key: cord-0686625-txo1gngl
authors: Szczygielski, Jan Jakub; Bwanya, Princess Rutendo; Charteris, Ailie; Brzeszczyński, Janusz
title: The only certainty is uncertainty: An analysis of the impact of COVID-19 uncertainty on regional stock markets
date: 2021-01-26
journal: Financ Res Lett
DOI: 10.1016/j.frl.2021.101945
sha: 1904be17abaf63e00d302a306dba0649acdc31b9
doc_id: 686625
cord_uid: txo1gngl

Uncertainty surrounding COVID-19 is widespread. We investigate the timing and quantify the impact of COVID-19 related uncertainty on returns and volatility for regional market aggregates using ARCH/GARCH models. Drawing upon economic psychology, COVID-19 related uncertainty is measured by searches for information as reflected by Google Trends. Asian markets are more resilient than others. Latin American markets are most impacted in terms of returns and volatility. For most regions, there is evidence of an increasing impact of COVID-19 related uncertainty which dissipates as the crisis evolves. We confirm that Google Trends capture uncertainty by comparing this measure against alternative uncertainty measures.

While several pandemics and serious disease outbreaks have occurred in the past, such as the Spanish flu (1918) , SARS (2003) , MERS (2012) 1 and Ebola (2014) , the novel coronavirus outbreak of 2019-2020 ranks amongst the most severe and widespread, with infections recorded in more than 200 countries (World Health Organisation (WHO), 2020). The emergence of has resulted in a global economic crisis coupled with severe stock market declines. Prior studies show that not only are financial markets negatively impacted by diseases and crises in general, but that the intensity and timing of impact differs across countries (Nippani and Washer, 2004; McTier et al., 2013; Bekaert et al., 2014) . Ichev and Marinč (2018) report that Ebola outbreaks had a more significant impact on companies that had operations in, or were geographically nearer to, Ebola origins (West Africa). Claessens et al. (2010) documented that during the 2007-2008 financial crisis, countries with closer ties to the United States' (US) financial system or direct exposure to assetbacked securities were the first to be affected.

Research on the differential effects of COVID-19 across countries has identified varying intensities and timing. Liu et al. (2020) observe that Asian financial markets experienced an immediate downturn when the outbreak occurred. The impact on US and European markets was delayed, occurring several days after outbreaks of the virus in South Korea and Italy, 2 and less severe. Similarly, Gormsen and Koijen (2020) show that only once COVID-19 spread to Italy, Iran and South Korea, did US and German stock markets decline sharply. Gunay (2020) reports a structural break in volatility for Chinese stock returns earlier (30 th January 2020) than other countries. 3 Ru et al. (2020) find that market reactions to early COVID-19 outbreaks were more immediate and substantial in countries that suffered from SARS in 2003. Gerding et al. (2020) document that stock markets in countries with higher debt-to-gross domestic product ratios were more impacted.

Uncertainty surrounding COVID-19 is widespread, both with respect to the evolution of the disease itself and its economic impact (McKibbin and Fernando, 2020) . Moreover, COVID-19 related uncertainty has impacted both returns and volatility in the US (Baig et al., 2020; Bretscher et al., 2020; Ramelli and Wagner, 2020) and internationally (Liu, 2020; Papadamou et al., 2020) . However, no study has examined the differential impact of COVID-19 related uncertainty on regional markets around the world and the timing of these effects. We quantify the differential impact of COVID-19 related uncertainty on returns and variance for six regional market aggregates using the ARCH/GARCH framework and structural change analysis.

Drawing from economic psychology, which proposes that individuals respond to uncertainty about specific events by searching more intensively for relevant information (Liemieux and Peterson, 2011;  1 Severe acute respiratory syndrome and middle east respiratory syndrome respectively. 2 19 and 21 February 2020 respectively. 3 The US, Italy, Spain, Turkey and the United Kingdom. Dzielinski, 2012; Castelnuovo and Tran, 2017; Bontempi et al., 2019) , we measure uncertainty using Google Trends search data for terms related to . We contribute to the burgeoning literature on the impact of COVID-19 on financial markets. To the best of our knowledge, this is the first study that investigates the relationship between uncertainty reflected by Google search trends and COVID-19 for regional market aggregates. We find that returns for all regions are negatively impacted by global COVID-19 uncertainty and that COVID-19 uncertainty has volatility triggering effects for all regions with the exception of Arab markets. Furthermore, we find that a number of regions show a weakening of the impact of COVID-19 uncertainty as the crisis evolved. We go on to confirm that Google Trends are a proxy for uncertainty which drives returns and triggers volatility.

Our primary data sample spans 1 January 2019 to 19 June 2020. 4 For regional markets, the MSCI All Country (AC) Asia, AC Europe, Emerging Frontier Markets (EMF) Africa, Emerging Markets (EM) Latin America, North America and Arabian Markets indices are used. Returns are defined as logarithmic differences in index levels. Data is of a daily frequency and is stated according to MSCI's local currency methodology, representing performance unimpacted by foreign exchange rate movements.

Following an analysis of Google Trends, we identify nine COVID-19 related terms associated with high search volumes globally. These are: "coronavirus", "COVID19", "COVID 19", "COVID", "COVID-19", "SARS-CoV-2", "SARS-COV", "severe acute respiratory syndrome-related coronavirus" and "severe acute respiratory syndrome". We construct a search term index 5 by combining search trends for the terms above. Individual index values are added and the sum is divided 4 1 December 2019 is chosen as the start of the COVID-19 crisis as this was the day on which the first case was reported . However, we use a longer sample for estimation purposes. 5 Data obtained from Google Trends is the sum of the scaled total number of searches between 0 to 100 based upon a topic's proportion to all searches on all topics. by nine. The highest value is adjusted to 100, with remaining values adjusted accordingly relative to this base. Index values are then differenced ( Figure 1A , Appendix).

This figure plots levels in the combined COVID-19 search term index created from Google Trends search volumes for nine COVID-19 related search terms over the period 1 December 2019 to 19 June 2020. Levels of search volumes for individual COVID-19 related terms are also plotted.

We apply the ARCH/GARCH framework to measure the impact of changes in search volumes on both returns and conditional variance, a proxy for risk (Brzeszczyński and Kutan, 2015) . We begin with an ARCH(p) model and proceed to estimate an GARCH(p,q) model if the ARCH(p) specification exhibits residual heteroscedasticity. We also consider the IGARCH(p,q) model if ARCH and GARCH parameters sum to unity (Engle and Bollerslev, 1986) . Following preliminary specification testing, the following models are proposed: (1), is specified in the "mean" row. ARCH(p), GARCH(p,q) and IGARCH(p,q) specifications, equations (2a)/(2b)/(2c) respectively, follow after the "ARCH/GARCH" row.

Specification Mean: ∑

(1) ARCH/GARCH: June 2020, respectively. A residual market factor, derived from returns on the MSCI AC World Market index, is included to address potential underspecification (Burmeister and McElroy, 1991) . Additionally, a factor analytically derived factor set, ∑ , is incorporated into equation (1) to account for influences that may not be reflected by . Factors comprising the factor analytic augmentation, accounting for both contemporaneous and lagged relationships, are derived from regional return series and are then adjusted for the impact of and (Szczygielski et al., 2020) . 6 For parsimony, only significant proxy factors are retained. Finally, autoregressive terms, of order identified from an analysis of a residual correlogram are included to address remaining autocorrelation if required. To identify periods for which the impact of differs, breakpoints are identified using the Bai-Perron test (Carlson et al., 2000) . If breakpoints are detected, the variable, together with associated coefficients and shift dummies in equations (1) and (2a)/(2b)/(2c), are replaced with ∑ and ∑ respectively, with taking on a value of one or zero otherwise for segment between breakpoints. The equations are first estimated using maximum likelihood estimation (MLE). If residuals are non-normal, they are re-estimated using quasi-maximum likelihood (QML) estimation with Bollerslev-Wooldridge standard errors and covariance (Fan et al., 2014) .

Panel A, Table 3 reports coefficients on in the conditional mean ( ) and Panel B

reports the impact of on the conditional variance ( . 7 The results in Panel A, Table 3 indicate that returns for all regions are negatively and significantly impacted by . The results

in Panel B indicate that coefficients on in the respective ARCH/GARCH models, , are positive and statistically significant for five regions. The negative impact of on returns can be attributed to a combination of lower expected cash flows and heightened risk aversion. The adverse economic effects of COVID-19 uncertainty are likely to be associated with declining expected cash flows to firms (Ramelli & Wagner, 2020) . Also, heightened risk aversion attributable to uncertainty surrounding the pandemic means that investors will require a higher risk premium which 6 Szczygielski et al. (2020) show that a residual market factor may be insufficient to ensure residual correlation matrix diagonality, implying that a model omits factors with a systematic (common) impact. The inclusion of a factor analytic augmentation is shown to result in a diagonal residual matrix. 7 All estimation procedures converge and residuals are free of ARCH effects and serial correlation.

is reflected in the forward looking discount rate (Andrei & Hasler, 2014; Smales, 2020; Cochrane, 2018) . Together, lower expected cash flows and a higher discount rate translate into lower stock prices. 8

Although returns in North America are negatively impacted ( of -0.003417(3 rd )), this region does not show significant volatility triggering effects. However, the results in Panel B, Table 4 paint a different picture suggesting that North American markets experienced delayed volatility triggering effects. Similarly, while returns in Europe are also impacted ( of -0.003459(2 nd )), volatility triggering effects are relatively low ( of 0.1460(4 th )). Arab markets do not appear to experience heightened volatility associated with , although returns are impacted ( of -0.00188(5 th )). The lack of volatility triggering effects is surprising, given the economic dependency on oil of Arab markets and the consequent uncertainty surrounding their macroeconomic prospects (Ashraf, 2020) . However, an analysis of realized variance suggests that Arab markets showed extreme, but short-lived, heightened volatility around 7 to 9 March 2020. These dates coincide with COVID-19 cases surpassing 100 000 and a call by the WHO for more stringent actions to control the spread of COVID-19 (WHO, 2020). While may not be significant, forecasted conditional variance captures this volatility spike ( Figure 7A , Appendix).

Asian markets are relatively resilient to COVID-19 uncertainty of -0.001814(6 th ) and of 0.1300(5 th ) respectively). This may be attributable to experience that Asian countries have in dealing with pandemics (SARS and MERS outbreaks) (Lu et al., 2020; Wang and Enilov, 2020) . While these results differ from those of Liu et al. (2020) and Ru et al. (2020) , who report that Asian markets were severely impacted by COVID-19 infection numbers, this finding demonstrates the varying effect of COVID-19 uncertainty relative to infection numbers on stock markets.

Finally, the substantial impact of on returns and volatility in African and Latin American markets ( s of -0.00314(4 th ) and -0.003625(1 st ) and, s of 0.2680(2 rd ) and 0.5480(1 st ))

can be attributed to risk aversion in relation to developing markets in times of crisis and spillovers from developed markets (Frank and Hesse, 2009; Bekaert et al., 2014) . Both regions comprise two of the larger and more developed stock markets in the world, the Johannesburg Stock Exchange (JSE) (19 th ) and the Brazilian BM&F Bovespa (20 th ) (Haqqi, 2020) , which are highly integrated with global markets (Babu et al., 2016; Nashier, 2015) and therefore likely to readily reflect global developments (Szczygielski and Chipeta, 2015) . 9 In contrast, Arab markets, while comprising developing countries, 8 We would like to thank an anonymous reviewer for a comment relating to this issue as well as for other comments which helped in improving this paper. 9 Spearman rank-order correlations for realized volatility over the pre-COVID-19 and COVID-19 periods are compared across respective regions (see Table 2A , Appendix). All regions show stronger and (now always) significant positive correlation over the COVID-19 period. Latin American and African markets are now significantly and positively correlated with North American and Arab markets. Also, African markets are now correlated with Asian markets, which was not true prior to the COVID-19 period. Furthermore, correlation between African and European market volatility doubles. Volatility have been found to be less integrated globally (Marashdeh & Shrestha, 2010; Alotaibi & Mishra, 2017) , which is consistent with our findings in that they are less impacted by COVID-19 related uncertainty. Our results are generally consistent with previous studies on the differential impact of pandemics and crises on different regions (Claessens et al., 2010; Bekaert et al., 2014) .

in Arab markets is now significantly correlated with all regions although correlations become insignificant after adjusting for realized oil variance, which can be viewed as an important factor for this market. If volatility is interpreted as a proxy for information, this suggests that volatility in these markets now reflects spillovers from new sources of information (see Singh, Kumar and Pandey, 2010) . and variance for regional markets. Coefficients on in the conditional variance equation are scaled by 100 000. Panel A reports estimation results for the conditional mean, which also includes proxy factors derived from regional returns using factor analysis and adjusted for the impact of and . Panel B reports the results for the conditional variance. Values in brackets (…) rank the order of absolute impact according to the magnitude of the and coefficients. Panel C reports model diagnostics, with and being Ljung-Box tests statistics for joint serial correlation at the 1 st and 10 th orders. ARCH(1) and ARCH(10) are test statistics for the ARCH LM test for heteroscedasticity. Each model is estimated over the primary data period between 1 January 2019 and 19 June 2020 unless residuals show dependence structures in which case longer estimation periods are used. Pre-COVID-19 and COVID-19 periods are defined as 1 January 2019 to 30 November 2019 and 1 December 2019 to 19 June 2020 respectively. The asterisks ***,** and * indicate statistical significance at 1%, 5% and 10% levels of significance respectively.

is constructed from global Google Trends, we also consider value-weighted regional versions by replacing with 10 in Table 2 as an extension and robustness test. Table 4 show a similar pattern. Returns for all regions, with the exception of Arab markets 11 , are impacted negatively although to a lesser magnitude. For example, coefficients on for Latin and North America decrease to -0.000876 and -0.002296 respectively. The order of the magnitude of impact is approximately the same across regions although North American and Arab markets are now most and least impacted respectively. We attribute this effect to the dominance of US uncertainty. Specifically, uncertainty experienced by the US dominates the North American market, but also US uncertainty impacts all other regions (Chiang et al., 2015; Dimic et al., 2016; Smales, 2019) and hence with regional measures, US uncertainty is excluded resulting in a reduced impact. Volatility triggering effects associated with are also lower, with the exception of Latin America where is now associated with a coefficient of 1.0700 in the conditional variance. The generally greater impact of on both returns and conditional variance indicates that regional markets likely reflect not only regional uncertainty but also spillovers from the rest of the world (see discussion that follows). Importantly, it appears that global COVID-19 related uncertainty, as opposed to region or country-specific uncertainty, primarily matters most for stock markets and volatility (see Mumtaz and Mussom, 2019) . 12 This is broadly consistent with the 10 This measure is constructed by value weighting combined country specific search term indices for each country in each respective region. In constructing regional measures, we include all countries in each region, with the exception of Africa for which we include major constituents only (Egypt, South Africa, Kenya, Mauritius, Morocco, Nigeria and Tunisia) given the unavailability of country level market capitalizations for some minor constituents. 11 The positive relationship between Arab markets returns and becomes statistically insignificant and decreases further in absolute magnitude when the conditional variance is re-specified as a GARCH(1,1) process of 0.000114) and negative and statistically insignificant when the mean equation is re-estimated using least squares ( of -0.0000343). This suggests that the unexpected positive relationship between returns and the regional measure of COVID-19 uncertainty is not robust for Arab markets and is potentially attributable to the relatively noisier nature of the regional COVID-19 related uncertainty measure for Arab markets (see Figure 8A , Appendix). 12 Furthermore, using instead of while retaining original conditional mean and variance functional forms for comparative purposes generally results in lower log-likelihood values (with the exception of Europe and Arab markets, for which the log-likelihood values increase). For most regions, relying on global Google Trends to capture COVID-19 uncertainty produces a superior model fit (see Panel C, Table 1A ; Myung, 2003) . 1.0700*** (1 st ) 0.0421(5 th ) -0.0076 (6 th ) This table reports the abridged results for the impact of changes in regional COVID-19 related uncertainty as captured by Google Trends on the returns ( and variance ) for regional markets. Coefficients on in the conditional variance equation are scaled by 100 000. The asterisks, ***,** and * indicate statistical significance at 1%, 5% and 10% levels of significance, respectively. Figure 8A in the appendix presents a comparison of global and regional search term indices. Unabridged results are reported in Table 1A in the appendix. findings of Costola et al. (2020) that US, German, French, Spanish and UK stock markets respond more to Italian Google search trends than those in their own countries. Smales (2020) also finds that global search trends have a greater impact than regional search trends on the stock markets of the G20 countries. We conclude that, overall, the results of the analysis using are mostly qualitatively consistent with those for . 13 (Onali, 2020; Taylor, 2020) . Gunay (2020) also identified a structural break in volatility in North America and Europe in late February.

American countries under lockdown (Taylor, 2020) . We also identify a breakpoint for North America in January (20 January 2020) 15 and one for Latin America in mid-May (13 May 2020). infections on markets outside of Asia and is consistent with Ichev and Marinč's (2018) assertions that geographical proximity matters. It is only when these two regions become centres of the outbreak that volatility (and to a lesser extent returns) is most impacted in these markets. For returns in Latin 13 We investigate the direction of causality between regional returns and to determine whether market declines during the COVID-19 period contribute to COVID-19 related uncertainty or whether COVID-19 related uncertainty contributes to market declines. See Black (1976) and Bouchaud et al. (2001) for a discussion of the leverage effect which is concerned with the asymmetric relationship between volatility and returns. The results in Table 3A of the Appendix show that overwhelmingly Granger-causes regional market returns, with the exception of Africa for which there appears to be a bi-directional relationship. Although we do not undertake an extensive study of the intertemporal structure of returnrelationships, bi-directionality for this region continues at higher orders of lags although the F-statistic for the test of Granger-causality from returns on African markets to decreases as the number of lags is increased. 14 24 February 2020 in Europe and North America and 26 February 2020 for Latin America. 15 More cases outside of China were documented on 20 January (Japan, South Korea and Thailand), with the first US case reported on 21 January (Taylor, 2020) . America, the initial impact is less severe but more than doubles ( of -0.002338 to of -0.0054) after the end of February, before declining progressively ( of -0.003980 and of -0.002243, respectively). A similar pattern emerges with triggering heightened volatility after the end of February and further after late March (significant and of 0.6190 and 0.7780, respectively) before dissipating after mid-May. The dissipating effect of uncertainty on volatility thus occurs later in Latin American markets than in North American or European markets. The weakening impact of on volatility can potentially be attributed to the COVID-19 crisis being viewed by economic agents as a no longer novel but persistent situation. The decline in uncertainty reflected in Figure 1 can also mean that a higher risk premium is no longer needed as risk aversion has dissipated or decreased substantially and/or that the decline in expected cash flows due to the pandemic is not as severe as initially predicted by the markets. Alternatively, this decline may be attributable to government responses to the pandemic, such as lockdowns and/or economic stimulus packages. A role for government interventions in reducing uncertainty and volatility is suggested by Kizys et al. (2020) but not by Zaremba et al. (2020) . The latter is investigated further in Section 3.2. African markets implies that the impact of COVID-19 uncertainty is still high. For African markets, this is potentially attributable to the pandemic still being far from its peak (WHO, 2020). For Arab markets, this may reflect a return to persistently lower levels of volatility following a large but shortlived volatility spike in early March 2020.

13 and variance ( for regional markets, taking in account structural breaks. Segments identified using the Bai-Perron test of L+1 vs L sequentially determined breaks with robust standard errors (HAC) and heterogenous error distributions. Coefficients on in the conditional variance equation are scaled by 100 000. Panel A reports estimation results for the conditional mean, which also includes proxy factors derived from regional returns using factor analysis and adjusted for the impact of and . Panel B reports the results for the conditional variance. Panel C reports model diagnostics, with and being Ljung-Box tests statistics for joint serial correlation at the 1 st and 10 th orders. ARCH(1) and ARCH(10) are test statistics for the ARCH LM test for heteroscedasticity. Breakpoint identifies the date on which each structural change occurs during the COVID-19 period, where the beginning of the COVID-19 period is taken as 1 December 2019. Each model is estimated over the primary data period between 1 January 2019 and 19 June 2020 unless residuals show dependence structures in which case longer estimation periods are used. Pre-COVID-19 and COVID-19 periods are defined as 1 January 2019 to 30 November 2019 and 1 December 2019 to 19 June 2020 respectively. Asterisks ***,** and * indicate statistical significance at 1%, 5% and 10% levels of significance respectively.

To confirm that is indeed driving returns, we factor analyse the structure of returns during the pre-COVID-19 and COVID-19 periods. For both periods, a single factor is extracted. The higher mean communality for the COVID-19 period suggests that the extracted factor explains a greater proportion of shared variance. The higher KMO statistic also suggests that a greater proportion of shared variance is attributable to underlying factors.

Both measures point towards strengthened dependence, likely attributable to the global impact of COVID-19 (Uddin et al., 2020) . Spearman correlation between factor scores and is highly significant with a coefficient of -0.3240 (ordinary -0.5619). This implies that is indeed part of a composite factor set driving regional returns over this period. Figure 2 shows that the rolling correlation between factor scores summarising the drivers of returns and during the COVID-19 pandemic steadily grows in magnitude from early February, peaking between mid-March and late April, and decreasing thereafter. These increases (decreases) correspond to a growing (decreasing) negative impact on returns and higher (lower) periods of volatility attributable to , notably for Europe, Latin America and North America as identified using structural break analysis. To confirm that reflects uncertainty during the pandemic, we compare our measure against two other measures over the COVID-19 period. The first is the CBOE Volatility index (VIX) which we treat as a measure of stock market uncertainty (Bekaert et al., 2013) . Although this is the US version of the index, Smales (2019) shows that VIX captures global market uncertainty. Chiang et al. (2015) and Dimic et al. (2016) also utilise the US version of this index as a measure of global uncertainty. The second is the recently developed Twitter-based Market Uncertainty (TMU) index of Renault et al. (2020) . Figure 3 shows that COVID-19 search term index levels move closely with the two alternative measures of market uncertainty, although with somewhat of a lag especially between the end of January 2020 and the end of the sample period. Furthermore, changes in both measures become highly correlated with between the end of January 2020 and end of April 2020, implying that both reflect during this period (see Figure 9A and10A in the Appendix).

Given that these two measures appear to reflect COVID-19 related uncertainty over the COVID-19

period, we re-estimate the specifications in Table 2 Arab markets remain least impacted. As in Table 3 , North American markets experience the lowest volatility triggering effects in response to both alternative measures although they respond significantly to . Conversely, Latin American markets continue to be significantly and highly impacted. Overall, our results are largely consistent with those presented in Table 3 , providing support for the role of as a measure of uncertainty during the COVID-19 period. Given that shows a dissipating impact on returns and volatility in Table 5 and that Figure 2 suggests that the importance of diminishes, we set out to determine whether this can be attributed to government responses during the COVID-19 crisis. We first construct a response measure, , using the Oxford COVID-19 Government Response Tracker database 16 and then test model specifications by incorporating in place of in Table 2 after adjusting returns for the impact of . 1.9600*** (1 st ) This table reports the abridged results for the impact of changes in government responses to the COVID-19 crisis on the returns ( and variance ( ) or regional markets. Coefficients for in the conditional variance equation are scaled by 100 000. Values in brackets (…) rank the order of absolute impact according to the magnitude of coefficients on

The asterisks ***,** and * indicate statistical significance at 1%, 5% and 10% levels of significance respectively. Unabridged results are reported in Table 6A in the appendix. Regional returns are adjusted for . Table 8 show that returns for most regions, with the exception of Asia, respond negatively to government responses to the pandemic. While this measure also reflects economic support measures, it may be that containment measures (lockdowns and restrictions) dominate. This would explain a negative relationship. Four regions are significantly and negatively impacted with North America and Arab markets the most and least impacted respectively. Moreover, response measures are associated with significant volatility triggering effects in four regions, namely Asia, Latin

America, Europe and Arab markets, which show the greatest response by far. A potential reason for the positive impact is that these measures were implemented around the times and in response to significant COVID-19 related events which also had an adverse impact on stock markets and volatility, and therefore responses are a proxy for the immediate impact of these events. 17 These findings are in line with that of Zaremba et al. (2020) who find that stringent policy responses tend to increase return volatility in international markets. We therefore propose that the lessening importance of in Table 5 is attributable to a normalisation of economic agents' expectations.

Finally, we present variance forecasts obtained from ARCH/GARCH specifications against realized variance for the COVID-19 period. Plots in Figures 2A to 7A in the Appendix 18 show that our forecasts approximate the changing volatility dynamics and that the increases (decreases) in variance coincide with increases (decreases) in search volumes (see Figure 1A ).

Using the ARCH/GARCH framework, we demonstrate that COVID-19 uncertainty has impacted almost all regions in terms of lower returns and increased market volatility. Asian markets appear to be more resilient to COVID-19 related uncertainty, while European, North and Latin American markets experience a weakening of the impact over time. The evidence of a differential impact of COVID-19 across time and regions paves the way for further research into the reasons why such effects exist and as to why they dissipate over time. We confirm that our measure of COVID-19 related uncertainty reflects uncertainty by showing that it moves closely with alternative measures of uncertainty during the COVID-19 period. These measures, namely the VIX and TMU index, have a similar impact on returns and volatility over the COVID-19 period. Our results, together with the analysis of the structure of the return generating process show that COVID-19 uncertainty is part of the factor set driving regional returns although its role has lessened substantially. 17 Correlation analysis shows that and are contemporaneously correlated suggesting that responses occurred around the time of heighted COVID-19 related uncertainty (ordinary correlation of 0.2415, Spearman's correlation of 0.1496, statistically significant at 1% level of significance). 18 We use squared residuals from a least squares regression of the mean without breaks to proxy for realized variance.

Time varying international financial integration for GCC stock markets

Investor attention and stock market volatility. The Review of Financial Studies

Stock markets' reaction to COVID-19: cases or fatalities

Deaths, panic, lockdowns and US equity markets: The case of COVID-19 pandemic

Efficiency of emerging and emerged capital markets: An empirical study on BRICS and G7 nations

The global crisis and equity market contagion

Risk, uncertainty and monetary policy

Studies of Stock Market Volatility Changes

University of Bologna Working Paper No. 1134

Leverage effects in financial markets: The retarded volatility model

The supply channel of uncertainty shocks and the crosssection of returns: Evidence from the COVID-19 crisis. Georgia Tech Scheller College of Business Research Paper No

Public information arrival and investor reaction during a period of institutional change: An episode of early years of a newly independent central bank

The residual market factor, the APT, and mean-variance efficiency

Structural uncertainty and breakpoint tests: An application to equilibrium velocity

Google it up! A google trends-based uncertainty index for the United States and Australia

Dynamic stock-bond return correlations and financial market uncertainty

Cross-country experiences and policy implications from the global financial crisis

Stock gyrations

Public concern and the financial markets during the COVID-19 outbreak. University of Venice Working Paper No. 3591193

Impact of financial market uncertainty and macroeconomic factors on stock-bond correlation in emerging markets

Measuring economic uncertainty and its impact on the stock market

Modelling the persistence of conditional variances

Quasi-maximum likelihood estimation of GARCH models with heavy-tailed likelihoods

Financial spillovers to emerging markets during the global financial crisis

The value of fiscal capacity in the face of a rare disaster

University of Chicago, Becker Friedman Institute for Economics Working Paper No

A new form of financial contagion: COVID-19 and stock market responses

20 Largest Stock Exchanges in The World

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Stock prices and geographic proximity of information: Evidence from the Ebola outbreak

From COVID-19 herd immunity to investor herding in international stock markets: The role of government and regulatory restrictions

Purchase deadline as a moderator of the effects of price uncertainty on search behavior

The COVID-19 outbreak and affected countries stock markets response

The effects of COVID-19 on Chinese stock markets: An EGARCH approach. The University of Sydney Working Paper No. 3612461

Weathering COVID-19 storm: Successful control measures of five Asian countries

Stock market integration in the GCC countries

The global macroeconomic impacts of COVID-19: Seven scenarios. Centre for Applied Macroeconomic Analysis Working Paper No. 19/2020

Do stock markets catch the flu?

The evolving impact of global, region-specific, and countryspecific uncertainty

Tutorial on maximum likelihood estimation

Financial integration between BRICS and developed stock markets

SARS: a non-event for affected countries' stock markets?

Covid-19 and stock market volatility

Direct and indirect effects of COVID-19 pandemic on implied stock market volatility: Evidence from panel data analysis

Feverish stock price reactions to covid-19

Twitter-based Uncertainty Indices

What do we learn from SARS-CoV-1 to SARS-CoV-2: Evidence from global stock markets

Price and volatility spillovers across North American, European and Asian stock markets

How Does Policy Uncertainty Influence Financial Market Uncertainty Across the G7?

Investor attention and global market returns during the COVID-19 crisis

Underspecification of the empirical return-factor model and a factor analytic augmentation as a solution to factor omission

Risk factors in returns of the South African stock market

A Timeline of the Coronavirus Pandemic. The New York Times

Stock market contagion of COVID-19 in emerging economies

The global impact of COVID-19 on financial markets

Coronavirus Disease (COVID-19) Situation Report -164

Infected markets: Novel coronavirus, government interventions, and stock return volatility around the globe

Writing -Original Draft, Writing -Review & Editing, Visualization, Supervision, Project administration Princess Rutendo Bwanya: Investigation, Writing -Original Draft, Writing -Review & Editing