown in Table 1 . Z value is shown in brackets, "**" means significant at the level of 5%, and "***" means significant at the level of 1%. The value is 0 before the December 31, 2019, and is after that date. The ARCH effect tests on the residuals after the establishment of the EGARCH model shows that the conditional heteroscedasticity has been eliminated. From the estimation results of Table 1 , it can be seen that the coefficients of the COVID-19 are significant and positive, indicating obvious overall impacts of the epidemic on the stock market, which increases the volatility and the risk of the stock market.

Furthermore, we separated COVID-19 into two stages and analyzed the differences in their impacts. In the variance equations of stock indexes, the variables and describing the two-stage epidemic are introduced in the form of Eq. (5), where is December 31 st in 2019, is February 18th, and is May 13rd.

The estimated results of epidemic variables and goodness-of-fit of the two-stage models for stock indexes are shown in Table 2 . For SSEC, the impact of the early stage is slightly greater than that of the slow decrease stage; for SZI and CSI300, the impact in the latter stage of the epidemic is greater and more significant, which shows that the shock of COVID-19 is gradual and continuous.

Furthermore, we took the government and the corporate bond indexes to study the COVID-19's impact on the bond market with the whole stage and two-stage models. The results show that only in the two-stage model of the government bond index, the coefficient of the rapid growth stage of the epidemic is significant at the level of 10%. In summary, the volatility of stock indexes and the government bond index have undergone structural changes after COVID-19. The outbreak of the epidemic increased the market volatility, especially for the stock market. However, the epidemic has almost no impact on the returns of stock and bond indexes. 

The results of each industry index are shown in Figures 2 and 3 . The straight line parallel to the y-axis divides the two stages of the epidemic. From the two figures, SSE Materials, SSE Energy and SSE Finance have more negative abnormal returns and the cumulative abnormal returns show a downward trend, while the SSE Pharmaceuticals has more positive abnormal return and the cumulative abnormal return increases. The cumulative abnormal returns of SSE Information and SSE Telecom increase rapidly in the previous stage of the epidemic, while the cumulative abnormal return of SSE Consumption increases significantly in the later stage. Figure 4 shows that the abnormal returns of industry sectors have significant differences during the epidemic. The returns of the pharmaceutical and the information industries are mainly positively affected and the impact on pharmaceuticals is significant. On the contrary, the finance sector suffers an obvious negative impact and almost all abnormal returns are extreme changes. In addition, COVID-19 has negative effects on the consumption sector in general, but most of deviations are abnormal changes. The optional consumption sector shows positive abnormal changes in the early stage. However, more negative deviations occur later, indicating a lag in the impact the sector received. In addition, SSE Telecom and SSE Industry also show slow reaction to the epidemic. Moreover, we analysed the cumulative abnormal returns to study the persistence of the impact. The 5-day and 10-day analysis uses the non-parametric method and other periods are judged by statistic Z in [5] . From the CAR results of Table 3 , the early stage of COVID-19 has little impact on industry sectors. For the materials, industry, optional consumption, consumption and utilities, the impact is not continuous significant. The SSE Industry shows less sensitive to information about the epidemic due to the relatively complete industrial system and strong risk-defence capabilities. The energy sector and the finance sector suffer significant negative impacts and the cumulative abnormal returns of the entire event window are still significant, indicating the long duration of the effect. In addition, the energy sector has obvious lag in response to the epidemic information, while the finance sector is sensitive. The pharmaceuticals and information indexes are continuously and significantly affected, and their response to epidemic information is relatively fast. 

The impact of COVID-19 is continuous and gradual on market fluctuations, and is different on various industries. The outbreak increases the risk of the stock market. There are differences in the direction, Materials Industry Energy Optional Consumption Pharmaceuticals Finance Information Telecom Utilities significance and duration of the impact for industry sectors. The main reason is that industries have different sensitivities for the information, and the transmission mechanisms and speeds are various. In general, except for some sectors, the impact of COVID-19 on industry sectors is limited.

Overall, investors should avoid excessive panic emotion and adjust investment strategies to effectively avoid risks based on the characteristics of various industries. Regulatory agencies and government departments should pay more attention to monitoring abnormal market fluctuations, and prevent and control financial risks in key industries in the capital market.

Energy (+)1.0000** (+)1.0298** (-)-1.6540**

(+)2.3901** (+)2.1869** (+)2.4175** Finance (-)0.0879* (-)0.2478 (-)-1.6543** (-)-1.5452* (-)-1.7276**

Utilities (+)0.6010 (-)0

Abnormal returns: abnormal change

Abnormal returns: extreme change (-)Abnormal returns: extreme change (-)Abnormal returns: abnormal change

Measuring tail dependence using multivariate regression quantiles

Analysis about the impact of emergencies on international oil price. Mathematics in practice and theory

Market anomalies and disaster risk: evidence from extreme weather events

The costs of terrorism and the benefits of cooperating to combat terrorism. The secure trade in the APEC region (STAR) Conference

The shock effect of earthquake disaster on Chinese stock market

Scenario forecasting model and prevention-control measurements of important public health event based evolutionary game. Systems engineering -Theory & Practice

The impact of terrorism on financial markets: an empirical study

Empirical analysis of the impact of SARS on Chinese stock market

Effects of major public health emergencies on China's insurance sector --case studies of SARS and the COVID-19 pandemic

Study on the impact of COVID-19 on global pork market and China's pork trade. Price:Theory & Practice

The economic effects of the 1918 influenza epidemic

Macroeconomic shock, financial risk transmission and governance response to major public emergencies

Monetary policy, stock asset prices and economic growth

Sentiment and stock prices: The case of aviation disasters

Investor sentiment in the stock market. The Journal of economic perspectives

Psychological or real? The effect of the Wenchuan earthquake on China's stock market

Behavioral finance explanation of the impact of SARS on stock market

Responding to global infectious disease outbreaks: Lessons from SARS on the role of risk perception, communication and management

Short-term economic impacts of influenza A (H1N1) and government reaction on the Mexican tourism industry: an analysis of the media

The economic impact of H1N1 on Mexico's tourist and pork sectors

To cope with the impact of SARS and maintain the steady growth of industrial economy. Macroeconomic management

Testing for changes in the unconditional variance of financial time series

This research was partly supported by the National Natural Science Foundation and the Engineering Research Centre of Education Ministry for National Financial Security in Central University of Finance and Economics.