Impact of Founder on the IPO Flipping Activity during Pre and Post-Global Financial Crisis

In the Initial Public Offering (IPO), when investor sentiment has changed, the trading pattern consistently shows unusual behavior such as selling pressure or excessive demand. This selling pressure is even more pronounced during the period of the Global Financial Crisis (GFC) experienced in 2008-09. Earlier studies have shown the influence of the founder is also vital in day-to-day business decision making. Studies that look into family-controlled firms and small firms (Claessens, Djankov & Lang, 2000), CEO-founder and firm performance (Jayaraman, Khorana, Nelling & Covin, 2000), and the persistence of founder influence (Nelson, 2003). There are limited studies on both founder and IPO during the GFC period. Moreover, this paper focuses on the IPO firms listed in the Main and ACE Market from the Malaysian stock market during the GFC period. Hence, this paper fills the gap by looking at the impact of the founder on flipping activities during GFC periods. As mentioned above, different investor sentiments create abnormal trading activity in the IPO market known as IPO anomaly (Aggarwal 2003; Ellis, 2006; Lowry, Officer & Schwert, 2010). One of the reasons that cause the abnormal trading activity is the liquidation from the investors (flippers) on the first day of trading. According to Aggarwal (2003), the flipper is defined as the investor who liquidates the share obtained from IPO. The process of liquidation of the IPO by flipper on the first day or first few days of trading immediately after the listing is Indonesian Capital Market Review 12 (2020) 69-84


Introduction
In the Initial Public Offering (IPO), when investor sentiment has changed, the trading pattern consistently shows unusual behavior such as selling pressure or excessive demand. This selling pressure is even more pronounced during the period of the Global Financial Crisis (GFC) experienced in 2008-09. Earlier studies have shown the influence of the founder is also vital in day-to-day business decision making. Studies that look into family-controlled firms and small firms (Claessens, Djankov & Lang, 2000), CEO-founder and firm performance (Jayaraman, Khorana, Nelling & Covin, 2000), and the persistence of founder influence (Nelson, 2003). There are limited studies on both founder and IPO during the GFC period. More-over, this paper focuses on the IPO firms listed in the Main and ACE Market from the Malaysian stock market during the GFC period. Hence, this paper fills the gap by looking at the impact of the founder on flipping activities during GFC periods.
As mentioned above, different investor sentiments create abnormal trading activity in the IPO market known as IPO anomaly (Aggarwal 2003;Ellis, 2006;Lowry, Officer & Schwert, 2010). One of the reasons that cause the abnormal trading activity is the liquidation from the investors (flippers) on the first day of trading.
According to Aggarwal (2003), the flipper is defined as the investor who liquidates the share obtained from IPO. The process of liquidation of the IPO by flipper on the first day or first few days of trading immediately after the listing is known as the flipping activity (Aggarwal, 2003;Che-Yahya, Abdul-Rahim & Mohd-Rashid, 2015). The flipping activity enables the flipper to benefit from the abnormal return on the first trading day. For instance, traders prefer to buy and sell the shares in a short period to gain an abnormal profit from the price difference. Sometimes, if the trading activities are excessive, it could negatively influence investors' decision making as they focus on long-term investment rather than speculation, especially on the first day of trading.
Besides, the anomaly of IPO initial return indicates that the investor sells the share when it is traded at a higher price than the offer price. The sharp increase in price enables the investor to obtain a considerable profit and causes higher flipping activity (Carter and Dark, 1993;Boehmer and Fishe, 2000;Habib and Ljungqvist, 2001;Islam and Munira, 2004;Ellis, 2006;Lowry et al., 2010). One of the possible reasons for the higher flipping activity is the investor's confidence level toward the firm's senior management structure. The high-level management structure represents founder, is defined as the founder held the top management position at the time of the IPO such as Chief Executive Officer, senior management executive, director, or no official role in the firm (Gao and Jain, 2012) In Malaysia, most of the businesses are family-based (Claessens, Djankov, and Lang, 2000). This reflects the decision made by the top management is more conservative, especially the founder. This unique feature allows investors to have a higher confidence level to the company performance, and it encourages the flipping activity on the first day of trading. Hence, the behavior of the investor will change if the same person holds the position of founder and the top management in a firm. In the context of this study, the role of founder and its impact on flipping activity will be examined based on the different positions held in a firm. This study proposes two possible explanations of the relationship between founder and IPO activities. First, investors are more confident with the positive signal that translates into a higher debut price of an IPO by the founder. Second, the higher debut price will attract irrational investors to liquidate the share and rational investors to buy the stock, developing conducive trading for flipping activity. This reflects the strong belief of a firm as the investor continues to hold despite the exit opportunity from an IPO to obtain profit. This paper fills the gap in the literature on the relationship between founder and flipping activity.
The motivation of this study comes from two fronts. First, the majority of Malaysian firms are family-owned businesses (Claessens, Djankov, and Lang, 2000). This allows family members to hold numerous vital positions. This will provide greater control in decision making, and the founder will be able to focus on the growth of the firm. Therefore, investors have a higher confidence level despite a more conservative decision-making process.
Second, the reputation of top management is also one of the critical considerations for investors. For instance, investors can assess the senior management's track record for creating shareholders' wealth, such as the growth of market share, cost reduction, or expansion of the new market. However, not all businesses are family-owned; some firms have a higher risk due to the change of the management team or corporate governance practice. This paper fills the research gap by examining the impact of the founder on the IPO flipping activity.
Besides, this study also examines both listed firms in the Main and ACE Markets across the GFC periods. The Main market consists of many large-cap firms with q total market capitalization of a minimum of RM500 million. Besides, the listing requirements are positive cash flow, the uninterrupted profit of three to five financial years, and the aggregate after-tax profit of at least RM20 million. These requirements are not required for ACE Market that comprises of new IT or technology startup firms. The firm age in the Main Market is also longer than the ACE Market. Due to these differences in firm size, age, and capital structure, it is expected the outcome will be different from the Main and ACE Market.
In our study, flipping activity refers to the liquidation process, which will be influenced by initial return, oversubscription ratio, firm age, offer period, firm size, and founder. Moreover, the role of the founder is emphasized because they play an essential role for a firm when they come to decision making, especially during GFC. This study shows the relationship between the independent variable and the dependent variable. It examines the Main Market and ACE Market in the effect of the founder of the flipping activity. This study contributes to the existing literature as the investors benefited from knowing the impact of the founder on IPO flipping activities in an emerging market.
The remainder of the paper is organized as follows. Section two reviews the past literature, followed by data and methodology. Section four discusses the results. The last session concludes the study.

Literature Review and Hypothesis Development
The research on pricing anomaly in an IPO has become the primary concern related to the flipping activity. Flipping activity refers to the liquidation of the share of an IPO within a short period, such as the first day of trading or subsequent trading. This is known as "stagging activity" (Bayley, Lee & Walter, 2006). It also means a quick profit to the investor and greater liquidity in the subsequent trading.
According to Booth and Chua (1996), the flipping activity provides greater liquidity in an IPO on the first day of trading. This benefits the investor with a lower cost of trading and optimizes the cost of capital for an issuer. However, the excess of flipping activity will be detrimental to the initial return of an IPO. Instead, the investor is encouraged to dispose of the share on a staggered basis.
The research from Miller and Reilly (1987), Shultz and Zaman (1994), Ellis, Michaelly and M O'Hara (2000) shows a positive relationship between flipping activity and the initial return of the IPO performance, especially for hot IPO is being flipped more than the cold IPO (Aggarwal, 2003). They argue that aftermarket trading volume as a proxy by flipping activity is essen-tial in explaining the predictive power of IPO performance.
The result contradicts Krigman, Shaw, and Womack (1999), they examine the underwriters' pricing error and predictive power of IPO performance. They find a negative relationship between flipping activity and initial return. This concludes that the underwriter sets the lowest offer price intentionally of an IPO to push for higher flipping activity. Therefore, H1 is developed.
H1: There is a positive relationship between initial return and flipping activity.
Besides, the impact of the market condition remains compelling in the flipping activity. According to Aggarwal (2003), an institutional investor flips more in cold IPO when providing price support on the first few days of trading. The result is consistent with Krigman et al. (1999), they show that flipping is more evident in the large trading volume for a weak IPO than in hot IPO. This concludes that flipping activity is not necessary for low trading volume in a weak IPO or high trading volume in hot IPO.
According to Che-Yahya et al. (2014), they study the impact of the lock-up period for the flipping activity. This study focuses on the length of time for significant shareholders or promoters to hold on to the IPO. They found a negative relationship between the lock-up period and flipping activity. This is because a more extended lock-up period causes significant shareholders to hold the shares for good IPO. Our study is different from Che-Yahya et al. (2014), first, the position of the founder and the senior management in a firm held by the same person. Second, the founder and the senior management play a role in business decision making, but there is no role for the shareholders. Lastly, shareholders concern more about the share performance, but the founder focuses more on business performance. Therefore, H2 to H6 are developed.
H2: There is a positive relationship between the oversubscription ratio and flipping activity. H3: There is a positive relationship between firm age and flipping activity. H4: There is a positive relationship between the offer period and flipping activity. H5: There is a positive relationship between firm size and flipping activity.
The founder-CEO has to lead the firm in terms of founding vision, positive image, organizational influence, and ownership stakes. The empirical evidence shows that IPO firms are going public with non-founder CEOs at the helm. Earlier studies The study on founder-CEO is different from a founder-led firm to nonfounder led firms such as investment choices, decision-making processes, governance structure, ownership structure, and firm performance (Jayaraman, Khorana, Nelling & Covin, 2000;Nelson, 2003).
According to Certo, Covin, Daily, and Dalton (2001) and Nelson (2003), half of the IPO firms that go for listing with founder-CEO tend to receive higher IPO initial returns compared to non-founder led firms.
Besides, Wasserman (2003) suggests that the founder-CEO departure is event-driven and tends to occur around essential milestones such as completion of product development and new financing for the firm. However, relatively little is known regarding how a significant event such as an IPO influences founder-CEO succession. Therefore, this shows an essential area of research in flipping activity (Wasserman, 2003). Thus, H6 is developed.
H6: There is a positive relationship between founder and flipping activity.
Based on the above discussion, the research gap is identified. In most of the IPO markets, the founder or senior management does not have the full authority to interfere in IPO trading, especially in Malaysia. The majority of Malaysian firms are family-based businesses, and this affects the selling decision of investors and influences flipping activity. However, none of the studies has examined the importance of the founder on the flipping activity.
In testing the hypothesis, this paper highlights that flipping activity is also influenced by other IPO feature such as initial return, oversubscription ratio, firm age, offer period and firm size (Beatty and Ritter, 1986;Dawson, 1987;Carter and Manaster, 1990;Ritter, 1984;Yong, 1996;Kautia, 2004;Guo and Brooks, 2009). This paper also examines the interaction between the founder and other variables such as initial return, oversubscription ratio, firm age, offer period, and firm size to the flipping activity. Therefore, H7 to H11 are developed.
H7: There is an interaction between the founder and initial return H8: There is an interaction between founder and oversubscription ratio H9: There is an interaction between the founder and firm age H10: There is an interaction between the founder and the offer period H11: There is an interaction between the founder and firm size The hypotheses, H1 to H6 is used to test the relationship in equation (1), we expect that initial return, oversubscription ratio, firm age, offer period, firm size and the founder have a positive relationship with flipping activity.
H7 to H11 is used to examine to what extent the founder influences the relationship between initial return, oversubscription ratio, firm age, offer period, firm size, and flipping activity. The firms are family-based businesses, and senior management focuses more on business performance. It is commonly believed that the interaction with the founder will have an impact on the flipping activity on the first day of trading.
Also, the existing literature is insufficient to support the argument that the founder has an impact on flipping activity based on the dif-ferent segments of the capital market, such as the Main and ACE Market. Hence, this study extends the hypothesis to test the effect of the founder in the Main market. Hence, H12 to H17 are developed.
H12: There is an interaction between the founder and Main market H13: Initial return in the Main market enhances the flipping activity H14: Oversubscription ratio in the Main market enhances the flipping activity H15: Firm age in the Main market enhances the flipping activity H16: Offer period in the Main market enhances the flipping activity H17: Firm size in the Main market enhances the flipping activity Similar hypotheses are also set up for the ACE market. Hence, H13 to H23 are developed.
H18: There is an interaction between the founder and ACE market H19: Initial return in the ACE market enhances the flipping activity H20: Oversubscription ratio in the ACE market enhances the flipping activity H21: Firm age in the ACE market enhances the flipping activity H22: Offer period in the ACE market enhances the flipping activity H23: Firm size in the ACE market enhances the flipping activity

Theoretical Framework
According to Jain and Tabak (2008), signaling theory from the sociological perspective suggests a framework to identify the factors that influence the founder of the entrepreneur firm. For instance, Davis (2005) highlights that investors are aware of the signaling theory that is used to evaluate IPO firms.
However, the choice of founder versus nonfounder CEO of the IPO firm is likely to be influenced by various factors. The track record of the founder will stimulate higher demand for the stock and lead to higher valuations. Hence, signaling theory provides the theoretical justification for the selection of factors that are hypothesized to influence the choice between founder versus non-founder CEO of IPO firm. This paper examines the interaction of the founder and other independent variables for the Main Market and the ACE Market during pre-GFC, GFC, and post-GFC period. The conceptual framework and relevant hypotheses are shown in Figure 1. The beginning period of 2006 marks the time when Bursa Malaysia has started to provide investors with more sophisticated trading tools, in addition to a more informative website with IPO like offer price, closing price, offer period, and the unit provided. For this study, Firm age and founder are obtained from firms' annual reports. Over-subscription ratio and trading volume are derived from the newspaper on the next day of IPO debut.

Data and Methodology
The multiple regression model is used to predict the value of IPO flipping activity based on the number of independent variables consist of IPO initial return, oversubscription ratio, firm age, offer period, firm size, and founder management. These variables, together with the dependent variable, are shown in the conceptual framework ( Figure 1).
The models are motivated by the idea that there exists more selling pressure of stock (flipping activity) during the period of GFC. Due to fear or other emotional distress, investors with a negative market outlook will engage in abnormal trading activity in the IPO market, either in the Main Market or ACE Market (Aggarwal, 2003).
As highlighted in earlier studies, a few control variables that are frequently used in different IPO studies such as initial return, oversubscription ratio, firm age, offer period and firm size are used for this paper (Dawson, 1987;Yong, 1996;Carter & Manaster, 1990;Ritter, 1984;Guo & Brooks, 2009;Wasserman, 2003). Refer to the above. Therefore, model 1 to model 6 is developed. Also, the interaction effect of founder and one of the control variables are used to gauge the flipping activities or selling pressure during the GFC period of the stock, especially for family-based business (Jain and Tabak, 2008;Gao and Jain, 2012).
In equation (1), the dependent variable is IPO flipping activity as measured by the liquidity, FA i = (volume/unit offer) x100. From equation (2) and (6), the interaction between the founder and other variables in Main and the ACE Mar-ket are studied. The founder is defined by the position of the founder or senior management held in a firm.
FA i = θ 0 +θ 1 IR i +θ 2 OSR i +θ 3 LogAge i +θ 4 LogOP i +θ 5 LogSize i +θ 6 DFounder i +θ 7 DAce i +θ 8 DFounder i *DAce i +ε i FA i = λ 0 +λ 1 IR i +λ 2 OSR i +λ 3 LogAge i +λ 4 LogOP i +λ 5 LogSize i +λ 6 DFounder i +λ 7 DAce i +λ 8 IR i *DAce i +λ 9 OSR i *DAce i +λ 10 LogAge i *DAce i +λ 11 LogOP i *DAce i +λ 12 LogSize i *DAce i +ε i Where: FA i is a flipping activity measured by a percentage that denotes the trading volume divided by the number of shares issued on the first day of trading of the ith company. (Aggarwal, 2003) IR i is an initial return measured by a percentage that denotes the change in price from the offer price to the closing price on the first day of trading of the ith company. (Dawson, 1987) OSR i is an oversubscription ratio measured by several times that denotes an IPO issue that is either over-demanded or under-demanded by the group of investors of the ith company. (Yong, 1996) LogAGE i is firm age measured by a number of years that denotes the from the establishment to its listing of the ith company. A log transformation is applied because of its positive skewness. (Carter and Manaster, 1990;Ritter, 1984) LogOP i is an offer period measured by the number of days that denotes the duration of the subscription of the ith company. A log transformation is applied because of its positive skewness. (Guo and Brooks, 2009) LogFSZ i is the firm size measured by Ringgit Malaysia (RM) that denotes the number of the unit offers multiply by the offer price of the ith company. A log transformation is applied because of its positive skewness. (Wasserman, 2003) DFounder i is the value of 1 if the senior management of the issuing firm at the time of the IPO is also the founder. (Jain and Tabak, 2008;Gao and Jain, 2012). DMain i = 1 represents the IPOs listed in the Main Market, and zero otherwise.
DAce i = 1 represents the IPOs listed in the ACE Market, and zero otherwise. There are two arguments for equation (3) to (6); Firstly, different sentiment enables investors to gain a quick profit through IPO liquidation on the first day of trading on Main or ACE Market. Secondly, the underwriter provides price support during a weak market sentiment. This will affect investor demand for IPO and also flipping activity. Equation (1) and (2) refer to the whole capital market, equation (3), and (4) focus on the Main Market and equation (5) and (6) focus on the ACE market.
Equation (1) is the baseline model. Model (3) and (5) extend the baseline model with the founder and its interaction with the Main or ACE market, respectively. Besides, Model (4) and (6) Table 1 shows descriptive statistics of IPO flipping activity, initial return, oversubscription ratio, firm age, offer period, and firm size in the whole, pre-GFC, during GFC and post-GFC periods. As observed, there is a low initial return and demand during GFC, but investors participate more frequently in IPO trading (flipping activity) as compared to pre-GFC and post-GFC. Due to the weak economy during GFC, the IPO demand (oversubscription ratio) is low. This is because the confidence level of investors toward the newly issued share is small, and the higher uncertainty of the IPO performance compared to pre-GFC and post-GFC.

Results and Discussions
Despite the low initial return during GFC, investors can obtain an average return of 17.09% in the long term, as shown in the whole period. The result is consistent with Leow and Lau (2018a) and Leow and Lau (2018b) as there are still positive initial returns among the new is-sues in the GFC period. This is vital for flipping activity because it will increase the liquidity of IPO.
Besides, the positive skew of flipping activity indicates that investors are keen on liquidating the IPO on the first trading day. The right-skewed distribution of initial return and oversubscription ratio shows an IPO closes at higher first-day returns when there is a greater demand. The skewness of firm age, offer period, and firm size indicates investors are more active to subscribe to an IPO for a well-established IPO firm because of the stable financial performance. The Jarque-Bera test shows the variables used in the analysis are normally distributed. Table 2 presents the correlations analysis between the explanatory variables. Notably, the variables such as IPO flipping activity, initial return, oversubscription ratio, firm age, offer period, and firm size used in the analysis are not highly correlated for the whole period, 76         pre-GFC, during GFC and post-GFC periods. Therefore, there is no multicollinearity in the model. As observed inTable 3 for the entire period from 2006 to 2016, firstly, the models show a positive relationship between oversubscription ratio and flipping activity and a negative relationship between firm size and flipping activity. Moreover, the interaction of founder and OSR is negative and significant. It means that founder participation in business will reduce the flipping activity.
A similar observation can be found in Table  4 during the pre-GFC period; the oversubscription ratio is positively related to the flipping activity. However, the negative interaction between founder and oversubscription ratio reduces the effect of the flipping activity. It can be explained that investors have higher confidence in the capability of the founder or senior management who participates in the business activities of the IPO firm. Therefore, investors have less tendency to engage in flipping activity.
For the main market, it is found that firm age increases while the offer period reduces the flipping activity. For the ACE market, firm age reduces while the offer period increases flipping activity. The IPO offer period has a positive impact on flipping activity. This is because the investor has more time to study the financial performance of listed firms. The result is consistent with Guo and Brooks (2009), a study on the duration between the offering and listing period for Chinese A-share IPOs. Table 5 presents the result for the GFC period. Firm age is the main factor that explains the flipping activity. Besides, the result also shows there is no difference between the Main Market and the ACE Market during GFC.
As shown in Table 6 for the post-GFC period, the oversubscription ratio has a positive impact on flipping activity. A renewed confidence has accelerated the first trading day. Besides, firm size has a negative effect on flipping activity. Although the confidence level has recovered. , investors prefer to hold on to the large scale of IPO. This creates a low flipping activity on the first day of trading. This also shows that the investors have changed their trading behavior 81 Table 7 (flipping activity) from GFC to the post-GFC period.

Conclusion
We summarize all the results from table 3 to 6 in Table 7. It can be observed that the founder, by itself, does not have any impact on flipping activities.
Among the key findings are: Firstly, the founder, by itself, does not have any effect on flipping activity. However, the interaction of founder and oversubscription ratio reduces flipping activity in the pre-crisis period. Secondly, the interaction of founder and firm age is significant during the GFC period. Thirdly, in post-GFC, the interaction between founder-firm age, founder-offer period are essential factors. Overall, it is found that firm age and IPO offer period have an impact on flipping activity in the Main and ACE Market.
Many listed firms are initially started by founder and subsequently managed by themselves or professional managers. Though these firms are listed eventually, the founder still involves making a day-to-day business decision. It is often believed that such firms are more conservative than others, and hence, these firms have more stable performance relative to others. It is not surprising that firm age and founder are an essential combination of our findings.
The result from sub-sample periods reveals almost similar findings, irrespective of whether it is Main or ACE Market. Besides firm age, the interaction between founder and IPO offer period is essential as this allows prospective investors to have more time to study the newly listed firms. These findings on emerging markets like Malaysia serve as an interesting case study for others in further understanding investors' behavior in emerging markets.