Do investors overreact to COVID-19 outbreak? An experimental study using sequential disclosures

Sulistiawan, Dedhy; Feliana, Yie Ke; Rudiawarni, Felizia Arni; Grigorescu, Adriana Article Do investors overreact to COVID-19 outbreak? An experimental study using sequential disclosures Contemporary Economics Provided in Cooperation with: VIZJA University, Warsaw Suggested Citation: Sulistiawan, Dedhy; Feliana, Yie Ke; Rudiawarni, Felizia Arni; Grigorescu, Adriana (2023) : Do investors overreact to COVID-19 outbreak? An experimental study using sequential disclosures, Contemporary Economics, ISSN 2300-8814, University of Economics and Human Sciences in Warsaw, Warsaw, Vol. 17, Iss. 1, pp. 43-57, https://doi.org/10.5709/ce.1897-9254.498 This Version is available at: https://hdl.handle.net/10419/297620 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ www.ce.vizja.pl 43 This work is licensed under a Creative Commons Attribution 4.0 International License. This paper aims to investigate market participants' reactions to sequential information, presenting firm-spe- cific news and market-wide information. Experimental study takes place in the COVID-19 pandemic era, as market-wide information representation. We also provide firm-specific information in the form of company fundamental information. The results show that participants, as representatives of retail investors, do not overreact to COVID-19. The recency effect dominates their decision-making. Neither firm-specific information nor market-wide information can eliminate the recency effect in decision making. Investors still provide valuations based on the latest information they receive. Another interesting finding in this study is that positive framing of information cannot mitigate the effects of bad news contained therein. Our findings contribute to the study of behavioral finance and corporate disclosure strategies. From the market participants' point of view, our results describe that investors' decisions are often not based on the information content but the latest information they received. From the company perspective, this research also contributes to the corporate disclosure strategy valued by investors based on how they disclose information to the public. 1. Introduction1. Introduction The main purpose of this paper is to evaluate the impact of sequential information on investors' decisionmaking. The recency effect studies present evidence that sequential information stimulates stock overvaluation (or undervaluation) (Libby & Tan, 1999; Pinsker, 2007, 2011; Tuttle et al., 1997). People tend to pay gradually increasing attention to later evidence due to bounded rationality (Gandré, 2020). When the sequence of good (bad) news is presented after the series of bad (good) news, people tend to react more positive (negative). Our study develops those studies using corporate announcements and Corona Virus Disease 2019 (COVID-19) news to represent bad news. The use of the COVID-19 issue enhances the context of stock investing in the pandemic era. COVID-19 issue is really bad news that substantially affects firm operations. According to CNN (2020), this news began to go global in early January 2020. Even at the beginning of March 2020, a panic situation emerged in many countries after Italy and South Korea experienced outbreaks. Even in the United States, the National Basketball Association (NBA) match, a competition Do Investors Overreact to COVID-19 Outbreak? An Experimental Study Using Sequential Disclosures ABSTRACT A12, G41, C91 KEY WORDS: JEL Classification: COVID-19, sequential information, experiment, recency effect, market-wide information, firm-specific information 1 University of Surabaya, Indonesia 2 National University of Political Studies and Public Administration, Romania Correspondence concerning this article should be addressed to: Dedhy Sulistiawan, Jl. Raya Kalirungkut 60293, Surabaya, Indonesia. E-mail: [email protected] Dedhy Sulistiawan1 , Yie Ke Feliana1 , Felizia Arni Rudiawarni1 , and Adriana Grigorescu2 Primary submission: 16.02.2022 | Final acceptance: 20.04.2022 44 Dedhy Sulistiawan, Yie Ke Feliana, Felizia Arni Rudiawarni, Adriana Grigorescu 10.5709/ce.1897-9254.498DOI: CONTEMPORARY ECONOMICS Vol. 17 Issue 1 43-572023 between the world's largest basketball teams, is delayed and reopens the matches, but the fans are forbidden to attend. Several governments around the world conduct lock-down policy. In our study, COVID-19 is the representation of bad news. It is market-wide information. Our study also uses firm-specific information to complement market-wide information by using earnings and fundamental information. Earnings announcements and other fundamental information are important information in the stock market (Francis et al., 2002; Hribar & McInnis, 2012;). In this study, we focus on presenting positive corporate news and negative news from the COVID-19 outbreak. Both of this information are distributed in the market and produce mixed-reaction. Most public firms announce their financial reports in March, while the COVID-19 outbreak spread throughout the world starting in March 2020. Those phenomena stimulate a research question. Using an experimental study, we expect to describe how joint information affect market participants decision when presenting sequentially. This study complements the previous study regarding joint information. Flannery and Protopapadakis (2002) present that macroeconomic factors affect aggregate stock returns. Furthermore, Mian and Sankaraguruswamy's (2012) research found that investors react differently to good news and bad news of earnings announcement based on the market sentiment. The study of Chen et al. (2018) shows that investors react more to macroeconomic disclosures than firm-specific information. Macroeconomic news attracts investors' attention. In effect, market participants pay less attention to earnings information when there is important market-wide news than at other times. However, in total, investors' attention to earnings information and macroeconomics information is greater when there is an important macroeconomic news announcement. In this research, the COVID-19 outbreak is macroeconomics news which is very important for the whole world. This study combines this market-wide information with firm-specific information that the company carried out during the COVID-19 outbreaks. We investigate further by completing the information released regarding the market-wide and firm-specific information with how the information is presented, that is, sequential. The discussion of sequential information is one of the important issues in our business and life. Outside the stock market, sequential information will stimulate bias decisions. Juries in the court tend to be affected by sequential evidence (Furnham, 1986). Auditors are also stimulated by recent information than previous ones because of the order-effect bias (Ashton & Kennedy, 2002; Ashton & Ashton, 1998). Some scholars also provide evidence that information sequentially presented affects job recommendation (Chen et al., 2019) and clicking behavior (Murphy et al., 2006). This study seeks to analyze the impact of sequential information between the positive fundamental news and COVID-19 outbreak news in the stock market setting. This paper is organized into four sections. The second part develops hypotheses, and the third section describes the research design, while the finding section shows and discusses the findings. The last section concludes the research findings and limitations. 2. Literature Review2. Literature Review Investing based on individual pieces of news may produce stock price movement. The good news (bad news) normally generates positive (negative) reaction. Ideally, people will have the same reaction when they receive the bundle of information. Based on the rational perspective, people who receive positive-negative and negativepositive information should produce the same reaction. The efficient market hypothesis supports that argument. However, our study refers to behavioral finance studies. We believe that human is not fully rational. The complexity and completeness of information in the stock market are problems for market participants because of human limitations in information processing. Belief revision theory, presented by Hogarth and Einhorn (1992), conducts a discussion about inconsistent disclosures, both positive and negative, produce more belief revision when sent to decision-makers after disclosing the opposite sign. After receiving contrary evidence, people tend to experience an increased sensitivity level after recognizing the existing belief. After reviewing good news, a piece of bad news will cause the decisionmaker to be more sensitive to the new evidence. The presentation of bad news is contrary information to what people have previously received. The amount of their belief revision is also increasing but in the opposite direction. The phenomenon of overreaction to the newest information is called recency effects. In the stock market, earnings announcements and www.ce.vizja.pl 45 Do Investors Overreact to COVID-19 Outbreak? An Experimental Study Using Sequential Disclosures This work is licensed under a Creative Commons Attribution 4.0 International License. other fundamental news are presented in step-by-step mode rather than distributed in a bundle of information. The news of COVID-19 is also distributed to the news portal and social media in a sequential presentation. In the stock market, investors receive the sequence of those disclosures differently. Good (bad) news followed by bad (good) news will produce belief revision from positive (negative) to negative (positive) reaction because the last information is more weighted than previous ones. The stock markets always generate return volatility because different expectations of market participants produce different reactions. The bad news generates a stronger effect than the good news, especially in 2020 due to the COVID-19 pandemic that was perceived as a threat (KANTAR, 2020; Onyele & Nwadike, 2021). It reconfirms the Jegajeevan theory (2012) that the bad news compared with a good one is having a higher effect on the asymmetric volatility of the return. The news influences the stock market and customer behavior, as the representative of retail investors. Gambetti et al. (2021) studied four aspects of the good and bad news on the market response. Even if there is no bias or negativity in media about economic events, the negative news is more persistent and sometimes more 'attractive' for the news providers due to the effects on the audience. The COVID-19 pandemic was called after few months from its debut a 'health crisis' followed by uncertain news that an economic recession is expected. This 'neutral' news triggered customer behavior reconsideration during and after the COVID-19 pandemic. Customer behavior has been analyzed considering psychological, sociological, and economic approaches as the main pillars of the spending behavior and decision (Valaskova et al., 2015). Facing news like a COVID-19 pandemic makes consumer behavior change on a scale from small to tremendous, or at least will have a time of frozen until the news will complete the picture of the future (Jo et al., 2021). Mehta et al. (2020), based on the changes in spending preferences, comes to a conclusion, at least for India, that consumer behavior in crisis times is shifting from materialism to spiritualism. The question is, for how long will it last? European, Japanese and Korean are the most skeptical compared with the optimistic Chinese about economic recovery and reach a new normal life (Kohli et al., 2020). The mobility investment (Garibaldi et al., 2021) seems to have different shifts in different countries, the pandemic news and the recession news effect in the investment-spending mindset. Those situations are not good for the business, including the stock market. Our study develops the idea that we use joint information between firm-specific information and market-wide information. Before valuing firms using specific information, the understanding of market-wide is crucial. Based on the valuation procedure, analysts or investors should weigh more on market-wide information than firmspecific information. Previous studies have shown that macroeconomics or market-wide information attracts investors' attention because many investors apply a topdown approach in their investment selection strategies. The initial step is to determine the risk-return trade-off based on fundamental macroeconomic information. So, market processes information based on market-wide information over the firm-specific information (Chen et al., 2018). Besides, market-wide information reveals information about economic fundamentals and discount rates (Li et al., 2014; Savor & Wilson, 2013). This method makes it easier for investors to separate systematic information from firm-specific information contained in earnings news. Discussing market-wide information and firm-spe- cific information, Mian and Sankaraguruswami (2012) examine the market response related to earnings news and sentiment representing market-wide information. Mian and Sankaraguruswami (2012) show that investors react differently to earnings announcements depending on market sentiment at that time. Higher (lower) sentiment generates (lower) higher value relevance. Their results show that investors weigh higher market-wide information than firm-specific information. Their research is also supported by Seok, Cho and Ryu (2019). The COVID-19 outbreak actually has a terrible impact on business. It should be more superior information than earnings announcements. However, because of the bias stimulated by the order of presentation, the latest information received tends to be considered more in the decision-making process. People overreact to the newest information. The recency effect studies have already been discussed in several papers, i.e., Libby dan Tan (1999), Tuttle et al. (1997), Pinsker (2007, 2011), and Sulistiawan and Rudiawarni (2019), but this research develops the process by combining firm-specific and market-wide informa- 46 Dedhy Sulistiawan, Yie Ke Feliana, Felizia Arni Rudiawarni, Adriana Grigorescu 10.5709/ce.1897-9254.498DOI: CONTEMPORARY ECONOMICS Vol. 17 Issue 1 43-572023 tion. Our paper also uses COVID-19 outbreak news to enhance the context of the study. We use fundamental information as positive news and COVID-19 information as the representation of bad news. Our alternate hypothesis is presented below. H: Positive fundamental information followed by COVID-19 outbreak news produce a more negative reaction than COVID-19 outbreak news followed by positive fundamental information. 3. Data and Methodology 3. Data and Methodology 3.1. Data Our study uses purposive sampling method. Participants were undergraduate students in the final semester of the business and economics faculties, considering they had obtained sufficient basic knowledge to conduct valuations. They are also representative of nonprofessional investors or noise traders. We use a randomized design to allocate participants into each experimental group. To examine the hypothesis, we run three experiments to build robust evidence. In the first experiment, the participants were undergraduate students from a private university. In the second experiment, apart from being undergraduate students from a private university, the participants are also members of an investment club. In the third experiment, participants consisted of undergraduate students from a public university. All three experiments are independent. We use different participants in each experiment to avoid maturity bias. The maturity bias that we anticipate in this study is a bias that occurs because the participants already know the experimental process before, so the results obtained are influenced by the psychological process they experienced from the previous experimental process and not due to the factors we investigated. 3.2. Methodology Methodologically, our study differs from previous studies. There are several unique features of this experiment. First, our study applies an online experiment, while previous studies used in-person laboratory experiments (Ashton & Kennedy, 2002; Pinsker, 2007; 2011). The limitation of the online experiment is the risk of the distraction of participants. The experiment is conducted three times for the duration of COVID-19 pandemic from April 2020 until January 2021. The role of the experimenter to isolate the experiment situation is impossible to be reached. We recognize this limitation. Second, it is implemented in Covid-19 pandemic situation, and we recognize it as market-wide issue. Third, we organized three experiments to produce robust results, while most studies of recency bias examine the hypothesis using only one or less than three experiment procedures (Ashton and Kennedy, 2002; Chen et al., 2019; Pinsker, 2007, 2011). In all experiments, we divide participants into two groups. The first group receives good news (positive fundamental information) followed by bad news (COVID-19 news). Conversely, the second one receives good news after bad news. The task of both groups is to value the company after they receive each piece of information. In our experiment, our main attention is on participants' valuation after the sixth information. As the final disclosure in the information order, the valuation after this sixth information captures the participants' decision-making process. Using the recency bias idea, participants who receive the same information will decide differently because of information order. As discussed in Tuttle et al. (1997), Pinsker (2007 and 2011), and Ashton and Kennedy (2022), participants who receive good news followed by bad news (+++---) tend to react to bad news. Conversely, participants tend to react to good news when receiving bad news followed by good news (---+++). After the sixth information, all participants from both groups have obtained all the same pieces of information, although receiving in a different order. Using a rational decision perspective, when both groups receive the same information, they react indifferently. However, when presented sequentially, we believe that the order information produces a recency effect (Pinsker, 2007, 2011). The first experiment uses positive fundamental information to represent firm-specific information and COVID-19 outbreak as negative market-wide information. The statements of bad news are presented using positive framing to reduce the negative reaction of the participants. Group 1 (Group 2) receives +++--- (---+++). They give a score from 0 to 10 after receiving each piece of information. We use the score of the sixth information to measure the recency bias. Rationally, the same information produces the same valuation. However, we expect Group 2 tend to produce a higher valuation because they www.ce.vizja.pl 47 Do Investors Overreact to COVID-19 Outbreak? An Experimental Study Using Sequential Disclosures This work is licensed under a Creative Commons Attribution 4.0 International License. focus on the good news. Although bad news is provided to participants using positive framing, we believe that the impact of bad news is more pronounced than the frame. The first experiment participants are the final year undergraduate students at a business school from a private university. One example of the good news, Group 1 receive this information, “the share price of FIDELIA Corp. will go higher as many analysts believe that the company offers excellent profit growth potential in the future.” One of the bad news presented using a positive frame stated, “the company lowered its revenue and net profit targets for this year. This decline was due to management's pessimism in dealing with public panic over the Covid-19. However, this pessimism could be an overreaction to the company's performance. Expectations that are too low are likely to result in positive surprises in the future”. The second experiment is conducted by presenting COVID-19 using negative framing. The second experiment participants are the final year undergraduate business school students from a private university, and they are also members of the investment club. Duplicating the first experiment, they also receive information presented using the same format, +++--- or ---+++. In this case, the bad news is given to participants using the negative frame. It is an example of the bad news used in the instrument for the second experiment, “The company lowered its revenue and net profit targets for this year. This decline was carried out due to management's pessimism in dealing with public panic over the Covid-19”. Like the first experiment, we still believe that participants weigh on the last information more than the previous one. Information +++--- (---+++) produce lower (higher) valuation. The third experiment uses the same procedure as the second experiment, but the participants are students from different universities. They are the final year undergraduate students of the faculty of business and economics from a public university. Information is presented in Table 1. Our online experiment is held to evaluate the impact of recency bias on sequential information of positive corporate news and COVID-19 outbreak as negative bad news. The use of final year bachelor students in business and economics is based on two reasons. First, they represent noise traders or retail investors. Second, some articles present evidence that no differences impact of sequential information between students and real stock traders (Pinsker, 2011; Tuttle et al., 1997). Table 1 Groups of Experiment Group 1 Group 2 Experiment 1 Three positive fundamental information are followed by three negative information about the COVID-19 outbreak using positive framing Three negative information about the COVID-19 outbreak using positive framing are followed by three positive fundamental information Experiment 2 Three positive fundamental information are followed by three negative information about the COVID-19 outbreak using negative framing Three negative information about the COVID-19 outbreak using negative framing are followed by three positive fundamental information Experiment 3 Three positive fundamental information are followed by three negative information about the COVID-19 outbreak using negative framing Three negative information about the COVID-19 outbreak using negative framing are followed by three positive fundamental information In each experiment, participants must make a final decision after the sixth information by conducting a self-review. We also evaluate this impact on participant valuation, although it is not hypothesized. 48 Dedhy Sulistiawan, Yie Ke Feliana, Felizia Arni Rudiawarni, Adriana Grigorescu 10.5709/ce.1897-9254.498DOI: CONTEMPORARY ECONOMICS Vol. 17 Issue 1 43-572023 The hypothesis was examined by comparing the two groups of participants' valuations on the sixth information (info6). We expect that info6grup2>info6grup1. We use both the independent sample t-test and unbalanced ANOVA. The dependent variable is the valuation of the sixth information. The independent variables are order information, sex, experience, understanding, and GPA. We expect all variables, except group, do not contribute to the participants' valuation. Although we use positive framing on bad news in the first experiment, we believe that participants still focus on the order of information rather than the content. The other feature in this experiment is the use of firmspecific information in good news and market-wide information in bad news. This feature is unique for two reasons. First, although prospect theory (Kahneman and Tversky, 1979) states that bad news is reacted more than good news, our study believes that information order still produces more bias. Second, from the basic valuation perspective, market-wide information is more superior information than specific information because market information will affect the discount rate and its valuation. 4. Results and Discussion4. Results and Discussion We conduct three experiments using different situations to produce robust results. 4.1. Experiment 1 The first stage experimental procedure was carried out on April 6, 2020. The data presented in Table 2. Group 1 received good news followed by bad news (with positive framing). In contrast, Group 2 received bad news (with positive framing) followed by good news. Table 2 presents descriptive data for experiment 1. The number of participants is 149. In this procedure, participants give an average rating of 5.2 per share. Respectively, the average info1, info2, info3, info4, info5, info5, and info 6 resulted in share price valuations of 5.11, 5.8., 5.36, 5.53, 5.96 and 5.59. Our main attention is on the sixth valuation of participants when they receive info6. When they weigh more current information than the previous one, Group 2 (Group 1) will produce a higher (lower) valuation. The analysis of the participants' valuation between groups 1 and 2 is presented in Table 3. This test is an important step in differentiating the reaction of participants in responding to the information provided. Group 1 received good news, namely positive news of firm-specific information, followed by bad news, which is information about the COVID-19 outbreak. Group 2 receives bad news followed by good news. If the treatment of information order is important to participants, it should impact the participants' decisions. The ratings between Groups 1 and 2 were no different when receiving initial information. It is indicated by the t-test value of 0.127, and it is not statistically significant. It is a good starting point that shows that both groups have the same response to the same information. The two groups also have the same answer variance. In Table 3, when there is no similarity of variance, the t-test value uses the assumption of variance between different groups. Based on the presentation of Info6 in Table 3, the response of Group 1 is 4.0946, while Group 2 is 7.08. The findings indicate that groups receiving the same information produce different judgment decisions. Both groups weigh the last information higher than the previous information. The hypothesis is supported. Group 1 receives good news, followed by bad news, and bad news is presented last. In contrast to Group 2, because bad news is followed by good news, participants tend to weigh good news. The valuation of Group 2 is higher than Group 1. The key issue in experiment 1 is positive framing for the bad news. Even though the COVID19 news representing bad news has been wrapped with positive framing, recency effect still occurs. If positive framing is successful, the impact of recency bias should be reduced. However, participants' response in this experiment does not provide evidence that positive framing of bad news in a sequence of information eliminates recency effect. Statistically, the mean of valuation between groups regarding gender, age, GPA, and experience did not differ significantly. Those results show that the valuation is not influenced by the participants' characteristics but rather because of the experiment's treatment. The ANOVA test results in Table 4 present that what influences the participant's assessment decisions is information order, not gender, GPA, age or investment understanding and experience. These results give evidence that the recency effect dominates the valuation process even though bad news has been created using positive framing in the instrument. www.ce.vizja.pl 49 Do Investors Overreact to COVID-19 Outbreak? An Experimental Study Using Sequential Disclosures This work is licensed under a Creative Commons Attribution 4.0 International License. Table 2 Descriptive Data: Experiment 1 N Minimum Maximum Mean Standard Deviation Group 149 1.00 2.00 1.5034 0.50168 Info0 149 3.00 10.00 5.2349 0.79176 Info1 149 2.00 10.00 5.1141 1.72239 Info2 149 2.00 10.00 5.8054 1.86966 Info3 149 1.00 10.00 5.3691 2.18208 Info4 149 0.00 10.00 5.5369 1.82521 Info5 149 0.00 10.00 5.9664 1.89386 Info6 149 0.00 10.00 5.5973 2.29252 Review 149 1.00 10.00 5.6980 1.51882 Und 149 0.00 1.00 0.7718 0.42108 Sex 149 0.00 1.00 0.2550 0.43735 Age 149 19.00 24.00 20.4631 1.71450 GPA 149 1.90 3.99 3.1408 0.47352 Exp 149 0.00 1.00 0.2215 0.41664 Participants Information Group 1 74 Experience No 116 2 75 Yes 33 Sex Female 111 Understanding No 34 Male 38 Yes 115 Note: Group 1(2) is participants who receive good (bad) news followed by bad (good) news. Info0 is the participants' valuation using the beginning information. Both groups receive the same information. Info1, info2, info3, info4, info5 dan info6 are participants' valuation using the first, second, third, fourth, fifth, and sixth information. Both groups receive different orders. Review is participants' valuation after participants evaluate the sixth information. Und is dummy variable of participants' understanding, 1 for understanding and 0, otherwise. Sex is a categorical variable, 1 for male and 0 for female. Age is participants' age. GPA is participants' grade point average. Exp is participants' experience in stock trading or investing, 1 for experience and 0 otherwise. 4.2. Experiment 2 The second experiment was carried out with the support of students who are members of the investment community. Participants of experiment 2 are undergraduate students from different study programs from experiment 1, but all participants of experiments 1 and 2 are from the same faculty at the same university. Descriptively, the results show that valuation based on different information sequences will produce different decisions. This second experiment was conducted to recheck the instrument's rigidity without using positive framing for the bad news. The valuation of Group 1 (2) is 3.8 (6.4) when they receive info6. The bad news is preferred in Group 1, whereas good news is more weighted in Group 2. Our study also adds a feature related to self-review. After receiving info6, participants are required to do a final valuation with no additional information. This selfreview made participants revise their valuation in the opposite direction from valuation info6, although group 2 is still higher than group 1. The visual appearance is 50 Dedhy Sulistiawan, Yie Ke Feliana, Felizia Arni Rudiawarni, Adriana Grigorescu 10.5709/ce.1897-9254.498DOI: CONTEMPORARY ECONOMICS Vol. 17 Issue 1 43-572023 Table 3 Differences Test between Groups: Experiment 1 Group N Mean Standard deviation Std. Error Mean Equality of variance t-value Info0 1.00 74 5.2432 0.85705 0.09963 0.127 2.00 75 5.2267 0.72733 0.08398 n.s Info1 1.00 74 6.0811 1.45015 0.16858 8.186 2.00 75 4.1600 1.41460 0.16334 *** Info2 1.00 74 6.7838 1.64083 0.19074 7.412 2.00 75 4.8400 1.55997 0.18013 *** Info3 1.00 74 6.7297 1.91817 0.22298 9.601 2.00 75 4.0267 1.48845 0.17187 *** Info4 1.00 74 4.7432 1.91648 0.22279 -5.817 2.00 75 6.3200 1.33720 0.15441 *** Info5 1.00 74 4.9459 1.75867 0.20444 -7.719 2.00 75 6.9733 1.43295 0.16546 *** Info6 1.00 74 4.0946 1.99430 0.23183 -10.462 2.00 75 7.0800 1.44970 0.16740 *** Review 1.00 74 5.0270 1.65499 0.19239 -5.927 2.00 75 6.3600 1.00861 0.11646 *** Und 1.00 74 0.7973 0.40476 0.04705 0.733 2.00 75 0.7467 0.43785 0.05056 n.s Sex 1.00 74 0.3108 0.46598 0.05417 1.552 2.00 75 0.2000 0.40269 0.04650 n.s Age 1.00 74 20.6892 0.79253 0.09213 1.608 2.00 75 20.2400 2.27109 0.26224 n.s GPA 1.00 74 3.0800 0.46284 0.05380 -1.564 2.00 75 3.2008 0.47935 0.05535 n.s Exp 1.00 74 0.2568 0.43983 0.05113 1.026 2.00 75 0.1867 0.39227 0.04530 n.s Note: Respectively, ***, **, *, represent statistical significance at the 1%, 5%, 10% levels. n.s is insignificant. Group 1 (2) is the group of participants who receive good (bad) news followed by bad (good) news. Both groups receive different orders. Info0 is the participants' valuation using the beginning information. Both groups receive the same information. Info1, info2, info3, info4, info5 dan info6 are participants' valuation using the first, second, third, fourth, fifth, and sixth information. Review is participants' valuation after participants evaluate the sixth information. Und is dummy variable of participants' understanding, 1 for understanding and 0, otherwise. Sex is a categorical variable, 1 for male and 0 for female. Age is participants' age. GPA is participants' grade point average. Exp is participants' experience in stock trading or investing, 1 for experience and 0 otherwise. www.ce.vizja.pl 57 Do Investors Overreact to COVID-19 Outbreak? An Experimental Study Using Sequential Disclosures This work is licensed under a Creative Commons Attribution 4.0 International License. in belief updating: The belief-adjustment model. Cognitive psychology, 24(1), 1-55. https://doi. org/10.1016/0010-0285(92)90002-J Hribar, P., & McInnis, J. (2012). Investor sentiment and analysts’ earnings forecast errors. Management Science, 58(2), 293-307. https://doi.org/10.1287/ mnsc.1110.1356 Jegajeevan, S. (2012). Return volatility and asymmetric news effect in Sri Lankan stock market.S t a ff Studies, 40(1), 37-57. http://doi.org/10.4038/ ss.v40i1.4680 Jo, H., Shin, E., & Kim, H. (2021). Changes in consumer behaviour in the post-COVID-19 era in Seoul, South Korea. Sustainability 2021, 13, 1-6. https://dx.doi.org/10.3390/su13010136 Kahneman, D. & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263-292. https://doi.org/10.2307/1914185 KANTAR. (2020, June 11). Understand consumer decision-making during COVID-19 and beyond. https://www.kantar.com/inspiration/coronavirus/understand-consumer-decision-making- during-covid-19-and-beyond Kohli, S., Timelin, B., Fabius, V., & Moulvad Veranen, S. (2020). How COVID-19 is changing consumer behavior – now and forever. McKinsey & Company. https://v.fastcdn.co/u/c81ab06a/53497572- 0-how-covid-19-is-chan.pdf KPMG. (2020). Responding to consumer trends in the new reality. COVID-19 pulse survey.. KPMG International. https://home.kpmg/xx/en/home/insights/2020/06/consumers-and-the-new-reality. html Li, N., Richardson, S., & Tuna, I. (2014). Macro to micro: Country exposures, firm fundamentals and stock returns. Journal of Accounting and Economics, 58, 1–20. https://doi.org/10.1016/j.jacceco.2014.04.005 Libby, R., & Tan, H. T. (1999). Analysts’ reactions to warnings of negative earnings surprises. Journal of Accounting Research, 37(2), 415-435. https://doi. org/10.2307/2491415 Mehta, S., Saxena, T., & Purohit, N. (2020). The new consumer behaviour paradigm amid COVID-19: Permanent or Transient? Journal of Health Management, 22(2), 291-301. https://doi. org/10.1177/0972063420940834 Mian, G. M., & Sankaraguruswamy, S. (2012). Investor sentiment and stock market response to earnings news. The Accounting Review, 87(4), 1357-1384. https://doi.org/10.2308/accr-50158 Murphy, J., Hofacker, C., & Mizerski, R. (2006). Primacy and recency effects on clicking behavior.Journal of Computer-Mediated Communication,11(2), 522-535. https://doi.org/10.1111/j.1083- 6101.2006.00025.x Onyele, O., & Nwadike, C. (2021). Modelling stock returns volatility and asymmetric news effect: A global perspective. Financial Risk and Management Review, 7(1), 1-15, https://doi.org/10.18488/ journal.89.2021.71.1.15 Pinsker, R. (2007). Long series of information and nonprofessional investors’ belief revision. Behavioral Research in Accounting, 19 (1), 197-214. https://doi.org/10.2308/bria.2007.19.1.197 Pinsker, R. (2011). Primacy or recency? A study of order effects when nonprofessional investors are provided a long series of disclosures. Behavioral Research in Accounting, 23(1), 161-183. https:// doi.org/10.2308/bria.2011.23.1.161 Quinn, S. (2020). NBA to limit locker room access due to coronavirus; owners to discuss contingency plans for games, per reports. CBS Sports. https://www. cbssports.com/nba/news/nba-reportedly-sent- memo-to-teams-indicating-theyshould-prepare- to-possibly-play-games-in-empty-arenas/ 11 Savor, P., & Wilson, M. (2013). How much do investors care about macroeconomic risk? Evidence from scheduled economic announcements. Journal of Financial and Quantitative Analysis, 48(2), 343–375. https://www.jstor.org/stable/43303804 Seok, S. I., Cho, H., & Ryu, D. (2019). Firm-specific investor sentiment and the stock market response to earnings news. The North American Journal of Economics and Finance, 48, 221-240. https://doi. org/10.1016/j.najef.2019.01.014 Sulistiawan, D., & Rudiawarni, F. A. (2019). Do stock investors need to discuss to reduce decision bias? Investment Management and Financial Innovations, 16(3), 1-9. http://dx.doi.org/10.21511/ imfi.16(3).2019.01 Tuttle, B., Coller, M., & Burton, F. G. (1997). An examination of market efficiency: Information order effects in a laboratory market. Accounting, Organizations and Society, 22(1), 89-103. https://doi. org/10.1016/S0361-3682(96)00026-8 Valaskova, K., Kramarova, K., & Bartosova, V. (2015). Multi criteria models used in Slovak consumer market for business decision making. Procedia Economics and Finance, 26, 174–182. https://doi. org/10.1016/s2212–5671(15)00913–2