Abstract
The purpose of this study is to investigate the initial public offering in Indonesia as one of the largest emerging markets in Asia. The focus of this study is IPOs’ behaviour, which is determined by underpricing and overpricing. The data consist of companies that launched IPOs on the Indonesian Stock Exchange (IDX) from January 2015 to June 2019. The methodology used in this study is an ordinary least square regression with a cross-sectional analysis. The result shows that IPO’s behaviour tends to be underpriced. However, overpricing might occur in several cases. A broad theoretical framework was used to initialize all the predictors of IPO, and we found that the firm age, gap of day, risk, hot periods, corporate social responsibility (CSR) and privatization showed a significant influence, while other variables, that is, IPO size, investor sentiment, rank lead underwriters, market volatility and board lists did not show significant results. This article adds to existing literature by providing a sample from 2015 until the second quarter of 2019 in Indonesia during a time of hot periods with market uncertainty and US–China trade war. This article adds to the signalling theory related to IPO by analysing the impact of CSR on IPO performance.
Keywords
Introduction
Pricing and post-listing performance of an initial public offering (IPO) have been an observed phenomenon by researchers for a long time. The success of IPO and also investors’ revenue are strongly influenced by the issuance price; hence, the IPO price setting plays an important role. Various studies on various capital markets in the world have found evidence of underpriced IPO. Underpricing is always interpreted as the difference between the IPOs’ offer price and the closing market price on the first day when entering the secondary market. Based on previous studies, underpricing tends to be more dominant than overpricing (Harris, 2018; Hoque, 2014; Killins, 2019; Rathnayake et al., 2019; Wang et al., 2019). The most common theory to explain the IPOs’ underpricing is information asymmetry theory in a proxy of shared price ex-ante uncertainty (Hoque, 2014; Rathnayake et al., 2019; Samarakoon, 2010). However, asymmetric information is not the only primary driver of the IPO phenomenon (Ritter & Welch, 2002). Over the years, there have been studies related to this context developed from Western markets to emerging markets. However, based on the author’s observations on google scholar, there are few articles in reputable journals that discuss IPOs in Indonesia.
During 2018, the Indonesian Stock Exchange (IDX) made a new achievement by ranking in the top 10 of the total global IPOs. Fifty-seven new issuers listed their companies in IDX and contributed 4% from 1384 of global IPOs, and become the largest IPOs since 1992, taking Indonesia to the first position in ASEAN IPO 2018, followed by Malaysia with 22 IPOs, and Thailand with 20 IPOs (Ernst & Young, 2019). The number of IPOs in Indonesia has increased significantly over the years: 37 IPOs in 2017, 13 IPOs in 2016 and 16 IPOs in 2015. In 2018, the number of IPOs was quite impressive considering there were a lot of uncertainties both in the global market and domestics market due to the US–China trade war. This achievement is an anomaly in the Indonesian market. There has been a very significant increase in the number of IPOs since the last few years, even though market conditions are currently in uncertainty due to internal and external factors. The significant increase in the number of IPOs resulted in an increase in IPO volume, that is, making it to the hot period, a period that rarely occurs in Indonesia. Therefore, this is the right moment to study IPO behaviour in Indonesia.
In this study, we use 12 variables, that is, issuer size, investor sentiment, privatization issues, offer risk, day gap, company age, market volatility (MVL), hot and cold issues, listed board effect, underwriter effect, the firm-specified characteristic by earning prior to IPO and the potential signals proxied by CSR (Corporate Social Responsibility), to acquire the explanation of IPO initial return in Indonesia.
This article adds to existing literature by providing a sample from 2015 until the second quarter of 2019 in Indonesia, a time of hot period with market uncertainty and US–China trade war. This article adds to signalling theory related to IPO by analysing the impact of CSR on the IPO performance. To the best of the author’s knowledge, such studies have not been conducted in Indonesia so far.
The rest of the article are organized as follows: The second section provides the literature review and hypothesis development. The third section shows the methodology adopted for the study. The fourth section presents the empirical results and discussion, and the fifth section provides the conclusion.
Literature Review
Initial Public Offering Initial Return
Underpricing is a commonly occurring phenomenon in companies that initialize IPO and this occurred in every country with various average underpricing conditions in each country (Loughran et al., 1994). Studies on underpricing are mostly conducted in Western countries (Allen & Faulhaber, 1989; Beatty & Ritter, 1986; Harris, 2018; Hoque, 2014; Killins, 2019; Ljungqvist, 2007; Loughran et al., 1994; Lyn & Zychowicz, 2003), which then spread to other countries such as Middle East and North Africa (MENA) (AlShiab, 2018) and Asia (Badru & Ahmad-Zaluki, 2016; Chen, 2014; Dhamija & Arora, 2017; Handa & Singh, 2017; Huang et al., 2019a, 2019b; Wong et al., 2017; Wang et al., 2019). A study by Loughran et al. (1994) in the USA from 1960 to 1962 showed an average underpricing of 15.2% from 177 companies with an IPO. Recent studies by Killins (2019) on the Canadian market showed an average underpricing of 1.45%, which consists of 73 IPO companies from 2010 to 2017. The average underpricing in Canada is relatively low when compared to other countries. In Asia, the underpricing phenomenon occurs in some countries such as China with an average of 42% (Huang et al., 2019), Malaysia with an average of 7% (Badru & Ahmad-Zaluki, 2016) and India with an average 22% (Dhamija & Arora, 2017). Research on Indonesian IPO shows an average underpricing of 23% from 231 IPO companies from 1995 to 2012 (Husnan et al., 2015). Many studies have been initialized and show the behaviour of underpricing IPOs. Although most studies show the dominations of underpricing, the results of IPOs’ overpricing might occur (Derrien, 2005). Research by Rathnayake et al. (2019) found an average overpricing of 17% to 18% originating from 32 IPOs in Sri Lanka. Research by Song et al. (2014) found an average overpricing of 44–53% depending on the measurement of the intrinsic value.
Theoretical Background
In the efficient market theory, stock prices might change when new information is found to be unpredicted (Ball & Brown, 1968). This information will be quickly responded by the market and the price changes towards a new equilibrium price. The speed of market response then becomes the main factor to create a capital market that provides a reasonable return to investors. However, efficient market is difficult to realize because of irregularities or anomalies that oppose the theory of market efficiency, increasing the level underpricing IPOs (Ţiţan, 2015).
The most common theory in explaining the phenomenon of underpricing is information asymmetry (Chen, 2014; Hoque, 2014; Huang et al., 2019). Companies, underwriters and investors are uncertain about the true value of the company’s shares (Rock, 1986). For Rock (1986), there were two categories of investors—informed investors and uninformed investors. Informed investors have a better understanding about the intrinsic value of shares. Informed investors dare to bid for higher share prices and only buy a good company IPO according to their own opinion. However, uninformed investors have difficulty in knowing the intrinsic value of stock prices due to the lack of relevant information. This results in higher ex-ante uncertainty. Uninformed investors, as a result, would not want to bid high prices on IPO, which inflicts underpricing. The amount of underpricing is directly related to the ex-ante uncertainty regarding stock prices. Beatty and Ritter (1986) state that ex-ante is positively related to underpricing. Several studies have proven the relationship of ex-ante uncertainty to underpricing (Badru & Ahmad-Zaluki, 2016; Killins, 2019; Rathnayake et al., 2019; Samarakoon, 2010). However, these studies were not in line with studies by Wang et al. (1992) and Song et al. (2014) which found a positive relationship between ex-ante uncertainty and overpricing. Some researchers use this ex-ante uncertainty proxy with the age of firm (Badru & Ahmad-Zaluki, 2016; Harris, 2018; Killins, 2019; Komenkul et al., 2017; Rathnayake et al., 2019), offer size (Komenkul et al., 2017; Rathnayake et al., 2019), offer risk (Badru & Ahmad-Zaluki, 2016; Rathnayake et al., 2019) and time lag (Anderson et al., 2015; Rathnayake et al., 2019).
Another model that can be used in explaining the phenomenon of IPO is investor sentiment theory (Ritter & Welch, 2002). When the market gives a positive reaction, the IPO stock returns will increase, resulting in underpricing (Rathnayake et al., 2019). The underpricing phenomenon is also explained by the windows of opportunity theory (Allen & Faulhaber, 1989; Derrien, 2005; Samarakoon, 2010). The increase in the volume of annual IPO shares will trigger a hot market. Investors become irrational and overly optimistic about IPO shares without the support of relevant data, which causes an increase in demand for IPO shares; the company will use this momentum to issue shares to get a benefit from abnormal returns (Ritter & Welch, 2002). However, Ljungqvist (2007) stated a different opinion. If the issuer uses the hot market moment to offer shares, the issuer should be able to set a higher offer price and lower the underpricing. Investors will respond optimistically from the price offered by the company, even though it is higher than it should be.
Signalling theory is the opposite theory to ex-ante uncertainty. In signalling theory, there are two types of companies, high-quality companies and low-quality companies. High-quality companies will signal to the public and deliberately set low IPO prices until underpricing occurs to show that the good quality of the company is different from companies with lower quality (Allen & Faulhaber, 1989; Ross, 1977). Low-quality companies will set prices higher so that they will get more funds from IPO (Welch, 1989). Previous studies investigated the signalling theory in the form of ownership structure (Chan et al., 2004; Downes & Heinkel, 1982), and restricted voting share and CEO compensation (Killins, 2019) to show corporate signals about the quality of the company. Research by Martinez-Conesa et al. (2017) show a significant positive relationship between CSR and firm performance. Companies that are proactive in implementing CSR have a higher performance. CSR can be a signal of company quality. Better companies have stronger abilities to perform CSR when compared to companies that are not performing CSR.
Determinants of Initial Public Offerings’ Initial Performance
Age of the firm is used as a proxy for ex-ante uncertainty. The age of the firm shows the length of time the company was operating prior to IPO. A company that has been operating for a long time has more information available than a new company (Loughran et al., 1994). The information availability can reduce ex-ante uncertainty. A study by Killins (2019) shows a significant negative relationship between the age of firm and initial return. However, studies by Badru and Ahmad-Zaluki (2016) and Rathnayake et al. (2019) show an insignificant relationship. In this study, the age of a firm is expected to have a negative relationship with initial return.
Studies by Anderson et al. (2015) and Rathnayake et al. (2019) use company size based on the board-listed category. In Indonesia, a listed board can show the specific characteristics of a company, which are divided into two, the mainboard and secondary board (Table 1). Larger companies have more information available than smaller companies and reduce the ex-ante uncertainty (Beatty & Ritter, 1986). Previous studies showed a positive relationship between firm size and initial return (Anderson et al., 2015; Badru & Ahmad-Zaluki, 2016). However, a study by Rathnayake et al. (2019) shows an insignificant relationship. We hypothesize a negative relationship between the listed boards and initial returns.
Offer size can be a proxy for ex-ante uncertainty (Beatty & Ritter, 1986). Offer size is obtained by multiplying the offer price by the number of shares offered at IPO (Rathnayake et al., 2019). A larger number of IPOs will result in a smaller return, which implies that a larger IPO has a lower risk (Badru & Ahmad-Zaluki, 2016). In this way, a larger IPO can reduce ex-ante uncertainty. Some studies show a negative relationship between offer size and initial return (Rathnayake et al., 2019; Samarakoon, 2010). However, other studies show no significant relationship (Badru & Ahmad-Zaluki, 2016; Lyn & Zychowicz, 2003). In this study, the following hypothesis is used: offer size has a negative effect on initial return.
Gap day is the gap from the listing date to the first day the company entered the secondary market. Longer time lag will increase ex-ante uncertainty and result in higher underpricing (Rathnayake et al., 2019; Samarakoon, 2010). Studies by Chan et al. (2004) and Rathnayake et al. (2019) show a significant positive relationship between time lag and initial return. This study follows Rathnayake et al. (2019) and makes the gap day variable as an independent variable. We hypothesized that there is a positive relationship between gap day and initial return.
Based on the risk-return trade-off theory, MVL is used to measure market risk and uncertainty (Rathnayake et al., 2019). MVL is calculated using the standard deviation of daily market returns for 30 days before IPO. The higher the volatility, the higher is the level of initial return (Rathnayake et al., 2019). In line with previous studies, we expect a positive relationship between MVL and initial return.
Offer risk used in this studies is based on previous studies by Abdul Rahim and Yong (2010), Badru and Ahmad-Zaluki (2016) and Rathnayake et al. (2019). Offer risk is a reciprocal of the IPO offer price (
Based on the signaling theory, we hypothesize that companies with good quality will signal a lower bid price to indicate the quality of the company, thereby increasing the initial rate of return. In this study, the signalling used is the dummy variable of CSR, with a value of 1 if the company performs CSR and a value of 0 if the company does not perform CSR. In a previous study by Martinez et al. (2017), there was a positive relationship between CSR and firm quality. Therefore, the existence of CSR can be an indication that the company has better prospects and quality than companies that do not perform CSR. In addition, the existence of CSR can be an indication that the company has a stronger ability. Building on this theory, we propose the hypothesis of a positive relationship between CSR and initial return.
When the market reacts positively, the IPO stock return will increase, resulting in underpricing (Anderson et al., 2015; Rathnayake et al., 2019). Investor sentiment is calculated using changes in the value of the Jakarta Composite Index (JCI) 1 month prior to IPO. Studies by Samarakoon (2010), Anderson et al. (2015), Harris (2018) and Rathnayake et al. (2019) showed a positive relationship. In line with the investor sentiment hypothesis, this study expects a positive relationship.
Hot market occurs when there is an increase in the volume of annual share offerings. In the hot market period, investors tend to be irrational and inflict higher initial returns (Ljungqvist, 2007; Ritter, 1991). This study also includes the volume of stock offerings as an independent variable and a positive relationship is expected. Hot market is measured using a dummy variable where 1 denotes hot market and 0 denotes cold market. Studies by Ritter (1991), Samarakoon (2010), Harris (2018) and Rathnayake et al. (2019) found positive results in hot market and initial return. In line with the windows of opportunity theory, this study expects a positive relationship between hot markets and initial returns.
Underwriters are proxied by rank lead underwriters (Badru & Ahmad-Zaluki, 2016; Wong et al., 2017; Killins, 2019). Underwriters will try to suppress the price of an IPO to reduce losses on the purchase of unsold stocks, resulting in underpricing (Ljungqvist et al., 2006). Rank lead underwriter is measured by a dummy variable, number 1 denotes top reputable underwriters, while number 0 denotes non-reputable underwriters. In a previous study by Benveniste et al. (2003), it was found that underwriters have a significant positive effect on initial returns that supports agency theory. However, Megginson and Weiss (1991) found negative relationships, which support the theory of ex-ante uncertainty. Based on the agency hypothesis, we build the hypothesis that rank lead underwriters have a positive effect on underwriters.
Initial public offering shares are not only from independent or conventional companies but also from companies owned by the government or private companies. Biais and Perotti (2002) revealed that the government would set a lower price for the bid price for the political benefit of public voting. Studies such as those by Samarakoon (2010) show a positive relationship, while studies by Lyn and Zychowicz (2003) and Rathnayake et al. (2019) show insignificant results. In this study, it is expected that there is a positive relationship between privatization and initial return.
Objectives
The main objective of this study is to investigate the IPO behaviour, by focusing on the IPO’s initial return and the market-adjusted abnormal return. Furthermore, we analyse the initial return of IPO in Indonesia, whether it is underpricing or overpricing. The underpricing determinant is explained using a broad theoretical framework that can provide an explanation for the initial performance of IPOs in Indonesia, and how the market reacts to Indonesian IPOs.
Rationale
The investigation of Indonesian markets becomes important in light of the global distinctions and upsurge in volumes in IPO market. These conditions generated considerable research interest in Indonesian IPO market. According to Ernst and Young, 2019, IDX made a new achievement by ranking in the top 10 of the total global IPOs in 2018. The study also becomes important because the influences of CSR variables on underpricing are expected to be different in emerging markets, owing to different government rules when compared to developed markets. This research is expected to provide an overview for potential investors about the factors that can affect the initial return of investment of IPO companies. Therefore, this it can be used for consideration of decision-making.
Data Sources, Variables and Methodology
The data consist of companies that initialized IPOs on IDX from January 2015 to June 2019. This period was chosen for this study because it was a time of hot period with market uncertainty and US–China trade war. The total number of companies initializing IPO was 155, but 2 companies had to be eliminated from the sample because they did not fit the data. The final sample used in this study consists of 153 companies. Data related to IPO companies such as offer price, age, size, lag, CSR information, rank underwriters and listing dates were extracted from prospectus reports or company websites and annual reports. Data related to share price, both the company’s stock price and the JCI price, were obtained from IDX.co.id.
Sample Distribution.
Two measurements of initial return are used, initial return and market-adjusted abnormal return (Rathnayake et al., 2019). Factors that are expected to influence initial return are shown in Table 2. Initial return is obtained from the percentage of the difference between the bid price and the market closing price on the first day (Equation 1). Market-adjusted returns are calculated by adjusting the market return index with the initial returns (Equation 2).
where P1 denotes the closing price on the first trading day of the company, P0 denotes the offer price of the company and IR denotes the initial return on the first trading day. Rm denotes the first trading day of the company, Rm = [(Pm1 − Pm0)/Pm0] × 100, Pm1 = the closing market index value on the first trading day of the ith stock and Pm0 = the closing market index value on the listed day of the ith stock.
List of Variables.
The research methodology was adapted from previous studies such as those by Samarakoon (2010) and Rathnayake et al. (2019) by analysing the initial return and independent variables. Then, we conducted univariate regression and stepwise multiple regression. Multiple regressions are used as follows:
Result
Indonesian Initial Public Offering Characteristics
The initial return in Indonesian IPOs from 2015 to the second quarter of 2019 was 84.13%, which indicated underpricing that consisted of 129 companies out of 153 companies. Overpricing also occurs in some companies, although not much when compared to companies that experience underpricing. The results presented in Table 3 show that 20 companies are overpricing with an average of 25.19%, while the rest showed the same price on the first day of return and offering. Similar results are also shown in market-adjusted abnormal return (MAAR). The average of underpricing and overpricing also shows significant differences.
IPO Performance of Underpricing and Overpricing.
Characteristics of Indonesian IPO from 2015 to 2nd Quarter of 2019.
The period between 2017 and the second quarter of 2019 can be categorized as a hot market marked by an increase in the annual share volume when compared with other periods observed. The conclusion regarding the hot market is also strengthened by the fact that Indonesia succeeded in occupying the top 10 global IPOs in 2018 with a significant increase in the number of IPOs compared to previous years. The symptoms of underpricing appear to have increased significantly from 2017 to 2019 which is the Hot Market period (Table 4). The average underpricing in 2017–2019 is in the range of 70–80%. When categorized by the industrial sector, the highest average underpricing occurred in the consumer goods industry sector followed by the infrastructure and finance sectors. The minus number denotes overpricing, but in 2019, it is observed that overpricing is very low compared to other years.
Indonesian Initial Public Offering Performance by the Determinant
Initial return and MAAR show almost the same results in Table 5, except for the AGE category. Younger firms tend to experience underpricing with the highest average with a significant result. This is in line with Killins’ (2019) study which shows a significant negative relationship between firm age and the level of underpricing. Therefore, it can be concluded that younger firms tend to experience higher underpricing. However, underpricing has not only happened to younger firms. Firms with longer operating age also experience underpricing with a significance level of 5% for IR and 10% for MAAR. This indicates that older companies also experience underpricing.
IPO Performance by the Determinant.
There is a positive relation between small size and large size with a significance level of 1%. Underpricing on small size shows high results with an average of 86.27% (Table 5). This implies that the larger offer size tends to experience a lower level of underpricing and has a lower risk (Badru & Ahmad-Zaluki, 2016). The gap in the range of 1–10 days is divided into three subcategories, and there is no significant relationship between the three subcategories, either shorter or longer gaps. This indicates that underpricing occurred with shorter gaps and longer gaps, although higher underpricing occurred at gaps above 10 days with an average IR and MAAR of 73.01% and 71.67% (Table 5).
Investor Sentiment (SENT) is divided into positive sentiment and negative sentiment. Table 5 shows that higher underpricing occurs when there is negative sentiment, but it does not show a significant difference with positive sentiment. The negative differences between positive sentiment and negative sentiment on IR and MAAR were 10.44% and 9.96%, respectively. RISK is divided into subcategories. It can be observed from Table 5, that higher RISK in the range 0.0076–0.01 experiences higher underpricing with an average IR and MAAR of 96.11% and 97.09%, respectively, with a significance level of 10%. A high RISK can also indicate that the offer price tends to be high with a low return on the first day. The results show that IPO with higher risk shows higher underpricing and decreases with decreasing risk level. This is in line with Rathnayake et al.’s (2019) study and supports risk-return trade-off theory.
Underwriter rank shows that companies with non-reputable underwriter’s experience high underpricing with an average of IR and MAAR of 82.63% and 81.83%, respectively (Table 5). The difference between reputable and non-reputable underwriters is significant, with an IR and MAAR difference of 1%. This shows a negative relationship between the underwriter and the initial return. A reputable underwriter will issue shares from a better company, reducing ex-ante uncertainty (Megginson & Weiss, 1991). MVL in Table 5 shows a unique pattern, with a bumpy up and down pattern. With an MVL level of less than 50, the average IR and MAAR are 81.69% and 80.75%, respectively. The level of underpricing is then reduced by MVL in the range of 51–100 with an average of IR and MAAR of 80.52% and 78.80%, respectively. The pattern then returns to MVL with a range of 101–105 with high back underpricing such as MVL less than 50 and then decreases again to a larger MVL. A significant IR average is found in MVLs in the range of 51–100 with a significance level of 10%.
A total of 10 issues companies are categorized as private companies, which are government companies. PRIV precisely shows a significant negative relationship between IR and MAAR. The average initial return on privatization was lower at 5.98% and 5.25% compared to conventional companies with averages of IR and MAAR of 65.97% and 66.13%, respectively. In Table 5, since BRD shows the size of the company, it is clear that companies in the mainboard category experience lower underpricing compared to companies in the development board category with a significant difference of 1%. The level of underpricing that occurred in companies with the main board category was 39.96%, while in the development board category, it was 76.47%. There is a negative relationship between the mainboard and the initial return. This indicates that a larger company will experience lower underpricing when compared to smaller companies.
A fairly high level of underpricing also occurs in companies that initialized IPOs during the hot market period, with an average underpricing of 79.32%. The average IR at the time of cold markets is smaller at 15.50%, which is significant at the level of 1% (Table 5). This supports the windows of opportunity theory which states that investors tend to be irrational and increase the volume of IPOs. The company will take this moment to issue its shares. Significant differences also exist from the differences between companies that perform CSR, but the direction obtained is negative. The average IR for companies that perform CSR is 54.75%, while it is 92.66% for companies that do not perform CSR (Table 5). This shows that companies with the ability to perform CSR will experience lower underpricing. Further, this does not support signalling theory, but supports the ex-ante uncertainty theory companies with better firm quality characterized by performing CSR and reducing the level of underpricing.
Univariate Regression
The next step is to regress the 11 variables which are the factors that influence underpricing. Regression is performed by regressing the independent variables one by one with the dependent variables, that is, IR and MAAR. Robust standard error is used in the regression model. The results of the regression between the factors and the dependent variables IR and MAAR do not show a significant difference.
Univariate Regression for Initial Return.
Univariate Regression for Market-adjusted Abnormal Return.
It can be seen from Table 7 (univariate regression) that LnSIZE, LnRISK, RANK, PRIV, BRD and HOT have a significant influence on IR and MAAR with a significance level of 0.01. LnAGE and CSR variables have a significant influence on IR and MAAR with a significance level of 0.5, while SENT, MVL and LnGAP do not have any significant influence.
In Table 7, LnAGE, LnSIZE and BRD have a negative relationship, which supports ex-ante uncertainty. Companies with longer history will reduce the level of underpricing. The size of the company indicated by the category of mainboard and development board shows that companies in the mainboard category have lower underpricing levels. Similarly, a larger offer size will reduce the level of underpricing. RANK shows the opposite direction of our hypothesis. RANK has a negative coefficient, which indicates that reputable underwriters will actually reduce the level of underpricing. RANK’s proxy does not support agency theory, but instead supports the ex-ante uncertainty theory. This indicates that the rank underwriter will issue shares of good companies and reduce ex-ante uncertainty, thereby reducing the level of underpricing.
LnRisk shows a positive coefficient, which indicates a higher level of risk will increase the level of underpricing. PRIV also shows the opposite direction of the hypothesis with a negative coefficient. This indicates that conventional companies have a higher level of underpricing than state-owned private companies. CSR has a significant negative influence, but it goes against the direction of the hypothesis. CSR has a negative influence which indicates that good companies that perform CSR will actually reduce the level of underpricing. This declines the signalling theory that good companies will actually have a high level of underpricing, which is a good signal for the company. HOT shows a positive relationship and supports the windows of opportunity theory.
Multiple Regression
As shown in Table 8, (correlation matrix), there is no mutually substitute relationship between independent variables with numbers below 0.5. IR and MAAR show a high number, but the coefficient occurs between the dependent variables that do not affect the regression.
Correlation Matrix.
Table 9 shows the estimation results from multiple regression, and there is no significant difference from the IR and MAAR regression results. There are five variables that have a significant influence on underpricing. From the results shown in Table 9, LnAGE shows a negative coefficient on underpricing with a significance level of 10%.
Estimation Results of Multiple Regression.
These results are in line with Killins’ (2019) study which shows that firm age has a negative relationship with the level of underpricing. This indicates that companies with longer operation will reduce the level of underpricing. Therefore, our study supports the ex-ante uncertainty theory which is proxied by firm age. The longer the operational life of the company will reduce ex-ante uncertainty and reduce the level of underpricing.
LnGAP has a positive influence on IR and MAAR with a significance level of 5%. This is in line with the studies by Chan et al. (2004) and Rathnayake et al. (2019). The difference in days between the offer period and the first day in the secondary market will increase the level of underpricing, thereby supporting ex-ante uncertainty.
Tables 5–9 show consistent results on the LnRISK variable, that is, the higher the level of risk, the greater is the level of underpricing. In Table 9, LnRISK has a positive coefficient with a significance of 5%. The results are in line with studies conducted by Badru and Ahmad-Zaluki (2016) and Rathnayake et al. (2019) and support the risk return trade-off theory. PRIV has the opposite effect of the proposed hypothesis. Although not in line with the studies by Samarakoon (2010) and Rathnayake et al. (2019), we actually found that government-owned private companies actually have a negative influence on the level of underpricing with a significance level of 1%. This indicates that state-owned companies actually tend to have more potential to experience underpricing when compared to conventional private-owned companies. CSR has the opposite effect on the hypothesis. The results show that companies performing CSR had lower levels of underpricing. The research results reject signalling theory, but they support ex-ante uncertainty theory which states that a better company, characterized by the company’s ability to perform CSR, can reduce ex-ante uncertainty theory and reduce the level of underpricing. HOT denotes the hot market period with an increase in volume on the annual IPO. Table 9 shows that HOT has a positive coefficient with a significance level of 1%. This is in line with the studies of Samarakoon (2010) and Harris (2018) and accepts the windows of opportunity hypothesis. Investors become irrational during the hot market and market returns can be higher than usual; hence, the company will use this moment to issue its shares (Ljungqvist, 2007).
Conclusion and Implications
Studies on IPO show IPO underpricing behaviour, not only in Western markets but also in developing markets. Studies by Husnan et al. (2015) in Indonesian IPOs showed an average underpricing of 23% from 231 companies in 1995–2012. In this study, we used a sample of 153 IPOs from 2015 to 2nd quarter of 2019 and the results show a high level of underpricing with an average 84.13%. Overpricing occurred in some companies with an average of 25.19% from 20 companies. We used two measurements of underpricing, IR and MAAR. The two measurements showed the same result.
We used a broad theoretical framework to explain the phenomenon of IPO. OLS and cross-sectional regression models were used to investigate the determinants of IPO underpricing. We found that the firm age, gap of day, risk, hot periods, CSR and privatization showed a significant effect. However, other variables such as IPO size, investor sentiment, rank lead underwriters, MVL and board lists did not show significant results. In this article, we add CSR and find that CSR has a significant effect on the initial return of an IPO. CSR decreases the level of underpricing, indicating the companies with CSR activities tend to have a better prospect and quality than companies without CSR activities. In addition, the existence of CSR indicates that the company has a stronger financial ability, even though it shows a hypothesis that is contrary to the signalling theory. However, CSR is a factor that can strengthen the ex-ante uncertainty theory which states that companies with stronger finances, as shown by CSR activities, will get a lower level of underpricing because companies dare to bid for higher prices during the IPO period. For further study, we suggest examining the relationship between CSR and IPO initial returns using different measurements.
Footnotes
Acknowledgement
The authors are grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of the article. The usual disclaimers apply.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship and/or publication of this article: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
