Covid-19 vaccine approvals and stock market returns: The case of Chinese stocks.

This paper investigates the Chinese stock market reactions to the announcements of Covid-19 vaccine approvals. These announcements generally impacted stock prices, but the impacts appeared to be heterogeneous across sectors. Particularly, firms in the manufacturing, wholesale, retail, and information technology sectors were persistently benefited. We also find that firms with poorer performance, smaller sizes, and greater ages reacted more positively compared to others.

Introduction

The coronavirus (SARS-CoV-2) spreads worldwide rapidly. By March 16, 2022, the total number of confirmed cases was over 462 million, and the cumulative death toll was over 6 million.1 The pandemic has significantly impacted the world economy. Because there was no specific drug in the early pandemic stage, many countries adopted preventive measures such as quarantine, social distancing, lockdowns, and travel restrictions to combat the spread of the virus. Traditional service industries, such as the tourism and catering industries, have been struck. On the other hand, because of the rising demand for medical resources across the globe, the pharmaceutical and healthcare industries might grow because of the pandemic. Covid-19 vaccines are important weapons against the coronavirus. Therefore, the approvals of Covid-19 vaccines provide hope and confidence in economic recovery.

In this paper, we first study how vaccine approvals affect the stock market in terms of cumulative average abnormal return (CAAR). Then we investigate their impacts on stocks in different sectors. Lastly, we explore the relationships between firm characteristics and cumulative abnormal returns (CARs). We find that vaccine approvals affect the stock market positively. In general, they generate positive CAARs for almost all sectors. We also find that certain firm characteristics, such as ROA, tangible assets ratio, and size, are negatively correlated to CARs. At the same time, age and financial leverage are positively correlated to CARs.

Previous studies often find that infectious diseases, such as SARS and flu, affect stock markets in a negative way (Chen et al., 2018, McTier et al., 2013). A growing number of studies investigate the effects of the Covid-19 outbreak on the stock markets (for example, see Eichenbaum et al., 2021, Phan and Narayan, 2020, Cao et al., 2021, Harjoto et al., 2021, Heyden and Heyden, 2021). Yan (2020) shows that stock prices in China fall sharply after the lockdown of Wuhan. He et al. (2020) and Wang et al. (2021) study Covid-19’s impacts on stock prices across different sectors in China. They find that the impacts are heterogeneous. Wang et al. (2021) also find that listed firms with higher debt levels were more negatively affected. Yi et al. (2021) show that corporate social responsibility rating is positively associated with CAR. Interestingly, Arteaga-Garavito et al. (2021) find that in many countries, the release of Covid-related news helps equities in general, regardless of whether the news is positive or negative.

Our paper makes several contributions. First, we investigate the impacts of the news of several Covid-19 vaccine approvals on the Chinese stock returns. Second, we provide a sectoral analysis, which deepens the understanding of how stocks in different sectors respond to this type of news. Third, we explore how firms’ characteristics correlate to their CARs in these events.

The rest of the paper is organized as follows: Section 2 describes the data and methodology. Section 3 presents the results and analysis. Section 4 concludes.

Data and methodology

In China, the government usually does not disclose the approval of a vaccine before issuing an official announcement. Therefore, the vaccine approvals are unexpected to the market participants. We examine the market reactions to the following events.2

We collect financial data of listed firms from the China Stock Market & Accounting Research Database.3 Firms in the financial industry and firms under special treatment are excluded.4 We employ the event study methodology. In particular, we use the market model (see Brown and Warner (1985)) to calculate CARs. The market index we use in this paper is CSI300.5 where is the rate of return of firm on day , and is the rate of return of the market index on day . The estimation window we use is , and we denote the estimates of and by and . The abnormal return (AR) of firm on day is Let . The CAR of firm from day to day around date (day ) is The CAAR from day to day around is where is the number of firms. Our benchmark model is specified as follows: where is the return on assets, is the tangible assets ratio, is the log of financial leverage, is the gap between the event year and the IPO year, and is the log of total assets.6 We control for time-fixed effects, province-fixed effects, and industry-fixed effects. We focus on the cases where , and .

Empirical results

Table 1 Panel A reports the stock market reactions to the four events. The CAARs are positive across the , , and event windows; in addition, they are all significant at the 1% level. These indicate that the market reacted positively to the announcements of Covid-19 vaccine approvals. Table 1 Panel B reports the results from the constant mean return model (see MacKinlay (1997)), and it shows that our results are robust.

Table 1

Cumulative average abnormal returns.

Window	CanSino&SP (Wuhan)			Zhifei			SinoPharm (Beijing)			SinoVac
	[-3, 3]	[-4, 4]	[-5, 5]	[-3, 3]	[-4, 4]	[-5, 5]	[-3, 3]	[-4, 4]	[-5, 5]	[-3, 3]	[-4, 4]	[-5, 5]
Panel A (Market Model):
CAAR	0.015	0.050	0.074	0.028	0.040	0.029	0.021	0.020	0.018	0.035	0.050	0.058
	***	***	***	***	***	***	***	***	***	***	***	***
	(0.001)	(0.002)	(0.002)	(0.001)	(0.002)	(0.002)	(0.002)	(0.000)	(0.000)	(0.002)	(0.002)	(0.002)
Panel B (Constant Mean Return Model):
CAAR	0.019	0.032	0.062	0.033	0.037	0.031	0.009	0.016	0.014	0.010	0.015	0.023
	***	***	***	***	***	***	***	***	***	***	***	***
	(0.001)	(0.002)	(0.002)	(0.001)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.001)	(0.002)	(0.002)
N	3867			3893			3954			3982

Standard errors are in parentheses. *** , ** , * .

We divide firms into different sectors using the China Securities Regulatory Commission (CSRC) Industry Classification (2012 Edition). Table 2 presents the sectoral CAARs over the three event windows. In the table, MR is the mean return of the sector (on the left of the CAARs), and is the number of firms in the sector (on the left of the stars).7

Table 2

Sectoral cumulative average abnormal returns.

Window	MR/N	CanSino&SP (Wuhan)			Zhifei			SinoPharm (Beijing)			SinoVac
		[-3, 3]	[-4, 4]	[-5, 5]	[-3, 3]	[-4, 4]	[-5, 5]	[-3, 3]	[-4, 4]	[-5, 5]	[-3, 3]	[-4, 4]	[-5, 5]
Agriculture	0.0036	−0.010	0.033	0.072	0.007	0.029	0.010	0.057	0.031	0.017	0.052	0.063	0.066
	42		**	***		*		***			***	***	***
Mining	0.0033	−0.023	0.018	0.066	−0.028	−0.027	−0.034	0.121	0.099	0.084	0.047	0.063	0.068
	74	**	*	***	***	**	***	***	***	***	***	***	***
Manufacturing	0.0028	0.010	0.042	0.063	0.023	0.037	0.018	0.019	0.019	0.015	0.046	0.063	0.072
	2749	***	***	***	***	***	***	***	***	***	***	***	***
Electric&Heating	0.0034	0.028	0.058	0.086	0.089	0.118	0.125	0.033	0.024	0.015	−0.003	0.004	0.002
	1281	***	***	***	***	***	***	***	*
Construction	0.0023	0.038	0.081	0.122	0.047	0.069	0.049	0.008	0.013	0.008	0.002	0.011	0.015
	99	***	***	***	***	***	***
Wholesale&Retail	0.0025	0.017	0.043	0.076	0.033	0.049	0.034	0.029	0.031	0.033	0.019	0.029	0.032
	173	***	***	***	***	***	***	***	***	***	**	***	***
Transportation	0.0012	0.001	0.022	0.053	0.021	0.035	0.004	0.014	0.014	0.012	0.003	0.007	0.015
	111		***	***	***	***		**	*				*
Hotels&Catering	0.0016	0.007	−0.001	0.005	0.053	0.068	0.044	−0.073	−0.052	−0.058	−0.014	−0.003	−0.008
	8		***	***	***	***		**
Info. Technology	0.0023	0.033	0.038	0.089	0.018	0.051	0.032	0.029	0.029	0.018	0.023	0.047	0.058
	343	***	***	***	***	***	***	***	***	**	***	***	***
Real Estate	0.0016	0.022	0.042	0.072	0.041	0.047	0.034	0.013	0.026	0.029	0.005	0.007	0.015
	114	***	***	***	***	***	***	*	***	***
Business Service	0.0018	0.028	0.075	0.103	0.050	0.063	0.045	0.035	0.048	0.027	−0.001	0.008	0.007
	58	***	***	***	***	***	***	***	***	*			*
Scientific Research	0.0035	0.000	0.025	0.039	0.038	0.048	0.041	0.007	0.008	0.010	0.016	0.022	0.027
	66			*	***	***	**
Public Mgmt	0.0051	0.097	0.133	0.166	0.105	0.111	0.109	−0.029	−0.032	−0.029	−0.007	−0.004	−0.001
	78	***	***	***	***	***	***	*
Residential Service	0.0092	0.119	0.167	0.261	0.070	0.113	0.073	0.111	0.135	0.215	0.184	0.140	0.089
	1			*							*
Education	0.0002	0.010	0.056	0.097	0.044	0.059	0.042	0.003	0.003	−0.013	0.016	0.034	0.047
	12		**	***	*	**
Health	0.0023	−0.110	−0.101	−0.137	0.014	0.025	0.023	0.084	0.091	0.122	0.058	0.053	0.057
	10	***	***	***				**	**	**
Sport&Entertain.	0.0016	0.009	0.039	0.064	0.043	0.059	0.046	0.038	0.040	0.040	−0.037	−0.037	−0.024
	56		***	***	***	***	***	***	***	***	***	***	**
Conglomerates	0.0026	0.032	0.088	0.114	0.047	0.052	0.031	0.000	−0.004	−0.007	0.044	0.066	0.084
	10	**	***	***	**	**					*	**	**

*** , ** , * . The mean return of the sector is on the left of the CAARs, and the number of firms in the sector is on the left of the stars.

Overall, the short-term reactions to the events are positive for most sectors.8 By contrast, Wang et al. (2021) find that the short-term reactions to the Wuhan lockdown on January 23, 2020 are negative for the majority of sectors. Column 3 shows that sectors such as construction, business service, and public management were strongly boosted by the CanSino and SinoPharm (Wuhan) vaccine approval announcements, reporting CAARs of 12.2%, 10.3%, and 16.6, respectively. While we see mixed results for some sectors, results for manufacturing, wholesale&retail, and information technology are persistently significant and positive across events and event windows. Another observation from Table 2 is that there are more significant results in columns 1 to 6 than in columns 7 to 12. This indicates that stocks in some sectors no longer respond to this type of news as time passes. For example, stocks in sectors such as transportation and hotels&catering that were hit hard by the pandemic basically only react positively to the first two events. One possible explanation is that there was no indication that these sectors are going to recover in a short period even more good news on vaccines is coming.

Table 3 reports the summary statistics. The means of CARs are relatively small compared to their standard deviations (for example, the mean of CAR[-5, 5] is 4.4%, and the standard deviation of CAR[-5, 5] is 11%). This implies there might be substantial variations in CARs across firms. We report the Pearson correlation matrix in Table 4. The coefficients of the pairwise correlations among the explanatory variables are low in general.

Table 3

Summary statistics.

Variable	Observation	Mean	Std. Dev.	Min	Max
CAR[-3, 3]	13205	0.024	0.081	−0.48	0.644
CAR[-4, 4]	13205	0.04	0.095	−0.506	0.79
CAR[-5, 5]	13205	0.044	0.11	−0.688	0.892
ROA	13205	0.024	0.025	−0.006	0.566
Tan	13205	0.928	0.088	0.061	1
Age	13205	11.297	8.433	1	31
Lev	13205	0.147	0.423	−3.996	4.687
Size	13205	22.206	1.402	18.276	28.636

Table 4

Pearson correlation matrix.

	CAR[-3, 3]	CAR[-4, 4]	CAR[-5, 5]	ROA	Tan	Age	Lev	Size
CAR[-3, 3]	1
CAR[-4, 4]	0.904***	1
CAR[-5, 5]	0.826***	0.927***	1
ROA	−0.015*	−0.047***	−0.043***	1
Tan	−0.019**	−0.027***	−0.033***	0.053***	1
Age	−0.007	−0.009	−0.010	−0.147***	−0.051***	1
Lev	0.023***	0.023***	0.028***	−0.235***	−0.071***	0.174***	1
Size	−0.087***	−0.099***	−0.095***	−0.104***	−0.063***	0.497***	0.195***	1

*** , ** , * .

Table 5 reports the results of the regression analysis. In regressions (1), (4), and (7), we only control for time-fixed effects; in regressions (2), (5), and (8), we also control for province-fixed effects; in regressions (3), (6), and (9), we further control for industry-fixed effects. It appears that our results are quite robust across event windows and specifications: there are only minor differences in the results, as well as their significances.

Table 5

Regression results.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
	CAR[-3,3]	CAR[-3,3]	CAR[-3,3]	CAR[-4, 4]	CAR[-4, 4]	CAR[-4, 4]	CAR[-5, 5]	CAR[-5, 5]	CAR[-5, 5]
Intercept	0.1898***	0.1929***	0.2083***	0.2648***	0.2687***	0.2854***	0.3023***	0.3082***	0.3282***
	(0.015)	(0.015)	(0.016)	(0.017)	(0.017)	(0.019)	(0.020)	(0.020)	(0.022)
ROA	−0.1282***	−0.1260***	−0.1154***	−0.1587***	−0.1521***	−0.1324***	−0.1535***	−0.1420***	−0.1126**
	(0.041)	(0.042)	(0.042)	(0.043)	(0.043)	(0.044)	(0.047)	(0.047)	(0.047)
Tan	−0.0186**	−0.0176**	−0.0202**	−0.0311***	−0.0298***	−0.0350***	−0.0427***	−0.0418***	−0.0457***
	(0.008)	(0.008)	(0.008)	(0.009)	(0.009)	(0.010)	(0.010)	(0.011)	(0.011)
Age	0.0004***	0.0004***	0.0003***	0.0005***	0.0005***	0.0004***	0.0005***	0.0005***	0.0004***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Lev	0.0051***	0.0049***	0.0039**	0.0067***	0.0065***	0.0056***	0.0090***	0.0087***	0.0080***
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
Size	−0.0068***	−0.0069***	−0.0075***	−0.0090***	−0.0092***	−0.0097***	−0.0100***	−0.0103***	−0.0110***
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
FE
Time	YES	YES	YES	YES	YES	YES	YES	YES	YES
Province		YES	YES		YES	YES		YES	YES
Industry			YES			YES			YES
N	13205	13205	13205	13205	13205	13205	13205	13205	13205
R^2	0.0266	0.0278	0.0306	0.0324	0.0340	0.0369	0.0519	0.0534	0.0573

Standard errors are in parentheses. *** , ** , * .

*** , ** , * .

ROA, Tan, and Size are negatively correlated to CARs, while Age and Lev are positively correlated to CARs. In other words, firms that have high CARs have poor performance, low tangible assets ratios, and high financial leverages and are small and old. Without loss of generality, we take a closer look at the sixth column. The coefficient of ROA is −0.1324, and it is statistically significant at the 1% level. This implies a one standard deviation increase in ROA (0.025) from its mean reduces CAR[-4, 4] by 0.331%. Similarly, the coefficient of Size is −0.0097, and a one standard deviation increase in Size (1.402) from its mean reduces CAR[-4, 4] by 1.356%. By contrast, the coefficient of Age is 0.0004, and a one standard deviation increase in Age (8.433) from its mean increases CAR[-4, 4] by 0.337%.

Conclusion

This paper extends the Covid-19 literature in economics and finance. We explored the impacts of Covid-19 vaccine announcements on the Chinese stock market. We find that these announcements positively impacted stock prices in general. We also find that stocks in different sectors react differently to the announcements. Interestingly, firms with poorer performance, smaller sizes, and greater ages might benefit more from this type of positive public health announcement.

Date	Vaccine	Approval
February 25, 2021 (T_1)	CanSino, SinoPharm (Wuhan)	China
March 17, 2021 (T_2)	Zhifei	China
May 07, 2021 (T_3)	SinoPharm (Beijing)	WHO
June 01, 2021 (T_4)	SinoVac	WHO