COVID-19 Government restriction policy, COVID-19 vaccination and stock markets: Evidence from a global perspective

Abstract

We use the COVID-19 stringency index to investigate the relationship among COVID-19 government restriction policy, COVID-19 vaccination and stock markets. We find that the impact of the change rate of COVID-19 stringency index on stock returns turns from significant in the pre-vaccination period to insignificant in the post-vaccination period. Bad news from COVID-19 restriction policy cause more stock volatilities than good news. The advent of COVID-19 vaccination weakens the linkage of COVID-19 stringency index and stock market, while COVID-19 stringency index only plays a partially mediate role in the correlation between COVID-19 cumulative vaccination rate and stock market performance.

1. Introduction

Since the outbreak of COVID-19 pandemic in early 2020, policymakers around the world have imposed more or less “lockdown” policies in response to the public health event. Some scholars concern whether these policies that restrict people's movements are legal from a legal and ethical point of view (e.g., Florey,2020; Thuong,2021; Kaldunski, 2022), or how necessary and effective they are in curbing the spread of COVID-19 pandemic from a medical point of view (e.g., Syafarina et al., 2021; Wang & Pagán, 2021; Susilo et al., 2021). More scholars are more interesting in the impact of these restriction policies on social and economic life in hindsight, such as climate (e.g., Akritidis et al., 2021), work and travel patterns (e.g., Hiselius & Arnfalk, 2021), consumption and sales patterns (e.g., Huang & Gou, 2022), unemployment rates (e.g., Ployhart et al., 2021), special crowd (e.g., Stolz et al., 2021), audit quality (e.g., Harjoto & Laksmana, 2022), etc.

In this study, we investigate the impact of COVID-19 government restriction policy on stock markets and the role of COVID-19 vaccination from a global perspective, covering 50 countries from 31 January 2020 to 30 September 2022. Our findings show that COVID-19 government restriction policy has impact on stock market, and this linkage is time-varying. The bad news from COVID-19 government restriction policy causes more volatilities in stock markets than good news. The advent of COVID-19 vaccination leads changes in the relationship. Besides, there is evidence that the COVID-19 vaccination rate has significant impact on stock markets, while COVID-19 restriction policy plays a partially mediate role.

Our study has some contributions as follows. First of all, unlike the literature that purely studies the relationship between vaccinations and the stock market (e.g., Apergis et al., 2022; Behera et al., 2022), or purely studies the impact of COVID-19 government restrictions on stock market (e.g., Anh & Gan, 2020; Narayan et al., 2021; Caporale et al., 2022; Kizys et al., 2021; Scarcella et al., 2022; Xie & Zhou, 2022), we systematically examine the impact of COVID-19 government restrictions on stock market and the role played by COVID-19 vaccination. We not only explore the role of the advent of COVID-19 vaccination, but also examine whether COVID-19 restriction policy can play a mediate path in the impact of COVID-19 vaccination rate on stock markets. Second, different from the mentioned previous studies above, our study provides evidence from a global perspective by using a larger sample. Third, different from Abdullah et al (2022), Mishra et al (2022) and Scherf et al (2022) which also use the COVID-19 stringency index proposed by the University of Oxford to evaluate government restriction policy, we depart its total volatility into good and bad ones to investigate its impact on stock markets.

The remainder of this paper is as follows. Section 2 is literature review and hypothesis development, Section 3 introduces the methodology, Section 4 describes sample data, Section 5 details the empirical results, and Section 6 concludes the study.

2. Literature review and hypothesis development

Since the stock market is a barometer of national economy, taking the stock market as the research object can well reflect the impact of COVID-19 pandemic on the economy. There has been a string of studies on what happens to stock markets during COVID-19 pandemic. Throughout the existing literature, a small part of studies pay attention on the impact of COVID-19 restriction policies on stock market in given countries, such as the United States (Martins & Cró, 2022; Mishra et al., 2022), Vietnam (Anh & Gan, 2020), Malaysia (Xie & Zhou, 2022), G20 countries (Caporale et al., 2022), G7 countries (Narayan et al., 2021). Using different methods and different empirical sample, different findings are generated on the relationship between COVID-19 restriction policy. Some scholars argue that stock returns are negatively affected by government restrictions (Abdullah et al.,2022; Aharon & Siev,2021; Kheni & Kumar,2021; Caporale et al.,2022), but some other scholars argue that stock markets react positively to government lockdown policies (Martins & Cró, 2022; Bouri et al.,2022; Wang et al.,2021; Narayan et al., 2021; Xie & Zhou (2022). However, Abdullah et al (2022) find that the long-term and short-term negative effect is asymmetric, Anh & Gan (2020) find than Vietnam's stock returns react negatively COVID-19 pre-lock down while positively during the lockdown period. Narayan et al. (2021) find that lockdowns have a positive effect on the G7 stock markets, while Caporale et al. (2022) find that the stock markets of G7 countries are affected negatively by government restrictions. Scherf et al (2022) also find that COVID-19 restrictions lead to different reactions OECD and BRICS countries. Ashraf (2020) argue that government restrictions have a positive effect on stock market returns due to the reduction in COVID-19 confirmed cases, while a negative effect due to their adverse effect on economic activity.

Summarized from previous studies above, we can see that COVID-19 government restriction policy has impact on stock markets, but the direction of the effect (positive or negative) has not been verified uniform yet. Thus, based on the previous literature review above, we propose the first hypothesis in our study as follows.

• H1.COVID-19 government restriction policy has significant impact on stock market.

To test this hypothesis, four specifical sub-hypothesis are constructed as below:

• H1a.The change rate of COVID-19 government restriction policy have significant impact on stock return.

• H1b.The total volatility of COVID-19 government restriction policy has significant impact on the total volatility of stock market.

• H1c.The bad volatility of COVID-19 government restriction policy has significant impact on the bad volatility of stock market.

• H1d.The good volatility of COVID-19 government restriction policy has significant impact on the good volatility of stock market.

With the advent of massive COVID-19 vaccination, there also has been a small amount of literature on the impact of COVID-19 vaccination on stock markets (e.g., Demir et al., 2021; Apergis et al., 2022; Mishra et al., 2022), or how to relax restrictions at the pace of vaccination (e.g., Bauer et al., 2021). Yu et al (2022) find that the dynamic linkage between COVID-19 pandemic and stock volatility turns weak after the announcement of COVID-19 mRNA-based vaccine. Mishra et al (2022) suggest that, in the long run, COVID-19 government restriction policy has a negative impact on stock market, whereas vaccinations have a positive effect on it. Hong et al. (2021) argue that the quick arrival of vaccines lowers the impact of COVID-19 pandemic shocks. Based on these previous, we propose the second group of research hypotheses as below:

H2. The advent of COVID-19 vaccination moderates the relationship between COVID-19 government restriction policy and stock market.

H2a. The advent of COVID-19 vaccination moderates the relationship between the change rate of COVID-19 government restriction policy stock return.

H2b. The advent of COVID-19 vaccination moderates the relationship between the total volatility of COVID-19 government restriction policy and the total volatility of stock market.

H2c. The advent of COVID-19 vaccination moderates the relationship between the bad volatility of COVID-19 government restriction policy and the bad volatility of stock market.

H2d. The advent of COVID-19 vaccination moderates the relationship between the good volatility of COVID-19 government restriction policy and the good volatility of stock market .

In previous literature, Mishra et al (2022) suggest that, in the long run, COVID-19 government restriction policy has a negative impact on stock market, whereas vaccinations have a positive effect on it. Demir et al (2021) argue that vaccination programs decrease the volatility of energy stocks. Apergis et al (2022) find that the vaccination program in Canada reversed the negative impact of COVID-19 pandemic on stock returns and the positive impact on stock volatility. Similarly, Cong Nguyen To et al. (2021) and Rouatbi et al. (2021) also document that COVID-19 vaccination has a negative impact on stock market volatility, helping to stabilize the financial markets. It can be seen that COVID-19 vaccination has impact on stock markets. However, except Demir et al (2021), other previous literature above does not discuss the path by which COVID-19 vaccination influences stock market. Thus, we further propose the third group of hypotheses to investigate whether COVID-19 restriction policy can play a mediate role in the COVID-19 vaccination and stock markets as follows:

H3. The COVID-19 cumulative vaccination rate has impact on stock market, and COVID-19 restriction policy plays a mediate role in this effect.

H3a. The COVID-19 cumulative vaccination rate has impact on stock return, and the change rate of COVID-19 restriction policy plays a mediate role.

H3b. The COVID-19 cumulative vaccination rate has impact on stock market's total volatility, and the total volatility of COVID-19 restriction policy plays a mediate role

H3c. The COVID-19 cumulative vaccination rate has impact on stock market's bad volatility, and the bad volatility of COVID-19 restriction policy plays a mediate role

H3d. The COVID-19 cumulative vaccination rate has impact on stock market's good volatility, and the good volatility of COVID-19 restriction policy plays a mediate role

3. Methodology

3.1. COVID-19 stringency index

In this study, we use the COVID-19 Stringency Index launched by the University of Oxford to measure the strictness of ‘lockdown style’ policies that governments have taken to tackle COVID-19, which is calculated as below:

$$R{I}_{i,t}=\frac{1}{9}\left(S{C}_{i,t}+W{C}_{i,t}+CP{E}_{i,t}+RO{G}_{i,t}+P{T}_{i,t}+SAH{O}_{i,t}+ROI{M}_{i,t}+IT{C}_{i,t}+PI{C}_{i,t}\right)$$ (1)

where RI _i,t means the total score of COVID-19 Stringency Index for country i at time t, equaling to the average score of the nine subindices, namely, School Closures (SC _i,t), Workplace closing (WC _i,t), Cancel Public Events (CPE _i,t), Restrictions on gatherings (ROG _i,t),Public Transportation (PT _i,t),Stay at Home Order (SAHO _i,t), Restrictions on Internal Movement (ROIM _i,t), International Travel Controls (ITC _i,t) and Public Information Campaigns (PIC _i,t)1 ^, 2 ^, 3

3.2. Mean-GARCH (1,1) model

We use a GARCH (1,1) model with t distribution to estimate the total conditional volatility of stock market and COVID-19 stringency index for each country as below:

$$\text{Rlog}{\mathrm{R}}_{i,t}=100\times \left[\ln (1+R{I}_{i,t})-\ln (1+R{I}_{i,t-1})\right]$$ (2)

$$Slog{R}_{i,t}=100\times \left[\ln \left(S{P}_{i,t}\right)-\ln \left(S{P}_{i,t-1}\right)\right]$$ (3)

$$\{\begin{array}{c}{r}_{i,t}=\mu +{\epsilon }_{i,t},t=1,2,...,T\hfill \\ {\epsilon }_{i,t}={\sigma }_{i,t}{e}_{i,t},e_{i,t}\sim t(0,1,\upsilon )\hfill \\ {\sigma }_{i,t}^2=\omega +\alpha {\epsilon }^2{{}_{i,}}_{t-1}+\beta {\sigma }^2{{}_{i,}}_{t-1}\hfill \end{array}$$ (4)

where r _i,t = [SlogR_i,t, RlogR_i,t]′, RlogR_i,tmeans the change rate of RI _i,t, andSlogR_i,tmeans the log stock returns for country i at time t. SP _i,t and SP _i,t − 1 respectively mean stock price for country i at time t and t-1. $\upsilon$ is the parameter for t-distribution. μ,ω,αandβ are parameters of the GARCH (1,1) model, where ω ≥ 0, α > 0, and β > 0 (Bollerslev, 1986).

The likelihood function of the model parameters can be written as:

$$L\left(\theta |\epsilon \right)\propto \prod _{t=1}^T\frac{\Gamma \left(\frac{\upsilon +1}{2}\right)}{\Gamma \left(\frac{\upsilon }{2}\right)}{\left(\upsilon {{\sigma }_t}^2\right)}^{-\frac{1}{2}}{\left[1+\frac{{\epsilon }_t}{\upsilon {{\sigma }_t}^2}\right]}^{-\frac{\upsilon +1}{2}}$$ (5)

The total conditional volatility is calculated as:

$$T{V}_{i,t}=\frac{\upsilon }{\upsilon -2}{{\sigma }^2}_{i,t}$$ (6)

where TV _i,t = [STV _i,t, RTV _i,t]′, represents the total conditional volatilities of SlogR_i,t and RlogR_i,t respectively.

3.3. Good volatility and bad volatility

Similar to Feunou & Okou (2019), Hussain (2020) and Lyu et al. (2021), we depart the total volatility into good and bad ones as below:

$$\mathit{SG}{V}_{i,t}=\{\begin{array}{cc}\mathit{ST}{V}_{i,t},& \text{if}\text{Slog}{\mathrm{R}}_{i,t}\ge 0\\ 0,& \text{else}\end{array}$$ (7)

$$\mathit{SB}{V}_{i,t}=\{\begin{array}{cc}\mathit{ST}{V}_{i,t},& \text{if}\text{Slog}{\mathrm{R}}_{i,t}\le 0\\ 0,& \text{else}\end{array}$$ (8)

$$\mathit{RG}{V}_{i,t}=\{\begin{array}{cc}\mathit{RT}{V}_{i,t},& \text{if}\text{Rlog}{\mathrm{R}}_{i,t}\le 0\\ 0,& \text{else}\end{array}$$ (9)

$$\mathit{RB}{V}_{i,t}=\{\begin{array}{cc}\mathit{RT}{V}_{i,t},& \text{if}\text{Rlog}{\mathrm{R}}_{i,t}\ge 0\\ 0,& \mathit{else}\end{array}$$ (10)

where SGV _i,t and SBV _i,tare respectively good and bad volatilities of SlogR_i,t, RGV _i,t and RBV _i,tare respectively good and bad volatilities of RlogR_i,t for country i at time t.

3.4. Empirical model design

First, a binary OLS regression model is specified as the baseline model to estimate the relationship between COVID-19 stringency indexes and stock markets as below:

$$\text{STOC}{\mathrm{K}}_{i,t}=a_1+a_2\text{COVI}{\mathrm{D}}_{i,t}+{\delta }_{i,t}$$ (11)

where STOCK_i,t = [SlogR_i,t, STV _i,t, SGV _i,t, SBV _i,t]′,and COVID_i,t = [RlogR_i,t, RTV _i,t, RGV _i,t, RBV _i,t]′. Our first hypothesis H1 and its four sub-hypotheses (H1a, H1b, H1c, and H1d) are tested by this model. All the total sample and subsamples grouped by pre-and post-vaccination are considered.

Second, we construct fixed effects models to test the role of the COVID-19 vaccination, namely the second hypothesis H2 and its four sub-hypotheses (H2a, H2b, H2c, and H2d) in this study. They are specified as follows:

$$\text{STOC}{\mathrm{K}}_{i,t}=a_1+a_2\text{COVI}{\mathrm{D}}_{i,t}+\sum Time+\sum Country+{\delta }_{i,t}$$ (12)

$$\text{STOC}{\mathrm{K}}_{i,t}=a_1+a_2\text{COVI}{\mathrm{D}}_{i,t}+a_3\text{Vac}\_\text{du}{\mathrm{m}}_{i,t}+\sum Time+\sum Country+{\delta }_{i,t}$$ (13)

$$\text{STOC}{\mathrm{K}}_{i,t}=a_1+a_2\text{COVI}{\mathrm{D}}_{i,t}+a_3\text{Vac}\_\text{du}{\mathrm{m}}_{i,t}+a_4\text{COVI}{\mathrm{D}}_{i,t}\times \text{Vac}\_\text{du}{\mathrm{m}}_{i,t}+\sum Time+\sum Country+{\delta }_{i,t}$$ (14)

where $\text{Vac}\_\text{du}{\mathrm{m}}_{i,t}$ is a dummy variable. $\text{Vac}\_\text{du}{\mathrm{m}}_{i,t}$ is 1 if the observation is in the post-vaccination period, otherwise, it is 0. $\text{COVI}{\mathrm{D}}_{i,t}\times \text{Vac}\_\text{du}{\mathrm{m}}_{i,t}$ is the multiplicative interaction.∑Time is the fixed effect of time and ∑Countryis the fixed effect of country.

Third, we investigate the impact of COVID-19 vaccination rate on stock markets and the moderate role of stringency index. Namely, our third hypothesis H3 and its four sub-hypotheses (H3a, H3b, H3c, and H3d) are tested. The regressions are specified as below:

$$\text{COVI}{\mathrm{D}}_{i,t}={\mathrm{c}}_1+c_2\text{AL}{\mathrm{O}}_{i,t}+\sum Time+\sum Country+{\vartheta }_{i,t}$$ (15)

$$\text{STOC}{\mathrm{K}}_{i,t}=d_1+d_2\text{AL}{\mathrm{O}}_{i,t}+\sum Time+\sum Country+{\vartheta }_{i,t}$$ (16)

$$\text{STOC}{\mathrm{K}}_{i,t}=d_1+d_2\text{AL}{\mathrm{O}}_{i,t}+d_3\text{COVI}{\mathrm{D}}_{i,t}+\sum Time+\sum Country+{\vartheta }_{i,t}$$ (17)

$$\text{STOC}{\mathrm{K}}_{i,t}=d_1+d_2\text{AL}{\mathrm{O}}_{i,t}+d_3\text{COVI}{\mathrm{D}}_{i,t}+d_4\text{AL}{\mathrm{O}}_{i,t}\times \text{COVI}{\mathrm{D}}_{i,t}+\sum Time+\sum Country+{\vartheta }_{i,t}$$ (18)

where ALO_i,tis the cumulative COVID-19 vaccination rate of country i at time t. ALO_i,t × COVID_i,t is the multiplicative interaction.

4. Data and basic statistics

4.1. Sample selection and data source

We collect daily stock price data and population data from the CEIC database, obtain daily data of the cumulative number of people of COVID-19 vaccination (at least one dose) and COVID-19 Government Stringency Index data from the Wind database, spanning the period from 31 January 2020 to 30 September 2022.4

We select all the countries that have both stock data and COVID-19 vaccination data, and then delete those countries of which the average number of missing values of stock data per week is more than one or the total number of missing values for the COVID-19 vaccination data is greater than 290. We use interpolation to fill in the missing observation of stock data. For COVID-19 stringency index data, we fill in the missing values before the first nonzero value with 0 and fill the null after the first nonzero value with the previous nonzero value. Eventually, there are 50 countries, totally 34,800 observations in our final sample. The details of the selected countries and stock indexes are summarized in Table 1 .

Table 1.

List of the selected countries and stock index series.

Country	ISO3	Stock index name
Argentina	ARG	Buenos Aires Stock Exchange (BCBA): Index: Merval
Australia	AUS	Australian Securities Exchange (ASX): Index: All Ordinaries
Austria	AUT	Vienna Stock Exchange (VSE): Index: ATX
Belgium	BEL	Euronext Brussels: Index: BEL 20
Brazil	BRA	B3 S.A.: Index: BOVESPA
Cambodia	KHM	Cambodia Securities Exchange (CSX): Composite Index
Canada	CAN	TMX Group Limited (S&P/TSX): Composite Index
Chile	CHL	Santiago Stock Exchange (BCS): Index: S&P/CLX IPSA
Colombia	COL	Colombian Securities Exchange (CSE): Index: COLCAP
Croatia	HRV	Zagreb Stock Exchange (ZSE): Equity Index: CROBEX
Czech Republic	CZE	Prague Stock Exchange (PSE): Index: PX
Denmark	DNK	Nasdaq Copenhagen (NC): Index: OMX Copenhagen
Ecuador	ECU	Quito Stock Exchange (QSE): New: Index: ECUINDEX
Estonia	EST	Nasdaq Tallinn (NTSE): Index: OMX Baltic Benchmark GI
Finland	FIN	Nasdaq Helsinki (HSE): Index: OMX Helsinki
France	FRA	Euronext Paris: Index: CAC40
Germany	DEU	Deutsche Börse Group (Germany): Index: DAX
Greece	GRC	Athens Stock Exchange (ASE): Index: Athex Composite Share Price
Hungary	HUN	Budapest Stock Exchange (BSE): Equity Index: Domestic: BUX
India	IND	Bombay Stock Exchange (BSE): Index: SENSEX
Indonesia	IDN	Indonesia Stock Exchange (IDX): Index: Composite
Ireland	IRL	Euronext Dublin: ISEQ Equity Index: All Share
Japan	JPN	Japan Exchange Group Inc. (TSE): Index: TOPIX
Kazakhstan	KAZ	Kazakhstan Stock Exchange (KASE): Index
Latvia	LVA	Nasdaq Riga: Index: OMX Riga
Lithuania	LTU	Nasdaq Vilnius (NV): Index: OMX Baltic Benchmark GI
Luxembourg	LUX	Luxembourg Stock Exchange (LSE): Index: LuxX
Malaysia	MYS	Bursa Malaysia: FTSE Bursa Malaysia Index: Composite
Malta	MLT	Malta Stock Exchange: Share Index
Mexico	MEX	Mexican Stock Exchange: Index: INMEX
Mongolia	MNG	Mongolian Stock Exchange: Index: Top 20
Pakistan	PAK	Pakistan Stock Exchange Limited (PSX): Index: KSE All Shares
Peru	PER	Lima Stock Exchange (LSE): Index: S&P/BVL: General
Poland	POL	Warsaw Stock Exchange (WSE): Index: WIG
Portugal	PRT	Euronext Lisbon: Index: PSI General
Romania	ROU	Bucharest Stock Exchange (BSE): RON: Index: BET
Russian Federation	RUS	Moscow Exchange: US Dollar Denominated Indices: RTS Index
Singapore	SGP	Exchange Data International Limited: Index: FTSE Strait Times
Slovakia	SVK	Bratislava Stock Exchange (BSSE): Index: SAX
Slovenia	SVN	Ljubljana Stock Exchange (LJSE): Index: SBITOP
South Korea	KOR	Korea Exchange (KRX): Index: Korea Composite
Spain	ESP	Madrid Stock Exchange: Index: General: IGBM
Sri Lanka	LKA	Colombo Stock Exchange (CSE): Index: All Shares
Switzerland	CHE	SIX Swiss Exchange (SIX SE): Equity Index: All Swiss Shares
Tunisia	TUN	Tunis Stock Exchange (TSE): Index: TUNINDEX
Turkey	TCA	Borsa Istanbul (BIST): NPI: All Share
United Arab Emirates	ARE	Dubai Financial Market (DFM): Index
United Kingdom	GBR	Exchange Data International Limited: Index: FTSE All Share
United States	USA	S&P Globa: Index: S&P 500
Vietnam	VNM	Hanoi Stock Exchange (HSE): Index: HNX

Notes:This table reports details on stock market index chosen for each country in our study. Country names are displayed in the first column. The second column shows the corresponding the ISO Alpha-3 country code. The last column shows the selected stock index for each country, obtained from the CEIC database (https://insights.ceicdata.com).5^,6

4.2. Descriptive statistics and basic tests

As seen in Fig. 1, Fig. 2 , the global stock markets and COVID-19 stringency indexes experienced huge volatilities after the announcement of COVID-19 pandemic in March 2020. The volatility of global stock markets also experienced a peak when the Russia-Ukraine war breakout in February 2022.There are a common tendency for the COVID-19 stringency indexes to rise and then fall, and a common tendency for stock prices to fall and then rise before the Russia-Ukraine war.

Fig. 1.

Time-varying trends of stock market, COVID-19 restriction index and vaccination rate in different countries.

(Notes: This figure depicts the evolution of COVID-19 restriction index, COVID-19 vaccination rate and stock index in each country during 31 January 2020 to 30 September 2022. There are 50 subgraphs in this figure, respectively corresponding to the 50 selected countries in this study. See Table 1 for details of the ISO Alpha-3 country code of given countries. For each country's subgraph, the upper part depicts the daily COVID-19 restriction index (RI, black line), the daily log change rate of COVID-19 restriction index (RlogR, orange line) and the daily COVID-19 cumulative vaccination rate (ALO, green line), in which the right y-axis is for RlogR, while the left y-axis is for RI and ALO. ALO is calculated as the cumulative number of people who have received at least one dose of vaccine divided by the total population. For each country's subgraph, the lower part depicts daily close prices of stock index (SP, blue line) and daily stock log returns (SlogR, red line), in which the left y-axis is for SP and the right y-axis is for SlogR. There are four purple dashed vertical lines in each subgraph for specific dates. Specifically, the first purple dashed vertical line marks the date when WHO announced the COVID-19 pandemic on 11 March 2020, the second one marks the date when Pfizer Inc. and BioNTech SE announced the results of the Phase 3 clinical trial of their mRNA-based COVID-19 vaccine with an efficacy rate of 95% on 18 November 2020, the third one marks the first vaccination date of each countries (the first non-zero data for ALO in this study), and the last purple dashed vertical line marks the date when Russia launched military attack against Ukraine on 24 February 2022 . Descriptive statistics of stock market, COVID-19 restriction stringency index and COVID-19 vaccination for each selected country are shown in Appendix A1.

Fig. 2.

Average trends of stock market, COVID-19 restriction index and cumulative vaccination rate.

(Notes: In the upper subgraph, the red line depicts daily mean values of stock index prices (SP) of the 50 countries, the blue line depicts daily mean values of COVID-19 indexes (RI) of the 50 countries and the black line depicts daily mean values of COVID-19 cumulative vaccination rate (ALO) of the 50 countries. In the lower subgraph, the green line depicts daily mean values of the total conditional volatilities of stock markets and the orange line depicts daily mean values of the total conditional volatilities of COVID-19 restriction indexes of the 50 countries.)

For space limitation, we put the results of basic tests for RlogR and SlogR of each country in Appendix A2, which shows that the GARCH–type model with t distribution is suitable for capturing the conditional volatilities. And, the estimation results of the mean-GARCH (1,1) models are reported in Appendix A3.

Table 2 report descriptive statistics for all the variables in our final sample. The mean value of SP in the pre-vaccination period is smaller than that in the post-vaccination period, while the mean value of RI in the pre-vaccination period is larger than that in the post-vaccination period. The mean value of SlogR is negative in the pre-vaccination period while positive in the post-vaccination period. The mean value of RlogR is positive in the pre-vaccination period while negative in the post-vaccination period. All the mean values of STV, SBV, SGV, RTV, RBV and RGV in the post-vaccination period are lower than those in the pre-vaccination period. And, there are more bad volatilities than good volatilities in RlogR before COVID-19 vaccination, while much more good volatilities than bad volatilities after COVID-19 vaccination. Those comparing results suggest the overall downward trend of COVID-19 stringency indexes and the overall upward trend of stock prices, which is consistent with Figs. 1 and 2.

Table 2.

Descriptive statistics foe variables in the total sample and the two subsamples grouped by pre-and post-vaccination.

Variable	Obs.	Mean	SD	Min	p50	p25	p75	Max
Panel A: Total sample (31/01/2020–30/09/2022)
SP	34,800	11,252	21,011	92.640	3209	1322	8174	150,300
RI	34,800	48.450	22.770	0.000	49.070	33.330	67.130	100.000
SlogR	34,750	0.010	1.449	-48.290	0.043	-0.518	0.631	23.200
RlogR	34,750	0.281	11.100	-371.800	0.000	0.000	0.000	423.500
STV	34,750	1.249	0.812	0.286	1.084	0.787	1.435	27.550
SBV	34,750	0.590	0.837	0.000	0.000	0.000	1.041	25.760
SGV	34,750	0.671	0.859	0.000	0.581	0.000	1.127	27.550
RTV	34,750	29.830	116.500	0.001	0.964	0.211	3.161	702.500
RBV	34,750	22.770	102.100	0.000	0.610	0.046	2.497	702.500
RGV	34,750	28.840	114.700	0.000	0.820	0.202	2.970	702.500
ALO	23,950	51.710	31.830	0.000	60.890	20.240	78.040	108.300
Variable	Obs.	Mean	SD	Min	p50	p25	p75	Max
Panel B: Pre-vaccination (31/01/2020–30/11/2020)
SP	10,850	9068	16,821	92.640	2540	1078	6774	116,700
RI	10,850	54.910	24.890	0.000	58.330	40.740	73.610	100.000
SlogR	10,800	-0.011	1.880	-16.340	0.035	-0.590	0.754	13.020
RlogR	10,800	1.615	17.420	-188.100	0.000	0.000	0.000	423.500
STV	10,800	1.515	1.058	0.287	1.236	0.871	1.783	10.870
SBV	10,800	0.719	1.033	0.000	0.000	0.000	1.191	9.767
SGV	10,800	0.809	1.096	0.000	0.549	0.000	1.284	10.870
RTV	10,800	30.590	117.000	0.001	1.002	0.220	3.366	702.500
RBV	10,800	29.520	115.200	0.000	0.852	0.202	3.366	702.500
RGV	10,800	29.140	114.800	0.000	0.820	0.163	3.161	702.500
Panel C: Post-vaccination (01/12/2020–30/09/2022)
SP	23,950	12,242	22,586	148.900	3380	1380	8610	150,300
RI	23,950	45.520	21.100	0.000	44.840	29.940	62.040	96.300
SlogR	23,950	0.020	1.205	-48.290	0.046	-0.490	0.586	23.200
RlogR	23,950	-0.321	6.388	-371.800	0.000	0.000	0.000	249.400
STV	23,950	1.129	0.637	0.286	1.026	0.763	1.330	27.550
SBV	23,950	0.532	0.725	0.000	0.000	0.000	0.988	25.760
SGV	23,950	0.609	0.718	0.000	0.589	0.000	1.071	27.550
RTV	23,950	29.490	116.200	0.001	0.840	0.211	2.945	702.500
RBV	23,950	19.730	95.550	0.000	0.463	0.008	1.826	702.500
RGV	23,950	28.710	114.700	0.000	0.820	0.202	2.937	702.500
ALO	23,950	51.710	31.830	0.000	60.890	20.240	78.040	108.300

Notes: This table reports descriptive statistics for variables in the total sample and the two subsamples grouped by pre-and post-vaccination, respectively. SP means stock price; RI means COVID-19 government restriction stringency index; SlogR represents stock return; RlogR represents the log change rate of COVID-19 stringency index; STV means the total volatility of stock market; SGV means the good volatility of stock market; SBV means the bad volatility of stock market; RTV means the total volatility of COVID-19 restriction stringency index; RGV means the good volatility of COVID-19 restriction stringency index; RBV means the bad volatility of COVID-19 restriction stringency index; ALO means COVID-19 vaccination rate (at least one dose).The period of the total sample is from 31 January 2020 to 30 September 2022, the period of the pre-vaccination subsample is from 31 January 2020 to 30 November 2020, and the period of the post-vaccination subsample is from 1 December 2020 to 30 September 2022. The sample date of ALO is from 1 December 2020 to 30 September 2022. The specific first COVID-19 vaccination date differs in different countries. Since Pfizer Inc. and BioNTech SE announced the results of the Phase 3 clinical trial of their mRNA-based COVID-19 vaccine with an efficacy rate of 95% on 18 November 2020, Finland, United States, United Kingdom and other countries began vaccination from December 2020. Thus, we use 1st December 2020 as the cut-off point to divide the total sample into pre-vaccination and post-vaccination subsamples.

5. Empirical results

5.1. Do changes of COVID-19 stringency index affect stock returns?

As shown in Panel A Table 3 , the coefficient of RlogR for SlogR is -0.0097 in the total sample, which is statistically significant on 1%. As shown in Panel B and C of Table 3, the coefficients of RlogR are -0.0122 and -0.0009 in the pre- and post-vaccination subsamples, respectively. The coefficient of RlogR in the pre-vaccination subsample is statistically significant on 1% level, while that in the post-vaccination subsample is insignificant. These results highlight that the overall relationship between RlogR and SlogR is negative but this correlation turns weaker after the advent of COVID-19 vaccination, providing evidence that the hypothesis H1a can hold.

Table 3.

Estimation results of the total sample and the subsamples grouped by pre-and post-vaccination for the relationship between RlogR and SlogR (RlogR → SlogR).

Variable	RlogR→ SlogR
	(1)	(2)	(3)
	Total sample	Pre-vaccination	Post-vaccination
RlogR	-0.0097⁎⁎⁎	-0.0122⁎⁎⁎	-0.0009
	(-13.877)	(-11.8732)	(-0.7588)
Constant	0.0129*	0.0089⁎⁎⁎	0.0194⁎⁎⁎
	(1.662)	(0.4907)	(2.4866)
Obs.	34,750	10,800	23,950
Adj.R2	0.0055	0.0128	0.0000
F	192.6	140.9735	0.5758

Notes: This table shows results of the total sample and the subsamples grouped by pre-and post-vaccination for the relationship between RlogR and SlogR (RlogR → SlogR), which is specified in Eq. (11). SlogR represents stock return; RlogR represents the log change rate of COVID-19 stringency index. The period of the total sample is from 3 February 2020 to 30 September 2022, the period of the pre-vaccination subsample is from 31 January 2020 to 30 November 2020, and the period of the post-vaccination subsample is from 1 December 2020 to 30 September 2022. T-statistics in parentheses. ***, **, and * denote the significance level on 1%, 5% and 10%, respectively.

5.2. Do conditional volatilities of COVID-19 stringency index affect stock conditional volatilities?

As seen Panel A of Table 4 , the estimated coefficients of RTV for STV in the total sample, pre- and post-vaccination subsamples respectively are -0.0001, 0.0003 and -0.0003, all of which are statistically significant. Correspondingly, the estimated coefficients of RGV for SGV (Table 4, Panel B) respectively are -0.0001, 0.0001 and -0.0002, of which the coefficients are statistically significant except for the pre-vaccination subsample. The estimated coefficients of RBV for SBV (Table 4, Panel C) respectively are 0.000, 0.0001 and -0.0001, of which the coefficients are statistically significant except for the total sample. It means that our hypotheses, H1b, H1c and H1d can hold, respectively.

Table 4.

The impact of the volatilities of COVID-19 stringency index on the volatilities of stock market.

	Total sample	Pre-vaccination	Post -vaccination
	(1)	(2)	(3)
Panel A: RTV→ STV
RTV	-0.0001*	0.0003⁎⁎⁎	-0.0003⁎⁎⁎
	(-1.733)	(3.8610)	(-7.2768)
Constant	1.2508⁎⁎⁎	1.5046⁎⁎⁎	1.1366⁎⁎⁎
	(278.084)	(143.0778)	(267.7285)
Obs.	34,750	10,800	23,950
F	3.004	14.9069	52.9525
adj. R2	0.0001	0.0013	0.0022

	Total sample	Pre-vaccination	Post -vaccination
	(1)	(2)	(3)
Panel B: RGV→ SGV
RGV	-0.0001⁎⁎	0.0001	-0.0002⁎⁎⁎
	(-2.163)	(1.2330)	(-4.4361)
Constant	0.6735⁎⁎⁎	0.8055⁎⁎⁎	0.6141⁎⁎⁎
	(141.825)	(74.0000)	(128.4871)
Obs.	34,750	10,800	23,950
F	4.679	1.5202	19.6793
adj. R2	0.0001	0.0000	0.0008
Panel C: RBV→ SBV
RBV	0.0000	0.0001⁎⁎	-0.0001⁎⁎⁎
	(0.254)	(1.6960)	(-2.8519)
Constant	0.5900⁎⁎⁎	0.7150⁎⁎⁎	0.5347⁎⁎⁎
	(128.223)	(69.7154)	(111.8553)
Obs.	34,750	10,800	23,950
F	0.0646	2.8764	8.1334
adj. R2	-0.0000	0.0002	0.0003

Notes: This table shows estimation results for the impact of the volatilities of COVID-19 stringency index on the volatilities of stock market, using the regression model specified in Eq. (11) with the total sample, the pre- and post-vaccination subsamples, respectively. STV means the total volatility of stock market; SGV means the good volatility of stock market; SBV means the bad volatility of stock market; RTV means the total volatility of COVID-19 restriction stringency index; RGV means the good volatility of COVID-19 restriction stringency index; RBV means the bad volatility of COVID-19 restriction stringency index. Panel A, B and C of this table respectively report the results for the impact of total volatility of stringency index on total volatility of stock market (RTV→ STV), the impact of good volatility of stringency index on good volatility of stock market (RGV→ SGV), and the impact of bad volatility of stringency index on bad volatility of stock market (RBV→ SBV). 0.0000 means the value less than 0.0001. ***, **, and * denote the significance level on 1%, 5% and 10%, respectively.

Those results suggest that the volatilities of COVID-19 stringency index have impact on the volatilities of stock market, but this relationship is different between pre- and post-vaccination periods. It is positive before the advent of COVID-19 vaccination but negative after that, which is found via comparing the results shown in columns (2) and (3) of Table 4. In addition, this relationship is also different between good volatility and bad volatility. In the pre-vaccination subsample, although both the coefficient of RBV for SBV and the coefficient of RGV for SGV are positive, the former coefficient is significant at 5% level while the latter is not. It indicates that higher volatility of COVID-19 stringency index causes the higher volatility of stock market, and bad news from COVID-19 restriction policy causes greater volatilities than good news.

5.3. Does the advent of COVID-19 vaccination affect the relationship between COVID-19 stringency index and stock market?

Table 5 reports the estimated results for whether the advent of COVID-19 vaccination affect the relationship between COVID-19 stringency index and stock markets. In Panel A of this table, the estimated coefficient of RlogR for SlogR in column (1), (2) and (3) respectively are -0.0097, -0.0096 and 0.0105. The coefficient of RlogR×Vac_dum for SlogR is -0.0113. All of them are significant at 1% level. It suggests that the advent of COVID-19 vaccination has a reverse moderate effect on the relationship between RlogR and SlogR, supporting the hypothesis H2a.

Table 5.

Estimation results for the impact of the advent of COVID-19 vaccination on the relationship between COVID-19 stringency index and stock markets.

	(1)	(2)	(3)
Panel A: RlogR→ SlogR (Dependent variable-SlogR)
RlogR	-0.0097***	-0.0096***	0.0105***
	(-13.843)	(-13.741)	(3.451)
Vac_dum		-0.0120	-0.0106
		(-0.712)	(-0.629)
RlogR×Vac_dum			-0.0113***
			(-6.814)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Obs.	34,750	34,750	34,750
adj. R2	0.0041	0.0040	0.0053
F	191.6	96.07	79.61
Panel B: RTV→ STV (Dependent variable-STV)
RTV	0.0038***	0.0028***	0.0019***
	(9.403)	(7.157)	(4.714)
Vac_dum		0.3828***	0.3655***
		(47.362)	(43.848)
RTV×Vac_dum			0.0006***
			(8.331)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Obs.	34,750	34,750	34,750
adj. R2	0.0011	0.0617	0.0636
F	88.42	1169	803.8
Panel C: RGV→ SGV (Dependent variable-SGV)
RGV	-0.0001	-0.0001	-0.0005**
	(-0.378)	(-0.562)	(-2.250)
Vac_dum		0.1999***	0.1914***
		(20.821)	(19.334)
RGV×Vac_dum			0.0003***
			(3.499)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Obs.	34,750	34,750	34,750
adj. R2	-0.0014	0.0109	0.0112
F	0.143	216.8	148.7
Panel D: RBV→ SBV (Dependent variable-SBV)
RBV	0.0002*	0.0000	-0.0004**
	(1.960)	(0.236)	(-2.442)
Vac_dum		0.1872***	0.1801***
		(19.718)	(18.463)
RBV×Vac_dum			0.0003***
			(3.138)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Obs.	34,750	34,750	34,750
adj. R2	-0.0013	0.0097	0.0100
F	3.841	196.3	134.2

Notes: This table shows the estimation results for the impact of the advent of COVID-19 vaccination on the relationship between COVID-19 stringency index and stock markets. Vac_dum is a dummy variable set to one for the sample period during COVID-19 vaccination which is from 1 December 12 2020, to 30 September 2022. Columns (1), (2) and (3) report estimated results of the models specified in Eqs. (12), (13) and (14), respectively. 0.0000 means a value less than 0.0001. T-statistics in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

As shown in Panel B of Table 5, respectively. The coefficients of RTV for STV respectively are 0.0038, 0.0028 and 0.0019 in columns (1), (2) and (3), all of which are statistically significant at the 1% level. The coefficient of RTV×Vac_dum is positively significant. This empirical result implies that the advent of COVID-19 vaccination weakens the positive impact of RTV on STV, supporting the hypothesis H2b.

As reported in Panel C and D of Table 5, the coefficient of RGV×Vac_dum for SGV is 0.0003 and that of RBV×Vac_dum for SBV is 0.0003, which are all statistically significant at the 1% level. Comparing the coefficients of RGV for SGV in columns (1) and (3), it turns from -0.0001 to -0.0005 and from insignificant to significant, implying that the negative correlation between RGV and SGV is enhanced during the COVID-19 vaccination period. However, the estimated coefficient of RBV for SBV turns from 0.0002 to -0.0004, suggesting that the bad volatility of COVID-19 stringency index could not cause the bad volatility of stock markets after the advent of COVID-19 vaccination. The hypotheses H2c and H2d also can been supported. These results provide evidence that our hypothesis H2 can hold.

5.4. Does COVID-19 stringency index moderate the relationship between COVID-19 vaccination rate and stock market?

In this subsection, we further discuss the relationship between COVID-19 vaccination rate (ALO) and stock market performance, including ALO→SlogR, ALO→STV, ALO→SGV, and ALO→SBV.

As reported in Table 6 , the estimated coefficients of ALO for RlogR, RTV, RGV, and RBV respectively are -0.0069, -0.0117, -0.0121 and -0.1551, all of which are statistically significant at the 1% level, indicating that COVID-19 vaccination rate negatively correlates to COVID-19 stringency index.

Table 6.

The impact of COVID-19 vaccination on COVID-19 restriction index.

	Post-vaccination (01/12/2020–30/09/2022)
Panel A: ALO→RlogR (Dependent variable- RlogR)
ALO	-0.0069***
	(-5.040)
Constant	0.0376
	(0.457)
Obs.	23,950
adj. R2	-0.0010
Country FE	Yes
Time FE	Yes
F	25.40
Panel B: ALO→RTV (Dependent variable- RTV)
ALO	-0.0117***
	(-6.012)
Constant	30.0961***
	(258.970)
Obs.	23,950
adj. R2	-0.0006
Country FE	Yes
Time FE	Yes
F	36.14
Panel C: ALO→RGV (Dependent variable- RGV)
ALO	-0.0121***
	(-2.608)
Constant	29.3316***
	(105.793)
Obs.	23,950
adj. R2	-0.0018
Country FE	Yes
Time FE	Yes
F	6.800
Panel B: ALO→RBV (Dependent variable- RBV)
ALO	-0.1551***
	(-12.800)
Constant	27.7492***
	(38.303)
Obs.	23,950
adj. R2	0.0047
Country FE	Yes
Time FE	Yes
F	163.8

Notes: This table shows the estimation results for the impact of COVID-19 vaccination rate on COVID-19 stringency index, using the model specified in Eq. (15) .The sample period is from 1 December 2020 to 30 September 2022. ALO means COVID-19 vaccination rate (at least one dose); RlogR represents the log change rate of COVID-19 stringency index; RTV means the total volatility of COVID-19 restriction stringency index; RGV means the good volatility of COVID-19 restriction stringency index; RBV means the bad volatility of COVID-19 restriction stringency index. The impact of ALO on RlogR, RTV, RGV and RBV are reported in Panel A, B, C and D of this table, respectively. T-statistics in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

As seen from columns (1), (2) and (3) of Panel A in Table 7 , the estimated coefficients of ALO for SlogR respectively are -0.00219, -0.00220 and -0.00222, all of which are statistically significant at the 1% level, highlighting the negative correlation between ALO and SlogR, even after controlling RlogR and ALO×RlogR. The estimated coefficient of ALO×RlogR is 0.0001 and significant at the 5% level, indicating a positive interaction. Connecting with the corresponding result on the significantly negative relationship between ALO and RlogR shown in Panel A of Table 6, we can get the conclusion that RlogR plays a partly mediating role in the influence of ALO on SlogR. It means that the hypothesis H3a can hold.

Table 7.

The impact of COVID-19 vaccination on stock markets.

	(1)	(2)	(3)
	Post-vaccination (01/12/2020–30/09/2022)
Panel A: ALO→SlogR (Dependent variable- SlogR)
ALO	-0.00219***	-0.00220***	-0.00222***
	(-8.462)	(-8.490)	(-8.576)
RlogR		-0.00122	-0.00258*
		(-1.004)	(-1.878)
ALO×RlogR			0.00010**
			(2.134)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Observations	23,950	23,950	23,950
adj. R2	0.0009	0.0009	0.0010
F	71.60	36.30	25.72
Panel B: ALO→STV (Dependent variable- STV)
ALO	0.00223***	0.00224***	0.00224***
	(19.901)	(19.943)	(19.933)
RTV		0.00054	0.00051
		(1.441)	(1.363)
ALO×RTV			-0.00000
			(-0.432)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Observations	23,950	23,950	23,950
adj. R2	0.0142	0.0143	0.0143
F	396.1	199.1	132.8
Panel C: ALO→SGV (Dependent variable- SGV)
ALO	-0.000036	-0.000036	-0.000038
	(-0.242)	(-0.241)	(-0.254)
RGV		0.000009	-0.000011
		(0.044)	(-0.052)
ALO×RGV			-0.000002
			(-1.137)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Observations	23,950	23,950	23,950
adj. R2	-0.0021	-0.0021	-0.0021
F	0.0587	0.0303	0.451
Panel D: ALO→SBV (Dependent variable- SBV)
ALO	0.002262***	0.002262***	0.002262***
	(14.949)	(14.898)	(14.891)
RBV		0.000001	0.000002
		(0.015)	(0.025)
ALO×RBV			0.000000
			(0.059)
Country FE	Yes	Yes	Yes
Time FE	Yes	Yes	Yes
Observations	23,950	23,950	23,950
adj. R2	0.0072	0.0071	0.0071
F	223.5	111.7	74.48

Notes: This table shows the estimation results for the impact of COVID-19 stringency index on the relationship between COVID-19 vaccination and stock markets. The sample period is from 1 December 2020 to 30 September 2022. Columns (1), (2) and (3) report estimated results of the models specified in Eqs. (16), (17) and (18), respectively. 0.0000 means a value less than 0.0001. T-statistics in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

As shown in Panel B, C and D of Table 7, although the estimated coefficients for ALO×RTV, ALO×RGV and ALO×RBV are not statistically significant, the absolute coefficients of ALO for STV, SGV and SBV in column (3) are respectively larger than corresponding ones in column (1), suggesting that COVID-19 stringency index’ volatilities slightly strengthen the impact of COVID-19 cumulative vaccination rate on stock volatilities. This means that the hypotheses H3b, H3c, and H3d cannot fully hold. In sum, our hypothesis H3 is partially supported.

6. Conclusions

In this paper, we take the COVID-19 stringency index proposed by the University of Oxford as the proxy to investigate the impact of COVID-19 government restriction policy on stock markets and the role of COVID-19 vaccination, using a sample of 50 countries spanning from 31 January 2020 to 30 September 2022.

Our findings can be concluded as follows. Firstly, the relationship between COVID-19 stringency indexes and stock markets is time-varying. Before the advent of COVID-19 vaccination, we find that the changes of COVID-19 stringency indexes have negative impact on stock market returns, which keeps in line with Abdullah et al (2022), Aharon & Siev (2021), Kheni & Kumar (2021) and Caporale et al (2022). Our empirical results suggest that the total volatility of COVID-19 stringency indexes positively correlates to the total volatility of stock markets, on some extent, providing evidence that support the research results of Martins & Cró (2022), Bouri et al (2022),Wang et al (2021),Narayan et al., 2021; Xie & Zhou (2022). Different from previous literature, we find that the impact of the bad volatility of COVID-19 stringency index on stock market's bad volatility is stronger than that of the good volatility of COVID-19 stringency index on stock market's good volatility, giving more details on the relationship. After the advent of COVID-19 vaccination, the impact of COVID-19 stringency index changes on stock returns turns insignificantly negative even significantly positive, and the linkage between volatilities turns from significantly positive to insignificantly positive even significantly negative, which is consistent with Apergis et al (2022). Secondly, there is evidence that the advent of COVID-19 vaccination weakens the impact of COVID-19 stringency indexes on stock markets. On some extent, this result is consistent with Yu et al (2022), Mishra et al (2022) and Hong et al. (2021). Thirdly, COVID-19 cumulative vaccination rate significantly affects stock markets, and COVID-19 restriction stringency index negatively correlates to COVID-19 cumulative vaccination rate, while COVID-19 restriction stringency index only plays a partially mediate role in the relationship between COVID-19 cumulative vaccination rate and stock markets. That is, COVID-19 cumulative vaccination rate directly influences stock markets at most time.

Our paper is the first study to systematically analyze the relationship between COVID-19 government restriction policy, COVID-19 vaccination and stock market from a global perspective. Different from previous literature, it divides the total volatilities of stock markets and COVID-19 stringency indexes into good and bad ones, to investigate their relationship in more details. It explores the mediate role of the advent of COVID-19 vaccination, providing a possible reason for the time-varying correlations between COVID-19 stringency indexes and stock markets, as well as reconciling the two controversial points of view (positive or negative). Our study also explores the path that COVID-19 cumulative vaccination rate influences stock markets. It has important implications for understanding the impact of COVID-19 government restrictions on the stock market and the role of COVID-19 vaccination. Our paper has some important contributions to the literature, but also has its limitations. For example, in the process of sample selection, we delete some countries of which there are too many missing values in the COVID-19 vaccination data, resulted in the fact that some special countries such as China and Ukraine are not involved in this study. We could further individually investigate the relationship among COVID-19 government restriction policy, COVID-19 vaccination and financial markets in these countries in the future. Besides, when we examine the relationship between COVID-19 vaccination and stock markets, we only examine the role of the cumulative vaccination rate, ignoring the types of COVID-19 vaccine in given countries. Thus, it could be very interesting to further study the role of different types of COVID-19 vaccine in the relationship between COVID-19 cumulative vaccination rate and financial markets.

CRediT authorship contribution statement

Xiaoling Yu: Conceptualization, Methodology, Formal analysis, Data curation, Validation, Visualization, Writing – original draft, Writing – review & editing. Kaitian Xiao: Software, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

There is no potential conflict of interest

Appendix

Appendix A1. Descriptive statistics of stock market, COVID-19 stringency index and vaccination for each country

Country	ISO3	SP	SlogR	RI	RlogR	ALO	ALO
		Mean	Mean	Mean	Mean	Mean	Max
Peru	PER	19,541.30	-0.003	64.56	0.000	49.54	90.67
India	IND	49,047.01	0.050	63.08	0.176	42.91	74.37
Malaysia	MYS	1532.92	-0.013	59.35	0.084	53.50	86.84
Vietnam	VNM	264.00	0.129	59.12	0.000	45.89	92.44
Greece	GRC	801.95	-0.020	58.31	0.397	52.02	76.07
Indonesia	IDN	6060.45	0.025	57.75	0.494	39.43	74.62
Argentina	ARG	67,530.58	0.179	57.19	-0.359	58.90	91.44
Kazakhstan	KAZ	2917.34	0.042	56.64	0.271	32.91	57.71
Chile	CHL	4405.69	0.016	56.21	0.479	68.84	94.53
Canada	CAN	18,633.41	0.009	56.04	0.198	62.98	88.40
Sri Lanka	LKA	7708.19	0.074	54.91	0.153	49.87	79.99
Pakistan	PAK	29,333.86	-0.004	54.85	0.536	31.20	63.11
Colombia	COL	1325.66	-0.052	54.63	0.389	49.67	105.46
Ecuador	ECU	1312.31	-0.016	54.04	0.389	53.56	86.36
Brazil	BRA	107,860.73	-0.005	52.29	0.455	55.88	101.79
Austria	AUT	3007.11	-0.019	51.92	0.516	54.27	76.49
Tunisia	TUN	7039.69	0.023	51.88	0.540	37.23	75.28
Australia	AUS	6969.79	-0.009	51.83	0.088	53.27	88.01
United States	USA	3881.31	0.015	51.63	0.474	58.12	80.07
Germany	DEU	13,834.30	-0.010	50.20	0.127	55.40	77.32
Spain	ESP	809.50	-0.035	49.96	0.389	62.11	88.35
Cambodia	KHM	583.01	-0.068	49.60	0.441	56.83	90.89
United Arab Emirates	ARE	2751.50	0.027	48.34	0.329	74.34	101.02
Mexico	MEX	2812.48	0.009	48.12	0.421	43.40	75.25
Portugal	PRT	3796.78	0.025	47.95	0.359	66.56	95.74
United Kingdom	GBR	3830.42	-0.011	47.85	-0.051	62.15	79.27
Ireland	IRL	7245.47	-0.015	46.76	0.359	59.47	83.02
France	FRA	5935.37	-0.001	46.35	0.088	59.58	83.62
Turkey	TCA	1840.11	0.149	45.98	0.359	47.67	68.69
Romania	ROU	10,906.48	0.009	45.50	0.000	26.27	35.86
Singapore	SGP	2996.10	-0.001	45.49	-0.041	62.74	88.21
Russian Federation	RUS	1359.92	-0.052	45.18	0.177	33.52	58.31
Mongolia	MNG	29,346.21	0.085	44.81	0.359	52.86	69.33
South Korea	KOR	2657.23	0.003	44.79	0.197	55.47	88.00
Slovenia	SVN	1033.93	-0.001	44.15	0.359	43.73	60.89
Belgium	BEL	3782.83	-0.021	43.63	0.397	58.09	79.89
Luxembourg	LUX	1408.08	-0.006	43.54	0.509	54.44	76.95
Malta	MLT	3853.13	-0.036	43.51	0.359	79.42	108.25
Poland	POL	58,054.48	-0.030	43.43	0.335	43.22	60.21
Slovakia	SVK	364.71	-0.003	42.60	0.359	37.20	50.85
Czech Republic	CZE	1134.10	0.007	42.53	0.359	46.69	65.12
Japan	JPN	1815.79	0.012	42.36	0.310	51.82	82.40
Latvia	LVA	1137.57	0.003	39.77	0.359	46.43	71.37
Hungary	HUN	42,829.88	-0.019	39.13	0.000	50.75	66.44
Switzerland	CHE	715.09	-0.009	38.71	0.271	50.37	70.40
Denmark	DNK	1231.26	0.029	38.40	0.359	59.20	82.95
Finland	FIN	10,996.63	-0.002	38.36	0.193	58.87	81.61
Lithuania	LTU	1249.17	0.032	37.51	0.359	51.19	71.83
Croatia	HRV	1868.58	-0.010	36.67	0.359	39.11	56.48
Estonia	EST	1249.02	0.032	34.95	0.359	46.68	65.36

Notes: This table reports the mean values of stock index price (SP), stock return (SlogR), COVID-19 stringency index (RI) and COVID-19 stringency index change rate (RlogR), as well as the mean value and the max value of COVID-19 vaccination rate (ALO)for each country. It is ranked from largest to smallest according to COVID-19 stringency index (RI). The sample date of SP and RI is from 31 January 2020 to 30 September 2022, and that of SlogR and RlogR is from 1 February 2020 to 30 September 2022, including 696 and 695 observations for each country, respectively. The sample date of ALO is from 1 December 2020 to 30 September 2022, generating 479 observations for each country.0.000 means the value less than 0.001. Some max values of ALO are more than 100, possibly in that we take the middle-year estimated population in 2020 to calculate the vaccination rate.

Appendix A2. Stationarity, normal distribution and ARCH effect tests for time series

ISO3	ADF	PP	JB	LB (10)	Q (10)
Panel A: Stock return (SlogR)
ARE	-5.4604⁎⁎⁎	-23.1872⁎⁎⁎	6674.97⁎⁎⁎	57.6277⁎⁎⁎	57.1771⁎⁎⁎
ARG	-24.5102⁎⁎⁎	-24.7144⁎⁎⁎	991.55⁎⁎⁎	15.4159	15.2473
AUS	-5.9636⁎⁎⁎	-31.8471⁎⁎⁎	3757.59⁎⁎⁎	86.0459⁎⁎⁎	85.3245⁎⁎⁎
AUT	-5.8668⁎⁎⁎	-24.7329⁎⁎⁎	4318.59⁎⁎⁎	44.2293⁎⁎⁎	43.8596⁎⁎⁎
BEL	-6.1749⁎⁎⁎	-25.7423⁎⁎⁎	9587.94⁎⁎⁎	26.6425⁎⁎⁎	26.3460⁎⁎⁎
BRA	-6.5059⁎⁎⁎	-32.1848⁎⁎⁎	10,597.97⁎⁎⁎	109.6271⁎⁎⁎	108.6609⁎⁎⁎
CAN	-8.4331⁎⁎⁎	-31.4963⁎⁎⁎	23,760.98⁎⁎⁎	156.0669⁎⁎⁎	154.4162⁎⁎⁎
CHE	-7.9779⁎⁎⁎	-26.2962⁎⁎⁎	4332.86⁎⁎⁎	22.3210⁎⁎	22.0451⁎⁎
CHL	-11.9718⁎⁎⁎	-25.7245⁎⁎⁎	6865.77⁎⁎⁎	22.2553⁎⁎	22.0184⁎⁎
COL	-14.5445⁎⁎⁎	-22.6263⁎⁎⁎	20,125.99⁎⁎⁎	52.8565⁎⁎⁎	52.5028⁎⁎⁎
CZE	-6.6430⁎⁎⁎	-24.7882⁎⁎⁎	2797.73⁎⁎⁎	36.7453⁎⁎⁎	36.4233⁎⁎⁎
DEU	-6.9604⁎⁎⁎	-27.0506⁎⁎⁎	3742.32⁎⁎⁎	38.3751⁎⁎⁎	37.9347⁎⁎⁎
DNK	-11.1244⁎⁎⁎	-25.1307⁎⁎⁎	413.45⁎⁎⁎	16.0675*	15.8977
ECU	-14.2802⁎⁎⁎	-25.9505⁎⁎⁎	108,281.84⁎⁎⁎	10.6375	10.5429
ESP	-10.0198⁎⁎⁎	-27.6037⁎⁎⁎	7174.65⁎⁎⁎	48.4924⁎⁎⁎	48.0336⁎⁎⁎
EST	-6.1341⁎⁎⁎	-23.3261⁎⁎⁎	17,788.99⁎⁎⁎	44.5868⁎⁎⁎	44.2029⁎⁎⁎
FIN	-8.9017⁎⁎⁎	-25.1779⁎⁎⁎	1954.44⁎⁎⁎	23.4075⁎⁎⁎	23.1732⁎⁎
FRA	-6.9917⁎⁎⁎	-26.6523⁎⁎⁎	3693.38⁎⁎⁎	29.9414⁎⁎⁎	29.6138⁎⁎⁎
GBR	-27.1159⁎⁎⁎	-27.1263⁎⁎⁎	909.32⁎⁎⁎	6.4650	6.4051
GRC	-6.4222⁎⁎⁎	-27.6469⁎⁎⁎	6558.01⁎⁎⁎	47.8327⁎⁎⁎	47.3419⁎⁎⁎
HRV	-8.1962⁎⁎⁎	-30.0192⁎⁎⁎	44,746.78⁎⁎⁎	114.6849⁎⁎⁎	113.7140⁎⁎⁎
HUN	-6.3925⁎⁎⁎	-25.9674⁎⁎⁎	3374.50⁎⁎⁎	31.6940⁎⁎⁎	31.4364⁎⁎⁎
IDN	-9.6553⁎⁎⁎	-25.2816⁎⁎⁎	1082.92⁎⁎⁎	23.2011⁎⁎	22.9816⁎⁎
IND	-7.2923⁎⁎⁎	-27.3426⁎⁎⁎	9252.58⁎⁎⁎	55.0267⁎⁎⁎	54.3955⁎⁎⁎
IRL	-9.3030⁎⁎⁎	-25.4923⁎⁎⁎	946.95⁎⁎⁎	32.8917⁎⁎⁎	32.5298⁎⁎⁎
JPN	-15.0392⁎⁎⁎	-24.7462⁎⁎⁎	293.82⁎⁎⁎	10.9586	10.8694
KAZ	-7.5643⁎⁎⁎	-26.6985⁎⁎⁎	188.43⁎⁎⁎	52.2245⁎⁎⁎	51.7679⁎⁎⁎
KHM	-16.8247⁎⁎⁎	-21.1263⁎⁎⁎	7491.42⁎⁎⁎	67.8494⁎⁎⁎	67.2726⁎⁎⁎
KOR	-16.4328⁎⁎⁎	-26.8484⁎⁎⁎	1373.89⁎⁎⁎	16.6806*	16.5472*
LKA	-7.0788⁎⁎⁎	-22.4242⁎⁎⁎	776.42⁎⁎⁎	64.5057⁎⁎⁎	63.9964⁎⁎⁎
LTU	-5.5817⁎⁎⁎	-23.2541⁎⁎⁎	18,305.71⁎⁎⁎	44.7542⁎⁎⁎	44.3738⁎⁎⁎
LUX	-8.2455⁎⁎⁎	-27.1644⁎⁎⁎	505.02⁎⁎⁎	27.4143⁎⁎⁎	27.1391⁎⁎⁎
LVA	-9.2845⁎⁎⁎	-32.9038⁎⁎⁎	61,607.68⁎⁎⁎	77.3256⁎⁎⁎	76.7637⁎⁎⁎
MEX	-7.1336⁎⁎⁎	-25.9602⁎⁎⁎	295.69⁎⁎⁎	18.7196⁎⁎	18.4940⁎⁎
MLT	-9.5278⁎⁎⁎	-24.2931⁎⁎⁎	1437.99⁎⁎⁎	28.4369⁎⁎⁎	28.2145⁎⁎⁎
MNG	-9.7290⁎⁎⁎	-20.3031⁎⁎⁎	1130.13⁎⁎⁎	160.5000⁎⁎⁎	159.5223⁎⁎⁎
MYS	-13.2653⁎⁎⁎	-26.5568⁎⁎⁎	1813.75⁎⁎⁎	21.6274⁎⁎	21.3775⁎⁎
PAK	-10.1353⁎⁎⁎	-24.0705⁎⁎⁎	713.55⁎⁎⁎	26.4669⁎⁎⁎	26.2458⁎⁎⁎
PER	-7.5071⁎⁎⁎	-27.2868⁎⁎⁎	1573.37⁎⁎⁎	32.3538⁎⁎⁎	32.0230⁎⁎⁎
POL	-26.3910⁎⁎⁎	-26.6320⁎⁎⁎	5471.65⁎⁎⁎	12.6226	12.4657
PRT	-7.4840⁎⁎⁎	-25.8218⁎⁎⁎	2216.23⁎⁎⁎	23.5954⁎⁎⁎	23.3068⁎⁎⁎
ROU	-10.6955⁎⁎⁎	-26.4540⁎⁎⁎	5482.36⁎⁎⁎	44.7059⁎⁎⁎	44.2784⁎⁎⁎
RUS	-11.5112⁎⁎⁎	-29.1367⁎⁎⁎	243,391.18⁎⁎⁎	23.7524⁎⁎⁎	23.5860⁎⁎⁎
SGP	-9.2380⁎⁎⁎	-27.8930⁎⁎⁎	3109.46⁎⁎⁎	50.2825⁎⁎⁎	49.8498⁎⁎⁎
SVK	-13.1026⁎⁎⁎	-28.6683⁎⁎⁎	5225.38⁎⁎⁎	17.1323*	16.9760*
SVN	-7.4147⁎⁎⁎	-26.8719⁎⁎⁎	7726.77⁎⁎⁎	52.6866⁎⁎⁎	52.1906⁎⁎⁎
TCA	-16.0306⁎⁎⁎	-27.3758⁎⁎⁎	2389.24⁎⁎⁎	26.8972⁎⁎⁎	26.6695⁎⁎⁎
TUN	-10.7316⁎⁎⁎	-17.8677⁎⁎⁎	11,363.84⁎⁎⁎	117.2885⁎⁎⁎	116.6536⁎⁎⁎
USA	-7.2930⁎⁎⁎	-32.3292⁎⁎⁎	4356.24⁎⁎⁎	203.6719⁎⁎⁎	201.3908⁎⁎⁎
VNM	-6.6425⁎⁎⁎	-24.3327⁎⁎⁎	432.99⁎⁎⁎	45.3498⁎⁎⁎	44.8801⁎⁎⁎
ISO3	ADF	PP	JB	LB (10)	Q (10)
Panel B: The Change rate of COVID-19 restriction stringency index (RlogR)
ARE	-5.8298⁎⁎⁎	-29.3459⁎⁎⁎	492,144.84⁎⁎⁎	189.6265⁎⁎⁎	187.3396⁎⁎⁎
ARG	-4.5234⁎⁎⁎	-26.2120⁎⁎⁎	5,875,949.03⁎⁎⁎	6.0104	5.9583
AUS	-30.7283⁎⁎⁎	-31.1183⁎⁎⁎	128,086.95⁎⁎⁎	4.5076	4.4560
AUT	-7.3735⁎⁎⁎	-26.7332⁎⁎⁎	5,779,978.35⁎⁎⁎	23.6785⁎⁎⁎	23.3467⁎⁎⁎
BEL	-12.9152⁎⁎⁎	-25.9706⁎⁎⁎	5,925,033.73⁎⁎⁎	4.1556	4.1205
BRA	-8.0238⁎⁎⁎	-27.3299⁎⁎⁎	3,016,238.78⁎⁎⁎	64.1852⁎⁎⁎	63.6671⁎⁎⁎
CAN	-7.9454⁎⁎⁎	-23.9638⁎⁎⁎	1,756,978.55⁎⁎⁎	126.9919⁎⁎⁎	126.1025⁎⁎⁎
CHE	-8.7989⁎⁎⁎	-26.6253⁎⁎⁎	2,637,782.64⁎⁎⁎	33.7209⁎⁎⁎	33.4910⁎⁎⁎
CHL	-26.3320⁎⁎⁎	-26.3953⁎⁎⁎	12,538,580.46⁎⁎⁎	6.2441	6.1709
COL	-5.4901⁎⁎⁎	-26.2574⁎⁎⁎	5,495,078.33⁎⁎⁎	58.9309⁎⁎⁎	58.4003⁎⁎⁎
CZE	-5.7221⁎⁎⁎	-26.4977⁎⁎⁎	6,867,684.78⁎⁎⁎	32.4719⁎⁎⁎	32.0296⁎⁎⁎
DEU	-4.5818⁎⁎⁎	-27.6925⁎⁎⁎	145,305.52⁎⁎⁎	79.0177⁎⁎⁎	77.9998⁎⁎⁎
DNK	-5.5902⁎⁎⁎	-27.4440⁎⁎⁎	4,180,758.83⁎⁎⁎	41.7743⁎⁎⁎	41.3987⁎⁎⁎
ECU	-5.5527⁎⁎⁎	-25.2816⁎⁎⁎	2,322,520.49⁎⁎⁎	116.9917⁎⁎⁎	115.2554⁎⁎⁎
ESP	-8.8571⁎⁎⁎	-41.7276⁎⁎⁎	4,982,081.16⁎⁎⁎	4.1444	4.1168
EST	-15.9886⁎⁎⁎	-26.8666⁎⁎⁎	7,385,729.32⁎⁎⁎	19.4002⁎⁎	19.2585⁎⁎
FIN	-7.3842⁎⁎⁎	-27.1795⁎⁎⁎	279,023.69⁎⁎⁎	27.5565⁎⁎⁎	27.2917⁎⁎⁎
FRA	-4.9871⁎⁎⁎	-26.6226⁎⁎⁎	466,289.24⁎⁎⁎	55.3803⁎⁎⁎	54.8617⁎⁎⁎
GBR	-9.6380⁎⁎⁎	-23.1761⁎⁎⁎	305,206.20⁎⁎⁎	87.2926⁎⁎⁎	86.7115⁎⁎⁎
GRC	-10.0613⁎⁎⁎	-27.6621⁎⁎⁎	4,358,580.87⁎⁎⁎	36.2357⁎⁎⁎	35.9190⁎⁎⁎
HRV	-6.6524⁎⁎⁎	-26.3087⁎⁎⁎	3,455,503.74⁎⁎⁎	12.4907	12.2938
HUN	-26.2268⁎⁎⁎	-26.5554⁎⁎⁎	587,506.14⁎⁎⁎	9.9346	9.8263
IDN	-6.3067⁎⁎⁎	-24.6320⁎⁎⁎	11,309,628.67⁎⁎⁎	5.3646	5.3135
IND	-5.1760⁎⁎⁎	-27.8143⁎⁎⁎	844,495.99⁎⁎⁎	89.2634⁎⁎⁎	88.2426⁎⁎⁎
IRL	-4.7376⁎⁎⁎	-26.0270⁎⁎⁎	2,631,926.92⁎⁎⁎	56.3566⁎⁎⁎	55.5465⁎⁎⁎
JPN	-37.9665⁎⁎⁎	-28.8756⁎⁎⁎	1,172,814.74⁎⁎⁎	100.0043⁎⁎⁎	99.5536⁎⁎⁎
KAZ	-11.2014⁎⁎⁎	-22.1978⁎⁎⁎	5,043,829.11⁎⁎⁎	31.2014⁎⁎⁎	31.0366⁎⁎⁎
KHM	-8.6434⁎⁎⁎	-26.8934⁎⁎⁎	7,765,039.00⁎⁎⁎	20.6459⁎⁎	20.4280⁎⁎
KOR	-38.6867⁎⁎⁎	-41.6056⁎⁎⁎	2,459,523.28⁎⁎⁎	9.4677	9.4241
LKA	-12.8626⁎⁎⁎	-23.3273⁎⁎⁎	400,779.38⁎⁎⁎	33.6277⁎⁎⁎	33.3881⁎⁎⁎
LTU	-5.5125⁎⁎⁎	-26.3346⁎⁎⁎	3,206,227.99⁎⁎⁎	87.6902⁎⁎⁎	86.3945⁎⁎⁎
LUX	-6.7051⁎⁎⁎	-27.3673⁎⁎⁎	5,712,362.67⁎⁎⁎	79.0123⁎⁎⁎	78.3062⁎⁎⁎
LVA	-9.3288⁎⁎⁎	-29.9984⁎⁎⁎	721,615.27⁎⁎⁎	63.8076⁎⁎⁎	62.9695⁎⁎⁎
MEX	-4.4068⁎⁎⁎	-21.1467⁎⁎⁎	934,343.24⁎⁎⁎	47.5376⁎⁎⁎	47.3028⁎⁎⁎
MLT	-7.5718⁎⁎⁎	-27.4794⁎⁎⁎	5,245,946.89⁎⁎⁎	70.5002⁎⁎⁎	69.9002⁎⁎⁎
MNG	-26.3128⁎⁎⁎	-26.3260⁎⁎⁎	2,117,381.19⁎⁎⁎	0.0579	0.0571
MYS	-8.1019⁎⁎⁎	-27.1882⁎⁎⁎	173,666.43⁎⁎⁎	45.5252⁎⁎⁎	45.1914⁎⁎⁎
PAK	-10.4739⁎⁎⁎	-26.6515⁎⁎⁎	6,983,842.48⁎⁎⁎	8.3346	8.2837
PER	-4.0416⁎⁎⁎	-25.9545⁎⁎⁎	3,173,934.83⁎⁎⁎	14.5916	14.4402
POL	-7.5613⁎⁎⁎	-27.4376⁎⁎⁎	3,782,170.64⁎⁎⁎	51.4756⁎⁎⁎	50.9569⁎⁎⁎
PRT	-8.5032⁎⁎⁎	-32.4710⁎⁎⁎	536,952.34⁎⁎⁎	160.9973⁎⁎⁎	158.8881⁎⁎⁎
ROU	-4.7592⁎⁎⁎	-26.6248⁎⁎⁎	1,735,281.34⁎⁎⁎	11.4662	11.2972
RUS	-6.7695⁎⁎⁎	-26.5754⁎⁎⁎	59,455.31⁎⁎⁎	37.5476⁎⁎⁎	37.1315⁎⁎⁎
SGP	-16.6258⁎⁎⁎	-26.4544⁎⁎⁎	99,483.28⁎⁎⁎	12.5178	12.4297
SVK	-9.4446⁎⁎⁎	-24.9602⁎⁎⁎	6,048,910.76⁎⁎⁎	34.7908⁎⁎⁎	34.5396⁎⁎⁎
SVN	-6.7316⁎⁎⁎	-27.0894⁎⁎⁎	2,922,601.89⁎⁎⁎	55.9129⁎⁎⁎	55.2737⁎⁎⁎
TCA	-12.1502⁎⁎⁎	-26.7886⁎⁎⁎	6,778,284.61⁎⁎⁎	32.4824⁎⁎⁎	32.2429⁎⁎⁎
TUN	-7.6268⁎⁎⁎	-27.0036⁎⁎⁎	4,319,178.57⁎⁎⁎	29.3524⁎⁎⁎	29.0182⁎⁎⁎
USA	-5.3869⁎⁎⁎	-35.7915⁎⁎⁎	3,439,690.63⁎⁎⁎	19.9327⁎⁎	19.7082⁎⁎
VNM	-12.4014⁎⁎⁎	-25.9440⁎⁎⁎	59,654.49⁎⁎⁎	32.9342⁎⁎⁎	32.6555⁎⁎⁎

Notes: This table reports the results of Augmented Dickey–Fuller (ADF) test, Phillips–Perron (PP) test, Jarque–Bera (JB) and ARCH test (Ljung–Box statistics and Q-statistics). Panel A reports the results of stationarity, normal distribution and ARCH effect tests of Stock return series (SlogR), and Panel B reports those test results for the Change rate series of COVID–19 restriction index (RlogR) for each country, respectively. The Augmented Dickey–Fuller (ADF) test and the Phillips–Perron (PP) test are used to check the stationarity, the Jarque–Bera (JB) statistics which check the normality, the Ljung–Box (LB) statistics and the Q-statistics are to test ARCH effects ***, **, and * denote the significance level on 1%, 5% and 10%, respectively. As seen in this table, all the statistics of ADF test and PP test are significant at 1% level, highlighting the stationarity of time series. All the JB statistics are significant at 1% level, highlighting non-normal distributions. Most of the LB statistics and Q-statistics are statistically significant, implying significant heteroskedastic effects. These test results suggest that the GARCH–type model with t-distribution is appropriate for capturing the conditional volatilities of stock markets and COVID–19 stringency indexes.

Appendix A3. Estimation results of Mean-GARCH (1,1) models with t-distribution

ISO3	μ	ω	α	β	$\upsilon$	AIC	BIC	LL	Nobs
Panel A: Estimated results of Mean-GARCH (1,1) model with t-distribution for stock returns (SlogR)
ARG	0.175⁎⁎ (2.4)	0.474⁎⁎ (2.5)	0.172⁎⁎⁎ (3.3)	0.751⁎⁎⁎ (12.1)	6.306⁎⁎⁎ (4.1)	3050	3072	-1520	695
AUS	0.085⁎⁎⁎ (2.7)	0.067⁎⁎ (2.0)	0.225⁎⁎⁎ (3.1)	0.740⁎⁎⁎ (9.5)	6.163⁎⁎⁎ (4.8)	1937	1960	-964	695
AUT	0.107⁎⁎ (2.5)	0.096 (1.4)	0.139⁎⁎ (2.0)	0.837⁎⁎⁎ (10.5)	4.650⁎⁎⁎ (5.5)	2463	2485	-1226	695
BEL	0.030 (0.8)	0.026 (0.5)	0.067 (0.5)	0.916⁎⁎⁎ (6.0)	5.353⁎⁎⁎ (4.8)	2201	2224	-1096	695
BRA	0.054 (1.1)	0.191* (1.7)	0.164⁎⁎⁎ (3.0)	0.754⁎⁎⁎ (8.4)	9.816⁎⁎⁎ (2.9)	2470	2493	-1230	695
KHM	-0.043⁎⁎⁎ (-3.9)	0.060⁎⁎⁎ (2.7)	0.641⁎⁎⁎ (5.7)	0.359⁎⁎⁎ (2.9)	2.942⁎⁎⁎ (17.3)	1183	1205	-586	695
CAN	0.082⁎⁎⁎ (3.3)	0.088⁎⁎⁎ (2.8)	0.360⁎⁎⁎ (3.7)	0.598⁎⁎⁎ (6.7)	6.068⁎⁎⁎ (4.2)	1778	1801	-884	695
CHL	0.039 (0.8)	0.386⁎⁎⁎ (2.8)	0.202⁎⁎⁎ (2.7)	0.658⁎⁎⁎ (6.6)	5.004⁎⁎⁎ (5.0)	2489	2512	-1240	695
COL	0.009 (0.3)	0.285⁎⁎ (2.3)	0.384⁎⁎⁎ (2.9)	0.555⁎⁎⁎ (4.1)	3.690⁎⁎⁎ (7.0)	2188	2211	-1089	695
HRV	0.043⁎⁎ (2.4)	0.074⁎⁎ (2.3)	0.317⁎⁎⁎ (2.6)	0.649⁎⁎⁎ (6.1)	3.018⁎⁎⁎ (6.8)	1266	1289	-628	695
CZE	0.070⁎⁎ (2.5)	0.093⁎⁎ (2.3)	0.252⁎⁎⁎ (3.7)	0.701⁎⁎⁎ (9.3)	4.895⁎⁎⁎ (5.5)	1918	1941	-954	695
DNK	0.103⁎⁎ (2.4)	0.107 (1.5)	0.118⁎⁎ (2.3)	0.812⁎⁎⁎ (9.0)	11.530⁎⁎⁎ (2.7)	2208	2230	-1099	695
ECU	-0.003 (-0.7)	0.008 (1.0)	0.050⁎⁎⁎ (4.3)	0.950⁎⁎⁎ (21.1)	2.107⁎⁎⁎ (33.7)	159	182	-74	695
EST	0.062⁎⁎⁎ (2.7)	0.072⁎⁎⁎ (3.5)	0.371⁎⁎⁎ (4.9)	0.598⁎⁎⁎ (10.4)	4.522⁎⁎⁎ (6.0)	1567	1589	-778	695
FIN	0.084⁎⁎ (2.3)	0.061* (1.9)	0.138⁎⁎⁎ (2.8)	0.834⁎⁎⁎ (15.3)	5.522⁎⁎⁎ (5.1)	2173	2196	-1082	695
FRA	0.107⁎⁎⁎ (2.7)	0.066 (1.5)	0.191⁎⁎⁎ (2.8)	0.809⁎⁎⁎ (12.8)	4.155⁎⁎⁎ (5.8)	2265	2287	-1127	695
DEU	0.077⁎⁎ (2.2)	0.066 (1.6)	0.185⁎⁎⁎ (3.5)	0.815⁎⁎⁎ (16.4)	3.859⁎⁎⁎ (7.5)	2302	2325	-1146	695
GRC	0.099⁎⁎ (2.4)	0.155* (1.7)	0.163⁎⁎⁎ (2.7)	0.811⁎⁎⁎ (12.6)	3.209⁎⁎⁎ (6.5)	2363	2386	-1176	695
HUN	0.067 (1.5)	0.264⁎⁎⁎ (4.0)	0.234⁎⁎⁎ (4.6)	0.664⁎⁎⁎ (13.4)	5.716⁎⁎⁎ (4.7)	2393	2416	-1192	695
IND	0.139⁎⁎⁎ (3.7)	0.033 (1.6)	0.094⁎⁎ (2.5)	0.887⁎⁎⁎ (22.1)	5.454⁎⁎⁎ (5.4)	2122	2145	-1056	695
IDN	0.074⁎⁎ (2.5)	0.098⁎⁎⁎ (3.1)	0.211⁎⁎⁎ (4.1)	0.700⁎⁎⁎ (11.1)	7.024⁎⁎⁎ (4.2)	1859	1882	-925	695
IRL	0.043 (1.0)	0.061 (1.3)	0.099⁎⁎ (2.2)	0.882⁎⁎⁎ (17.1)	5.196⁎⁎⁎ (5.1)	2431	2454	-1211	695
JPN	0.049 (1.3)	0.248 (1.1)	0.157 (1.5)	0.648⁎⁎ (2.4)	9.617⁎⁎⁎ (2.8)	2091	2114	-1041	695
KAZ	0.071⁎⁎⁎ (2.9)	0.099* (1.8)	0.197⁎⁎ (2.4)	0.643⁎⁎⁎ (4.1)	5.802⁎⁎⁎ (5.2)	1499	1522	-745	695
LVA	0.025 (1.0)	0.258⁎⁎⁎ (3.3)	0.335⁎⁎⁎ (3.2)	0.508⁎⁎⁎ (5.3)	3.218⁎⁎⁎ (7.2)	1760	1782	-875	695
LTU	0.050⁎⁎ (2.1)	0.070⁎⁎⁎ (3.3)	0.390⁎⁎⁎ (4.7)	0.596⁎⁎⁎ (9.7)	4.169⁎⁎⁎ (6.4)	1543	1565	-766	695
LUX	0.071 (1.3)	0.138 (1.0)	0.109 (1.3)	0.840⁎⁎⁎ (6.5)	7.189⁎⁎⁎ (4.0)	2562	2585	-1276	695
MYS	-0.026 (-1.0)	0.048* (1.7)	0.102⁎⁎⁎ (2.7)	0.831⁎⁎⁎ (12.2)	6.317⁎⁎⁎ (4.6)	1644	1666	-817	695
MLT	-0.048⁎⁎ (-2.3)	0.108⁎⁎ (2.4)	0.190⁎⁎ (2.5)	0.683⁎⁎⁎ (6.9)	3.207⁎⁎⁎ (7.7)	1416	1439	-703	695
MEX	0.017 (0.4)	0.047 (0.9)	0.093 (1.6)	0.874⁎⁎⁎ (9.7)	22.321 (1.3)	2164	2187	-1077	695
MNG	-0.035 (-1.2)	0.172 (1.3)	0.376⁎⁎ (2.1)	0.594⁎⁎⁎ (3.1)	3.640⁎⁎⁎ (7.0)	1965	1987	-977	695
PAK	0.053* (1.9)	0.049⁎⁎⁎ (2.8)	0.118⁎⁎⁎ (3.5)	0.832⁎⁎⁎ (22.8)	4.678⁎⁎⁎ (5.6)	1762	1784	-876	695
PER	0.025 (0.6)	0.348⁎⁎⁎ (3.1)	0.155⁎⁎⁎ (2.8)	0.698⁎⁎⁎ (9.1)	3.938⁎⁎⁎ (6.7)	2348	2371	-1169	695
POL	0.053 (1.2)	0.118⁎⁎ (2.0)	0.154⁎⁎⁎ (2.9)	0.808⁎⁎⁎ (13.8)	4.651⁎⁎⁎ (5.3)	2354	2377	-1172	695
PRT	0.070 (1.5)	0.282⁎⁎ (2.5)	0.145⁎⁎⁎ (3.1)	0.696⁎⁎⁎ (7.6)	12.255⁎⁎ (2.5)	2327	2350	-1159	695
ROU	0.100⁎⁎⁎ (3.6)	0.070⁎⁎ (2.1)	0.184⁎⁎⁎ (3.4)	0.775⁎⁎⁎ (12.3)	3.974⁎⁎⁎ (6.5)	1822	1845	-906	695
RUS	0.162⁎⁎⁎ (2.7)	0.570 (1.6)	0.313⁎⁎⁎ (3.1)	0.654⁎⁎⁎ (5.9)	3.488⁎⁎⁎ (6.5)	2900	2923	-1445	695
SGP	0.024 (0.9)	0.071⁎⁎⁎ (2.7)	0.172⁎⁎⁎ (3.4)	0.744⁎⁎⁎ (11.6)	5.950⁎⁎⁎ (4.5)	1703	1725	-846	695
SVK	0.004 (0.5)	0.258 (1.4)	0.739⁎⁎⁎ (3.2)	0.261⁎⁎ (2.5)	2.319⁎⁎⁎ (30.7)	1401	1423	-695	695
SVN	0.088⁎⁎⁎ (3.6)	0.125⁎⁎ (2.1)	0.319⁎⁎ (2.5)	0.601⁎⁎⁎ (4.1)	3.786⁎⁎⁎ (6.3)	1675	1698	-833	695
KOR	0.051 (1.4)	0.139⁎⁎⁎ (2.6)	0.232⁎⁎⁎ (3.7)	0.682⁎⁎⁎ (8.9)	13.411⁎⁎ (2.3)	2136	2159	-1063	695
ESP	0.035 (0.9)	0.110⁎⁎ (2.1)	0.185⁎⁎⁎ (3.3)	0.781⁎⁎⁎ (15.3)	5.011⁎⁎⁎ (5.0)	2312	2335	-1151	695
LKA	0.153⁎⁎⁎ (3.6)	0.075 (1.5)	0.308⁎⁎⁎ (4.2)	0.692⁎⁎⁎ (7.7)	4.341⁎⁎⁎ (7.4)	2170	2193	-1080	695
CHE	0.070⁎⁎ (2.5)	0.058* (2.0)	0.215⁎⁎⁎ (2.9)	0.755⁎⁎⁎ (10.0)	4.682⁎⁎⁎ (5.7)	1794	1817	-892	695
TUN	0.038⁎⁎⁎ (3.1)	0.060⁎⁎⁎ (2.6)	0.462⁎⁎⁎ (2.9)	0.256 (1.2)	4.276⁎⁎⁎ (6.1)	545	568	-268	695
TCA	0.307⁎⁎⁎ (8.0)	0.520* (1.9)	0.296⁎⁎ (2.0)	0.618⁎⁎⁎ (4.8)	2.843⁎⁎⁎ (8.1)	2333	2356	-1162	695
ARE	0.072⁎⁎ (2.4)	0.066 (1.1)	0.249⁎⁎ (2.1)	0.719⁎⁎⁎ (4.9)	5.110⁎⁎⁎ (4.4)	1813	1836	-902	695
GBR	0.056* (1.9)	0.100* (1.8)	0.276⁎⁎⁎ (2.6)	0.718⁎⁎⁎ (7.3)	3.783⁎⁎⁎ (6.7)	2035	2058	-1012	695
USA	0.115⁎⁎⁎ (3.5)	0.060⁎⁎ (2.3)	0.270⁎⁎⁎ (3.5)	0.731⁎⁎⁎ (11.1)	6.586⁎⁎⁎ (4.2)	2173	2195	-1081	695
VNM	0.319⁎⁎⁎ (7.3)	0.436⁎⁎⁎ (4.0)	0.406⁎⁎⁎ (4.2)	0.514⁎⁎⁎ (8.0)	4.130⁎⁎⁎ (6.6)	2446	2469	-1218	695

ISO3	μ	ω	α	β	$\upsilon$	AIC	BIC	LL	Nobs
Panel B: Estimated results of Mean-GARCH (1,1) model for the change rate of COVID-19 stringency index (RlogR)
ARG	0.452⁎⁎⁎ (20.2)	1.904⁎⁎⁎ (24.6)	0.010 (1.1)	0.779⁎⁎⁎ (82.2)	2.982⁎⁎⁎ (166.5)	2576	2598	-1283	695
AUS	-0.148⁎⁎⁎ (-17.3)	0.037⁎⁎⁎ (10.4)	0.338* (1.9)	0.000 (1.0)	2.050⁎⁎⁎ (3873)	3599	3622	-1795	695
AUT	-0.020⁎⁎⁎ (-13.8)	0.210⁎⁎⁎ (11.0)	0.000 (0.0)	0.647⁎⁎⁎ (27.0)	2.050⁎⁎⁎ (517.0)	260	283	-125	695
BEL	0.002⁎⁎⁎ (6.9)	0.004⁎⁎⁎ (10.3)	0.173⁎⁎⁎ (11.7)	0.258⁎⁎⁎ (5.0)	2.050⁎⁎⁎ (886.6)	-2126	-2103	1068	695
BRA	-0.020 (-0.9)	0.024 (1.1)	0.000 (0.0)	0.421⁎⁎⁎ (3.2)	2.050⁎⁎⁎ (77.8)	-688	-665	349	695
KHM	0.004⁎⁎⁎ (25.9)	0.001⁎⁎⁎ (3.4)	0.000⁎⁎⁎ (3.7)	0.000 (0.8)	2.050⁎⁎⁎ (130.6)	-3082	-3059	1546	695
CAN	-0.097⁎⁎⁎ (-18.2)	0.190⁎⁎⁎ (4.8)	0.177⁎⁎⁎ (3.7)	0.112 (1.2)	2.140⁎⁎⁎ (373.1)	600	623	-295	695
CHL	0.056⁎⁎⁎ (4.0)	3.132⁎⁎⁎ (12.6)	0.004 (1.3)	0.000 (0.0)	7.469⁎⁎⁎ (7.4)	2342	2365	-1166	695
COL	-0.011⁎⁎⁎ (-10.6)	0.024⁎⁎⁎ (11.1)	0.000 (0.8)	0.002 (0.5)	3.668⁎⁎⁎ (14.5)	-272	-249	141	695
HRV	-0.146⁎⁎⁎ (-25.8)	1.055⁎⁎⁎ (15.8)	0.000 (0.3)	0.945⁎⁎⁎ (373.5)	2.050⁎⁎⁎ (1069)	1993	2016	-992	695
CZE	1498.453 (0.1)	0.003⁎⁎⁎ (7.8)	0.000 (0.0)	0.000 (0.0)	2.050⁎⁎⁎ (7.8)	44,706	44,728	-22,348	695
DNK	-0.013⁎⁎⁎ (-13.3)	0.289⁎⁎⁎ (11.3)	0.001 (1.1)	0.000 (0.0)	2.051⁎⁎⁎ (509.0)	-472	-449	241	695
ECU	-0.096⁎⁎⁎ (-23.0)	0.381⁎⁎⁎ (17.5)	0.007⁎⁎⁎ (4.2)	0.433⁎⁎⁎ (12.6)	2.369⁎⁎⁎ (315.2)	939	962	-465	695
EST	-0.000 (-0.8)	0.000 (0.1)	1.000⁎⁎⁎ (3.1)	0.000 (0.0)	2.050⁎⁎⁎ (90.7)	-2620	-2597	1315	695
FIN	0.553⁎⁎⁎ (32.7)	0.524⁎⁎⁎ (8.8)	0.091⁎⁎ (2.0)	0.018 (1.1)	3.841⁎⁎⁎ (10.3)	2391	2414	-1191	695
FRA	-0.009* (-1.8)	0.047 (1.1)	0.003* (1.7)	0.030 (0.6)	2.114⁎⁎⁎ (25.5)	-496	-473	253	695
DEU	0.734⁎⁎⁎ (20.3)	9.754⁎⁎⁎ (20.5)	0.443 (1.3)	0.000 (0.0)	3.469⁎⁎⁎ (87.0)	3011	3034	-1501	695
GRC	0.029⁎⁎⁎ (5.9)	2.180⁎⁎⁎ (16.9)	0.000 (0.0)	0.000 (0.0)	3.183⁎⁎⁎ (78.0)	2099	2122	-1045	695
HUN	-0.001⁎⁎⁎ (-11.0)	0.000⁎⁎⁎ (10.9)	0.001⁎⁎⁎ (3.2)	0.000 (0.1)	2.050⁎⁎⁎ (1161)	-3915	-3892	1963	695
IND	-761,800⁎⁎⁎ (-39.9)	0.247 (0.1)	0.000⁎⁎⁎ (3.3)	0.001 (0.1)	2.051⁎⁎⁎ (5368)	44,025	44,047	-22,007	695
IDN	0.032⁎⁎⁎ (15.3)	0.202⁎⁎⁎ (17.2)	0.010⁎⁎⁎ (5.0)	0.842⁎⁎⁎ (95.2)	2.063⁎⁎⁎ (983.8)	449	472	-219	695
IRL	-0.056⁎⁎⁎ (-6.1)	3.718⁎⁎⁎ (3.3)	0.920 (0.3)	0.065 (0.1)	3.537⁎⁎⁎ (43.9)	2565	2587	-1277	695
JPN	-0.006⁎⁎⁎ (-17.4)	0.016⁎⁎⁎ (20.1)	0.008⁎⁎⁎ (9.1)	0.401⁎⁎⁎ (9.3)	2.051⁎⁎⁎ (1367)	-2043	-2021	1027	695
KAZ	-0.022 (-0.6)	0.371 (0.5)	0.000 (0.6)	0.005 (0.2)	12.036 (0.9)	2294	2316	-1142	695
LVA	-0.369⁎⁎⁎ (-2.8)	0.017 (0.7)	0.040⁎⁎⁎ (11.6)	0.933⁎⁎⁎ (216.6)	5.363⁎⁎⁎ (10.3)	3244	3266	-1617	695
LTU	-0.041⁎⁎⁎ (-13.2)	0.494⁎⁎⁎ (17.9)	0.009⁎⁎⁎ (4.8)	0.211⁎⁎⁎ (4.1)	2.427⁎⁎⁎ (324.7)	960	982	-475	695
LUX	0.010⁎⁎⁎ (9.4)	1.041⁎⁎⁎ (24.8)	0.048⁎⁎⁎ (4.2)	0.487⁎⁎⁎ (21.5)	2.250⁎⁎⁎ (478.7)	1164	1187	-577	695
MYS	0.555⁎⁎⁎ (7.1)	6.484⁎⁎⁎ (8.9)	0.149⁎⁎⁎ (3.2)	0.021 (0.1)	5.305⁎⁎⁎ (23.1)	3053	3076	-1521	695
MLT	-0.000⁎⁎⁎ (-3.7)	0.001⁎⁎⁎ (19.5)	0.000⁎⁎⁎ (3.7)	0.000 (0.1)	2.050⁎⁎⁎ (1033)	-3999	-3976	2004	695
MEX	0.110⁎⁎⁎ (7.5)	11.12⁎⁎⁎ (28.7)	0.025 (1.4)	0.018 (0.1)	2.209⁎⁎⁎ (445.8)	2051	2074	-1021	695
MNG	0.005⁎⁎⁎ (5.2)	0.881⁎⁎⁎ (18.8)	0.004⁎⁎ (2.4)	0.121 (0.7)	2.247⁎⁎⁎ (412.9)	875	898	-433	695
PAK	-0.009⁎⁎⁎ (-3.4)	0.030 (0.9)	0.000 (0.0)	0.469⁎⁎ (2.2)	2.050⁎⁎⁎ (61.5)	-694	-671	352	695
PER	-10.712⁎⁎⁎ (-33.6)	0.000 (0.0)	0.000 (0.0)	0.998⁎⁎⁎ (7147)	4.802⁎⁎⁎ (18.3)	5722	5745	-2856	695
POL	162.006⁎⁎⁎ (436.2)	0.883⁎⁎⁎ (10.0)	0.000⁎⁎⁎ (131)	0.001 (0.4)	2.396⁎⁎⁎ (5107)	22,204	22,226	-11,097	695
PRT	-0.018⁎⁎⁎ (-18.2)	0.021⁎⁎⁎ (7.0)	0.002 (0.6)	0.005 (1.1)	2.061⁎⁎⁎ (402.1)	-292	-270	151	695
ROU	-279.560⁎⁎⁎ (-150.4)	2.020⁎⁎⁎ (5.4)	0.000⁎⁎⁎ (52.7)	0.000 (0.2)	2.442⁎⁎⁎ (2156)	23,873	23,896	-11,932	695
RUS	-0.319⁎⁎⁎ (-18.5)	0.394⁎⁎ (2.1)	0.012* (1.8)	0.024 (0.4)	2.975⁎⁎⁎ (5.7)	1865	1888	-928	695
SGP	-342.853⁎⁎⁎ (-250.8)	3.788 (0.1)	0.999⁎⁎⁎ (616)	0.001 (0.2)	2.400⁎⁎⁎ (3443)	11,277	11,299	-5633	695
SVK	-0.059⁎⁎⁎ (-10.3)	2.720⁎⁎⁎ (15.7)	0.299⁎⁎⁎ (4.0)	0.037 (0.8)	3.171⁎⁎⁎ (85.9)	2332	2355	-1161	695
SVN	-0.006⁎⁎⁎ (-6.4)	0.064⁎⁎⁎ (4.9)	0.000 (0.6)	0.000 (0.0)	2.050⁎⁎⁎ (225.2)	-745	-722	377	695
KOR	-0.218⁎⁎⁎ (-9.9)	0.948⁎⁎⁎ (10.9)	0.101⁎⁎⁎ (8.9)	0.821⁎⁎⁎ (65.0)	6.297⁎⁎⁎ (16.5)	3113	3136	-1552	695
ESP	0.037⁎⁎⁎ (4.4)	95.32⁎⁎⁎ (33.6)	0.000 (0.0)	0.000 (0.0)	2.050⁎⁎⁎ (1418)	2437	2460	-1213	695
LKA	-0.006⁎⁎⁎ (-18.2)	0.002⁎⁎⁎ (7.3)	0.000 (1.4)	0.000 (0.0)	2.050⁎⁎⁎ (437.6)	-2187	-2164	1098	695
CHE	-0.008⁎⁎⁎ (-16.6)	0.002⁎⁎⁎ (2.9)	0.011 (0.7)	0.000 (0.3)	4.199⁎⁎⁎ (4.6)	-1041	-1018	525	695
TUN	0.594 (1.6)	12.920* (1.7)	0.020 (0.8)	0.000 (0.0)	58.572⁎⁎⁎ (4.3)	4315	4338	-2152	695
TCA	-0.037⁎⁎⁎ (-13.4)	1.305⁎⁎⁎ (19.6)	0.026⁎⁎⁎ (3.5)	0.354⁎⁎⁎ (8.1)	2.213⁎⁎⁎ (369.8)	1293	1316	-641	695
ARE	-0.008⁎⁎ (-2.5)	0.154 (1.1)	0.004 (0.7)	0.001 (0.0)	9.780⁎⁎⁎ (3.3)	2399	2421	-1194	695
GBR	-0.251⁎⁎⁎ (-12.6)	1.878⁎⁎⁎ (14.9)	0.302 (0.5)	0.000 (0.0)	3.382⁎⁎⁎ (72.6)	2097	2120	-1044	695
USA	35,481⁎⁎⁎ (290.6)	0.004 (0.1)	0.000⁎⁎⁎ (520)	0.001 (0.9)	2.383⁎⁎⁎ (1848)	32,283	32,305	-16,136	695
VNM	0.000⁎⁎⁎ (7.1)	0.000⁎⁎⁎ (18.4)	0.078 (1.1)	0.000 (0.0)	2.096⁎⁎⁎ (878.6)	-4174	-4152	2092	695

Notes: This table reports the estimation results of Mean-GARCH (1,1) models with t-distribution, which is specified in Eq. (4). Panel A shows the estimated results for stock returns of each country, and Panel B shows the estimated results for the change rate of COVID-19 stringency index of each country. See Table 1 for details of the ISO3 code of given countries. μ,ω,α and β are parameters of the GARCH (1,1) model.$\upsilon$ is the parameter for t-distribution. T-statistics in parentheses. ***, **, and * denote the significance level on 1%, 5% and 10%, respectively.

Data availability

• Data will be made available on request.