key: cord-1001634-izsomd86
authors: Ahmad, Wasim; Kutan, Ali M.; Chahal, Rishman Jot Kaur; Kattumuri, Ruth
title: COVID-19 outbreak and firm-level dynamics in the USA, UK, Europe, and Japan
date: 2021-09-16
journal: International Review of Financial Analysis
DOI: 10.1016/j.irfa.2021.101888
sha: 02e5b198e407c8b37d19750be07257fc0e744fcd
doc_id: 1001634
cord_uid: izsomd86

This paper examines the impact of the coronavirus outbreak during its first and second waves for the USA, UK, Europe, and Japan. We explore the firm-level dynamics and exhibit the impact of coronavirus events on large and small firms and firms' idiosyncratic risk. We find that the intensity of the impact of the coronavirus outbreak events is not uniform firms. The Blank Swan events in March 2020 are still pronounced as compared to the second wave. The second wave analysis reveals the sign of recovery and receding effect of the pandemic. The idiosyncratic analysis shows the positive impact of the coronavirus and stringency measures on the idiosyncratic risk for most countries.

regions are trapped by the virus. The coronavirus pandemic has sought the attention of researchers with numerous studies covering macro and micro dimensions. The first version focussed on the macro dimension, followed by the micro dimension at a later stage. There is a need to incorporate the impact of the second wave and analyse the learnings from the first wave across countries. This study is a significant contribution in this direction as it examines the first and second waves of the coronavirus pandemic and provides a micro (firm) perspective to the analysis. The study focuses on the developed markets as these economies are major drivers of the global economy. 1 We examine the firms in two stages. At the first stage, we confirm the shock of coronavirus first and second waves on their stock market performance using the event-study approach. The result of the first wave shows significant impact than the second wave in the case of developed markets comprising the USA, UK, Europe, and Japan. We, then, shortlist the firms based on their size and employability to confirm whether the impact of coronavirus outbreak has been homogenous across firms (large and small). Employability implies the number of employees in a firm to ensure that we do not miss analysing the impact of the pandemic on their employment strength. One of the commonly pursued objectives is to find whether the rescue measures should have the components of first come and first serve or oriented to firms that are impacted. In the 1 Numerous studies have examined the impact of coronavirus at the macro and financial markets levels in the literature. For instance, Haroon and Rizvi (2020) for the US and world markets. Goodell and Huynh, (2020) examined the trading behaviour of US legislators during the coronavirus outbreak. Hernandez et al. (2020) analysed the impact on developing and emerging America. Sharif, Aloui, and Yarovaya (2020) on the effects of coronavirus outbreak on the US economy. repuation-based contagion of the coronavirus pandemic. Ali, Alam, and Rizvi (2020) examine reactions and channels of COVID-19 spread. Haroon and Rizvi (2020) investigated the relationship between pandemic news and stock market volatility. Corbet, Larkin, and Lucey (2020) on Chinese stock markets and bitcoin during the pandemic's peak. Hartley and Rebucci (2020) showcase the significant effect of quantitative easing on emerging and developed markets. Maneenop and Kotcharin (2020) examine the impact of coronavirus pandemic on the global airline stock performance. Chen and Yeh (2021) examine the reactions of US industries to both the global financial crisis of 2008 and COVID-19 pandemic. Heyden and Heyden (2021) showcase the impact of short-term market reactions to COVID-19 during the first phase. At sector level, Ahmad, Kutan and Gupta (2021) examine the US, UK, European sectors and conclude the devastating effect of the coronavirus outbreak during the second week of March 2020. confirm the network synchronicity with the implied volatility of US stock market with sectoral returns. Szczygielski et al. (2021) find the significant impact of COVID-19 uncertainty in the global industry returns.

Dow-Jones Industrial Average (DJIA) fell by 26%. It is noteworthy that in the second week of March 2020, the US Securities and Exchange Commission (SEC) had to apply circuit breakers four times to avoid the black swan events similar to the 1987 Black Friday.

As the developed markets have experienced the first wave, the second and subsequent waves have created panic among policymakers regarding the delay in recovery. Although the vaccine invention provides a timely respite, its uneven distribution and vaccine hesitance started the debate on poor economic recovery. Apart from emerging markets, the second wave has been harmful to the UK and Japan, where the regulatory structure faced enormous challenges to keep moving the economy. In this context, the firm-level analysis may provide an immediate direction to whether the intensity of the second wave has been the same as the first wave.

We find that the first wave of COVID-19, particularly during the Black Swan events of March, had a severe impact on firms' and the impact across large and small firms has been uniform for the USA and to some extent in Europe. But we do not observe such behaviour for the UK and Japan. The event-study analysis suggests that the impact of the first wave had a strong impact on Europe and the US markets. However, we do not observe a similar impact intensity during the second wave in these countries. The event-study analysis till April 2021 reveals the recovery and better performance in the stock markets of these economies. There is also a visible impact of vaccination drive and hope for a quicker recovery in our analysis. The idiosyncratic analysis shows a significant and positive impact of COVID-19, and stringency measures, on the idiosyncratic risk of firms. The analysis of large and small firms also conveys the same. Time-series and cross-sectional analysis further confirms the negative impact of cash flow, return on equity, market capitalization (size) on the idiosyncratic risk of the firms during the sample period that significantly covers the coronavirus outbreak period. has negatively impacted the firms' environmental performance. The COVID-19 disruptions on the stock also impacted the performance of efficient and inefficient firms, as examined by Neukirchen et al. (2021) . They find that the highly efficient firms recorded a higher jump in their stock returns than the crisis-period returns. Whether operating flexibility contributed significantly to the firm's performance during the first wave of coronavirus outbreaks has been examined by Liu, Yi, and Yin (2021) . They find that the operating flexibility indeed helped the firm, especially in those provinces which were badly hit by the pandemic in China.

For Europe, Huynh, Foglia, Doukas (2021) confirm the strong interconnectedness in the early phase of COVID-19 for the 46 large companies. Didier (2021) highlights the relevant issues in firms' financing and how difficult it has become for firms to remain in the hibernation model due to unprecedented uncertainty.

Overall, it is apparent that most studies cover the first wave of the Pandemic and have limitations regarding the firm selection and choice of variables. We still find our study different than the above discussed and contribute extensively to the literature. The rest of the study is organized as follows: Section 3 outlines the data and methodology. Section 4 focuses on analyzing results. Section 5 concludes the study.

We consider the daily data from May 1, 2019, till April 30, 2021, for the event study analysis. For Europe, we consider S&P (Standard and Poor's) for Europe, which has 186 constituents. For Japan and the UK, Nikkei-225 and FTSE-350 (Financial Times Stock Exchange), respectively. For the US market, we consider 503 stocks of S&P-500. All the sample data have been sourced from Thomson DataStream (Refinitiv). We calculate the J o u r n a l P r e -p r o o f abnormal returns from Fama and French (1992) , three-factor model. We download the factors series from the Fama and French's webpage. 2 We identify significant events related to the coronavirus outbreak and financial markets. In the first step, we adopt linear and nonlinear endogenous structural break models. As a linear model, we use the Bai and Perron (2003, hereafter BP) model which is based on the general-to-specific procedure. The key characteristic of this test is that it allows us to identify the unknown dates endogenously. It uses the sup F T (k, n) test that has the null hypothesis of no structural break (n = 0) against the alternative of a structural break (n = k).

The null hypothesis remains the same for the double maximum and sequential test criteria, which also add a methodological dimension to structural breaks. In the second step, we adopt the nonlinear framework of the Markov-switching Model (MSM) coined by Hamilton (1989) .

We use the MS-DR (Dynamic Regression) framework of Doornik (2013) due to the adoption of high-frequency data. The MS-DR has the same number of regimes and states, which makes it suitable for daily and monthly data. We specify the MS-DR model with switching intercept (means) and the variance 3 :

where, we assume that a market return is generated as an autoregression of order k with regime-switching in intercept (mean)  and variance ( 

The event-dates identified through the endogenous structural break test are listed in Table 1 . 5

It appears that the event dates identified by the structural break models seem valid. According to Mazur, Dang, and Vega (2021) Corbet, Larkin, and Lucey (2020) are different than our study because they select the dates based on the news coverage only.

We also cover 2021 till April when most countries experienced either the first wave or second wave and take into account the effect of vaccination derives in the sample countries and how the reactions of the markets at the beginning of 2021. The results suggest that during 2021, we identify two breaks, i.e., February 16 and March 4. On both dates, we observe that the stock markets had positive reactions. 

It is apparent that the coronavirus outbreak has generated the Black Swan events, and most of these events occurred in March 2020. The structural breaks during the 2021 period exhibit a sign of recovery and green shoots in these economies. We provide a detailed analysis of four dates viz., March 12, 18 in 2020, and February 16 and March 4 in 2021, analyze their impacts on different classifications of firms. We first classify the constituents of stock indices of the USA, UK, Europe, and Japan into large and small using two criteria. First, based on the number of employees, and second, based on the size. We do this exercise to confirm whether the impact of coronavirus outbreak is limited to only large firms or small firms are also equally impacted. Do we observe any variations in the effect of the pandemic of these firms?

The analysis may help understand the impact at the micro-level, which is a major research gap in the existing literature and also the main objectives of this study. J o u r n a l P r e -p r o o f Hartley and Rebucci (2020) , and Haroon and Rizvi (2020) . The finding implies that the coronavirus outbreak may decrease the possibilities of employment opportunities in the firms.

However, during 2021 events, industries, information technology, and financials as leading sectors across sample countries.

Overall, the classification of the top ten firms based on the number of employees suggests that the Black Swan event dates had a significant impact, small and large, on the US and European markets. The Japanese stocks (firms) exhibit a mixed effect on all event dates.

Based on this result, we conclude that the above result provided sufficient insights about the design of policy stimulus and recovery plans. It will be essential to track the performance of affected firms, and if the negative effect persists for a longer period, a proper stimulus package may give a new life. For instance, in the case of the US and Europe, policymakers should emphasize reviving the smaller firms to generate more employment than the larger firms. Although one may argue that the stock market fall is often linked to short-term gains or losses, it is also critical to monitor these firms' financial and operating performances. From the investors' perspective, the result suggests investment in large and small firms until March 31, 2020. However, the 2021 event dates significantly explain the impact of COVID-19

vaccine and vaccination plans in these economies.

We also classify the firms based on their size (market capitalization) to confirm the above results. 

It is apparent from above analysis that the coronavirus pandemic has impact the performance of firms and it would be wise to analyse these firms from systematic and nonsystematic risks perspectives as well. As size and employment strengths do matter for the micro-analysis, we undertake an idiosyncratic volatility analysis to find the extent of the impact of coronavirus outbreak on the idiosyncratic risk of firms. We adopt the following procedure: First, we calculate the idiosyncratic risk using the three-factor model using Equation (2). Second, we decompose the total risk into systematic and firm-specific risk, also known as the idiosyncratic risk. Following Fu (2009), the realized idiosyncratic risk (volatility) series is obtained using the standard deviation of the residuals from equations (2).

We have used a time-varying regression of equations (2), with a period of at least 20 daily observations in a month, to generate an idiosyncratic monthly series. Then the standard deviation of the residuals is used as the idiosyncratic risk component. Third, we specify the following regression to estimate the impact of firm-specific factors and controls. Table 4 shows the results of panel fixed-effect regression Equation (6). We find that firmspecific factors such as cash flow (Cash) and market capitalization (MC) exhibit a negative relationship with idiosyncratic risk. The results seem valid though the statistical significance varies across markets. Economically, the negative relationship implies that a 1% increase in the market capitalization in the case of US leads to a 10.8% decrease in the idiosyncratic risk for all countries. The result implies that as the firm's profitability increases, idiosyncratic risk declines and is much stronger during the crisis period. However, the coefficient of trading volume (Volume) picks a positive sign that implies that the increase in trading volume leads to an increase in idiosyncratic risk. It has happened during the crisis period as the sample period of monthly analysis covers the coronavirus outbreak period.

The coefficients of controls such as COVID-19 (COVID) cases and Stringency measures positively explain firms' idiosyncratic risk that implies coronavirus cases increased, which led to an increase in the idiosyncratic risk of firms. However, there is a caveat in the case of the USA. The coefficient of stringency exhibits a negative sign, suggesting an inverse relationship between idiosyncratic risk and COVID-19 related stringency index. According to J o u r n a l P r e -p r o o f Huang, Yang, and Zhu (2021) , top brands in the USA experienced higher stock returns, lower systematic risk and lower idiosyncratic risk as the COVID-19 restrictions increased during the COVID-19 outbreak period. Our result seems to be valid in this respect.

The coefficient of OVX shows a positive relationship, and it also suggests that the idiosyncratic risk of firms positively explains it. The impact of COVID-19 has been significant across sectors in the USA, as reported by Ahmad, Hernandez, Saini, and Mishra (2021) . This result is a new finding at the firm level for these countries.

However, we also estimate the Fama and Macbeth (1973) cross-sectional regression for all firms. Table 5 shows the results. We find that the signs of the coefficients are commensurate with the results of Table 4 discussed above.

We also estimate the results for large and small firms based on their size and number of employees, as we have done for the event-study analysis. This analysis is a robustness exercise to confirm the results of the event-study analysis and the results reported at the aggregate level in Tables 4 & 5 . We find that large firms based on market capitalization and number of employees are reported in Tables 6-7 and Tables 10-11 . The cash flow shows a significant and negative relationship with idiosyncratic risk for firms in the UK and Japan, whereas the coefficients are insignificant for the USA and Europe. We also draw a similar statistical inference for the cross-sectional analysis. At the cross-section level, we find the results are more pronounced and statistically significant. This inference signifies the importance of cross-sectional analysis for the firm-level analysis.

Similarly, for large firms based on employees, we find that the cash and market capitalization exhibit a negative relationship with firms' idiosyncratic risk. However, the coefficients are not significant for all the countries. The cross-sectional analysis confirms the above findings.

Coming to COVID-19 and stringency index variables, we find that the number of COVID-19 cases and Stringency measures positively associate with idiosyncratic risk. The Tables 8-9 (size) and Tables 12-13 (employees). The firm-specific factors do not enforce enough for these firms, as we find in the previous analysis. However, some of the coefficients are significant and consistent with previous analysis for the Europe and Japan.

The COVID-19 cases and Stringency measures show a positive relationship with idiosyncratic risk for the USA, UK, and Japan. However, Europe seems an exception for the COVID-19 cases and the USA for the stringency index as the coefficients of both countries imply a negative and significant relationship with idiosyncratic risk.

Overall, the micro-analysis reveals that the differential impact of the COVID-19 cases and stringency measures on firms' idiosyncratic risk, which is consistent with the event-study analysis reported above. In event-study analysis, we observed similar differences concerning the impact of events during the first and second waves. The general hypothesis that the coronavirus outbreak has impacted the firms uniformly is incorrect, and the intensity of the impact has varied across the types of firms. 

Overall, the stock (firm) level analysis reveals interesting patterns as far as the impact of the coronavirus outbreak is concerned. Our empirical setup contributes to the literature in the following manner. First, the applications of linear and nonlinear structural break models help identify the major event related to the coronavirus outbreak, including the black swan events reported during the first and second weeks of March and also the recovery phase of 2021.

Second, the analysis of the event-study approach confirms the significant impact of coronavirus outbreak events on the stock markets of sample countries. However, the analysis became interesting when we classified the firms into two categories, large and small, using the number of employees and the size. The results suggest notable differences with regards to the different phases of coronavirus shocks beginning February 2020. The differences between large and small firms are negligible for the US and Europe, implying that the coronavirus outbreak uniformly impacted the firms and stocks on March 12 and 18. Further studies can J o u r n a l P r e -p r o o f examine these issues. We observe an almost similar impact on small and large stocks for the UK, but large firms seemed more responsive than small stocks. For Japan, we observe the symmetric effect of the coronavirus pandemic across large and small firms. The incorporation of 2021 till April makes a difference to analysis as AR and CAR values as not as significant as we found in 2020.

Overall, the above-discussed results provide enormous opportunities for policy experts to trace the financial performance of small and large firms. A suitable remedy could be suggested to reduce the financial vulnerabilities of these firms. As aforementioned, the classification of firms' analysis seems useful from a policy perspective. The findings of our study can be linked to Kumar and Haydon (2020) , Goodell and Huynh (2020) and Hartley and Rebucci (2020) , and Haroon and Rizvi (2020), Huang, Yang and Zhu (2021) and Chebbi, Ammer and Hameed (2021) . Some results also differ with multi-country studies by Hu and Zhang (2021) and

However, the idiosyncratic analysis further substantiates the above findings as firmspecific factors establish a negative relationship with sample firms' idiosyncratic risk. The control variables such as COVID-19 cases and stringency measures positively explain firms' idiosyncratic risk, which is a significant finding. The analysis of large and small firms also confirms the differential impact of the coronavirus outbreak events based on their sizes.

Based on these results, we can say that the stock level analysis at the micro-level gives better clarity than the country and sectoral analysis, and the study can be extended to cover more countries and even emerging markets.

COVID-19 Crisis Poses Threat to Financial Stability

Black swan events and COVID-19 outbreak: Sector level evidence from the US, UK, and European stock markets

The US equity sectors, implied volatilities, and COVID-19: What does the spillover analysis reveal?

Death and contagious infectious diseases: Impact of the COVID-19 virus on stock market returns

Coronavirus (COVID-19)-An epidemic or pandemic for financial markets

Computation and Analysis of Multiple Structural Change Models

Fear from uncertainty: An event study of Khashoggi and stock market returns

Asset pricing and extreme event risk: Common factors in ILS fund returns

Stock market Friday recap: Dow loses 3,500 points on week, closes off lows, Fed pledges support

The COVID-19 pandemic and stock liquidity: Evidence from S&P 500. The Quarterly Review of Economics and Finance

Aye Corona! The contagion effects of being named Corona during the COVID-19 pandemic

The contagion effects of the COVID-19 pandemic: Evidence from Gold and Cryptocurrencies

Financing firms in hibernation during the COVID-19 pandemic

A Markov-switching model with component structure for US GNP

Common risk factors in the returns on stocks and bonds

The cross-section of expected stock returns

Risk, return, and equilibrium: Empirical tests

Chinese shares drop nearly 9% amid fears of coronavirus impact

Firms' environmental performance and the COVID-19 crisis

Does COVID-19 state aid reach the right firms? COVID-19 state aid, turnover expectations, uncertainty, and management practices

Did Congress trade ahead? Considering the reaction of US industries to COVID-19

A new approach to the economic analysis of nonstationary time series and the business cycle

COVID-19: Stock Market Reactions to the Shock and the Stimulus

COVID-19: Media coverage and financial markets behavior-A sectoral inquiry

An event study of COVID-19 central bank quantitative easing in advanced and emerging economies

Spillovers and diversification potential of bank equity returns from developed and emerging America

Statistics of extreme events in risk management: The impact of the subprime and global financial crisis on the German stock market

COVID-19 and Tail-event Driven Network Risk in the Eurozone

Brand equity and the Covid-19 stock market crash: Evidence from US listed firms

Impact of COVID-19 on stock price crash risk: Evidence from Chinese energy firms

COVID-19 pandemic and firm performance: Cross-country evidence

Market reactions to the arrival and containment of COVID-19: an event study. Finance research letters

Searching for safe-haven assets during the COVID-19 pandemic

The impact of COVID-19 on firm innovation: Evidence from Chinese listed companies

Risk contagion of COVID-19 in Japanese firms: A network approach

Procyclical ratings and market reactions

Effects of epidemic disease outbreaks on the financial performance of restaurants: Event study method approach

Industries Most and Least Impacted by COVID-19 from a Probability of Default Perspective

The impact of operating flexibility on firms' performance during the COVID-19 outbreak: Evidence from China. Finance research letters

Productivity of firms using relief policies during the COVID-19 crisis

The impacts of COVID-19 on the global airline industry: An event study approach

COVID-19 and the march 2020 stock market crash. Evidence from S&P1500

How has the cryptocurrency market performed during COVID 19? A multifractal analysis

Firm efficiency and stock returns during the COVID-19 crisis

Stock Markets and the COVID-19 Fractal Contagion Effects

Stock overreaction to extreme market events

New evidence on COVID-19 and firm performance

COVID-19 pandemic, oil prices, stock market, geopolitical risk and policy uncertainty nexus in the US economy: Fresh evidence from the wavelet-based approach

The impact and role of COVID-19 uncertainty: A global industry analysis

Mad March: how the stock market is being hit by COVID-19

Economic policy uncertainty dispersion and excess returns: Evidence from China

Financial markets under the global pandemic of COVID-19

Extreme co-movements and extreme impacts in high frequency data in finance

Stock market reaction to COVID-19: Evidence from US Firms' International exposure