key: cord-314113-m46j1379 authors: Arias, Jose title: COVID-19 effect on herding behaviour in European Capital Markets date: 2020-10-01 journal: Financ Res Lett DOI: 10.1016/j.frl.2020.101787 sha: doc_id: 314113 cord_uid: m46j1379 This article investigates whether COVID-19 pandemic had an effect on herding behaviour in Europe. Using a sample from the stock exchanges of France (Paris), Germany (Frankfurt), Italy (Milan), United Kingdom (London) and Spain (Madrid), over the period from January 03, 2000 to June 19, 2020, we found robust evidence that COVID-19 pandemic increased herding behaviour in the capital markets of Europe. The COVID-19 outbreak in Wuhan (China) in December 2019 has infected over 21 million people and caused nearly 766.000 deaths globally as of August 15, 2020 (Center for Systems Science and Engineering at Johns Hopkins University). The virus spread rapidly around the globe. The first confirmed case in Europe was on January, 24 th in France and spreading its roots to other countries such as Germany, on January, 27 th ; Italy, on January, 30 th ; the UK, on January, 31 th ; and Spain, on January, 31 th , (Georgeou and Hawksley, 2020) . The existing literature is limited on how pandemics impact financial markets (Goodell, 2020) . Al-Awadhi et al. (2020) report that daily growth in total confirmed cases and in the total number of deaths caused by COVID-19 impacted negatively on stock returns of 1579 stocks of companies included in the Hang Seng Index and Shanghai Stock Exchange Composite Index over the period from January 10 to March 16, 2020. Similarly, Ali et al. (2020) show that in countries such as the US, UK, Germany and South Korea stock market volatility increased significantly from the epidemic (December 2019 to March 10, 2020) to the pandemic period (post March 10, 20202 ). Mazur et al. (2020) find that natural gas, food, healthcare, and software stocks earn high positive returns whereas equity values in petroleum, real estate, entertainment, and hospitality sectors fall dramatically for 1,500 firms of the Standard and Poor's for the month of March 2020. In this context, an aspect not investigated so far is whether COVID-19 has an effect on herding behavior. Herding behaviour is a process where investors imitate the actions of others (Hirshleifer and Hong, 2003) . COVID-19 may affect herding behaviour in two directions. First, investors faced with the decline of the economy and medical and social uncertainty, consider the information available to maintain and / or invest in capital markets on the basis of their beliefs, or second, they take into account other agents who are more informed and follow their behaviour. In the latter case, the most informed agents, knowing the behaviour of the least informed, could carry out strategies to arbitrate the market for their own benefit and to the detriment of others. Given the above, we investigate whether the COVID-19 pandemic increases herding behaviour in capital markets in Europe. The paper proceeds as follows. Section 2 introduces the data and methodology. Section 3 presents the empirical results. Section 4 concludes the article. We collect data on stock prices for firms listed on the CAC 40 Index (France), DAX 30 Index (Germany), FTSE MIB (Italy), FTSE 100 Index (United Kingdom), Ibex 35 Index (Spain) over the period from January 03, 2000 to June 19, 2020 1 . We concentrate on these markets because these were the first countries to report COVID-19 in Europe 2 . The level of herding is more evident when daily data is used (Tan et al., 2008 January in the Federal state of Bavaria; in Italy Chinese tourists were diagnosed on 30 January on board a cruise ship; in the United Kingdom the first cases were reported on 31 January in the city of York; and in Spain the first patient with COVID-19 was identified on 31 January in La Gomera (Georgeou and Hawksley, 2020) . As in other studies (Mobarek et al., 2014; Tan et al., 2008) we use the return dispersion method. The calculation of the return dispersion measure requires the calculation of an average market portfolio return , . We use the equally-weighted average of stock returns as a proxy for , . Specifically, to detect herding behavior the model proposed by Chang et al. (2000) is used, which is a modification of the model proposed by Christie and Huang (1995) . Chang et al. (2000) suggest using the following cross-sectional absolute deviation (CSAD) model: where , is the market return (equal-weighted average stock return) and CSADt is a measure of return dispersion calculated as: where | , | and , are the absolute value of market return and individual stock return of stock i, respectively. Specifically, to assess the effect of covid-19 on herding we used the following specification: The covid-dummy (Dcovid) equals one from January 24, 2020, to France; January 27, 2020, to Germany; January 30, 2020, to Italy; and January 31, 2020, to United Kingdom and Spain; and zero before that. Eq. (3) is used to assess the presence of herding in Europe before and after the reporting of the first case of COVID-19. Significantly negative values of 3 and 4 would indicate the presence of herding following (before) of COVID-19. To assess the robustness of the results we utilized different periods (2001.01.30-2020.06.12; 2010.01.04-2020.06.12; 2015.01.05-2020.06.12; 2018.01.05-2020.06.12 ). In addition, we consider three effects used in the literature that can affect herding behaviour: asymmetric effects of market return, high and low volatility states and domestic market trading volume (Tan et al., 2008; Mobarek et al., 2014) . In the first case, we examine whether there is any asymmetry in herd behaviour when the market is rising or falling, pre and post COVID-19. In the second case, we consider high volatility when the observed volatility becomes higher than the moving average of volatility over the previous 30 days and low volatility when it does not exceed the moving average over the same period (Chang et al. 2000) . The volatility is calculated as the standard deviation of market return times the square root of 252 trading days. In the third case we calculate for each market as the sum of the daily volumes of all listed stocks in that market. 3.1. Descriptive statistics Table 1 reports the descriptive statistics for the CSAD measure and the average market return, calculated using both equal weights for each of sample markets. Results in Table 1 show that for all stock exchanges the mean values and standard deviations of CSAD are higher after COVID-19. A higher mean value suggests significantly higher market variations across stock returns. A higher standard deviation may suggest that markets have unusual cross-sectional variations due to unexpected events (Chiang and Zheng, 2010) . Dev. The results could be explained by less informed investors following more informed ones in a clear example of herding behavior which in turn led to the erratic behavior of the capital markets. Fear and uncertainty over the effects of the pandemic would drive the less informed agents to abandon their beliefs and follow the more informed ones. 3 In the case of low volatility, the number of observations is not sufficient to make robust estimates. 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 Epidemics and economics: New and resurgent infectious diseases can have far-reaching economic repercussions An examination of herd behavior in equity markets: an international perspective An empirical analysis of herd behavior in global stock markets Following the pied piper: do individual returns herd around the market? The contagion effects of the COVID-19 pandemic: Evidence from gold and cryptocurrencies‖ State Responses to COVID-19: a global snapshot at 1 COVID-19 and finance: Agendas for future research‖ The impact of COVID-19 pandemic upon stability and sequential irregularity of equity and cryptocurrency markets Herding behaviour in the Chinese and Indian stock markets COVID-19 and the march 2020 stock market crash. Evidence from S&P1500. Finance Research Letters A cross-country analysis of herd behavior in Europe Herding behavior in Chinese stock markets: An examination of A and B shares Christian Espinosa Méndez: Conceptualization, Methodology, Software, Data curation, Writing-Original draft preparation, Visualization, Investigation, Validation, Writing-Reviewing and Editing.José Arias: Data curation, Writing-Original draft preparation, Visualization, Investigation, Validation, Writing-Reviewing and Editing.