The Post-issue Market Performance of Initial Public Offerings: Empirical Evidence from the Malaysian Stock Markets

Abstract

This study examines the long-run performance of the initial public offerings (IPOs) listed in the Malaysian main and alternative ‘Access, Certainty and Efficiency’ (ACE) markets at the economic and sectorial levels. Using event- and calendar-time study methods and monthly data from January 2000 to December 2011, we provide novel evidence on the existence of under performance anomaly in the Malaysian markets and more intensely in the ACE markets. We demonstrate robust evidence on the distinction in sector-specific characteristics from the aggregate market characteristics. While the consumer products and industrial sectors dominate the overall underperformance, the construction, property and technology sectors significantly overperform. The findings are robust to a wide range of other sensitivity checks including parametric and non-parametric tests.

JEL Classification: G14, G15, G30, G34, G32, G38

Keywords

Initial public offerings (IPOs)long-run performance underperformance emerging economy

1. Introduction

The long-run abnormal performance of initial public offering (IPOs) is extensively reported in the literature. One possible explanation for long-run underperformance is the divergence of opinion hypothesis proposed by Miller and Reilly (1987), which assumes investors to be overly optimistic about the future cash flow and growth potential of IPOs. Such over-optimism drives investors to value the IPO stocks at higher than their intrinsic values and as more information are revealed over time, the optimistic investors will downward adjust the stock valuation towards their intrinsic values, and this narrows the gap in their divergence of opinion to be more in line with pessimistic investors. As stock price drops, return subsequently drops over the long run, and Ritter (1991) hypothesised that IPOs with the highest initial returns will perform worst in the long run.

The ongoing debates on IPOs revolve around whether under- performance exists and if the abnormal returns are indeed sensitive to the method employed in calculating the stock returns. A more recent strand of research conducted in the area extended the analysis by comparing the abnormal performance of IPOs across countries, types of IPOs and nature of listing boards, to better understand if IPOs’ behaviour may be attributed to those different characteristics.

Gregory, Guermat, and Al-Shawawreh (2010) conducted a study on IPOs listed on the main and alternative boards of the London Stock Exchange and found that despite IPOs underperforming on all markets, the main market IPOs (MM IPOs) tend to beat the performance of the IPOs of alternative boards. Another study by Vismara, Paleari, and Ritter (2012) examined IPOs listed on different boards in France, Germany and Italy and found similar behaviour exists in France and Italy. In Germany, however, the alternative market IPOs (AM IPOs) perform better than the MM IPOs. Similarly, in the context of developing countries, Komenkul, Brzeszczynski, and Sherif (2012) studied the Thai market and found IPOs in the Market for Alternative Investments (MAI) outperform their main market (known as the Securities Exchange of Thailand [SET]) counterparts. Komenkul et al. (2012) employed the cumulative abnormal returns (CAR) and buy-and-hold returns (BHAR) approaches using different benchmarks in calculating the abnormal returns of the IPOs, and they further found the results to differ based on the methods adopted.

In the context of Malaysia, the alternative market, known as the Malaysian Exchange of Securities Dealing and Automated Quotation (MESDAQ) market, was established in 2002 to provide a listing avenue for younger, smaller and growth-driven firms. In 2009, the MESDAQ market was renamed as ‘Access, Certainty and Efficiency’ (ACE) market. The main and ACE markets differ in terms of size, age and financial status, with the listing requirements being more stringent for the main market. The leniency for the ACE market listing implies that the stocks carry more ex ante uncertainties compared to their main market counterparts. Furthermore, most of the ACE market stocks are from the technology sector, that are often associated with higher ex ante risks. Hence, it is expected that IPOs listed in the two markets will portray different long-run behaviour (Ahmad-Zaluki & Kect, 2012). However, a direct comparison between the main and the ACE markets is missing in the literature. Hence, this study fills the gap by examining whether IPOs investments in the main Malaysian stock market has superior performance compared with the ACE (alternative) market and whether sector-specific characteristics differ from the aggregate market characteristics, in predicting and determining the IPO performances. Most of the literature compares between technology IPOs to their non-technology counterparts (refer to e.g., Gao & Jain (2011), Ahmad-Zaluki & Kect (2012) and Saade (2015), among others). Although Ahmad-Zaluki, Campbell, and Goodacre (2007) study cross-sectional returns of Malaysian IPOs by sector grouping, we extend this line of enquiry by comparing the abnormal returns for 3 years and eight discrete sector groupings.

Another contribution of this study is in terms of methodology. We ensure robustness in our abnormal returns models by applying the self-constructed size/book-to-market (BTM) matching portfolio and industry indices as benchmark, in addition to the traditional market index. We employ both parametric and non-parametric techniques to determine the significance of the abnormal returns. Further, for the calendar-time approach, we employ three asset pricing models, namely, the Fama–French three-factor (FF3F) model, the Carhart momentum model (MOM) and the liquidity-adjusted capital asset pricing model (LCAPM). Our study focuses on newer data set, using IPOs listed from January 2000 to December 2011.

The Malaysian context deserves attention considering its rankings as the third best emerging economy behind China and South Korea and twentieth in the Global Competitiveness Index (Schwab, 2013). Among Southeast Asian economies, only Malaysia and Singapore made it to the top 20 (Schwab, 2013) in this prestigious index. Furthermore, with the GDP growth rate of 4.7 per cent, Malaysian economy is expanding faster than Singapore (3.9%) and Thailand (1.8%) (The World Bank, 2015). It is important to also note that the Malaysian capital market is forecasted to grow at a compounded annual growth rate (CAGR) of 8.5 per cent from 2010 to 2020 (Securities Commission Malaysia, 2011) and expected to double in size from MYR 1.28 trillion in 2010 (US$366 billion) to MYR2.43 trillion (US$695 billion) in 2020. To further ensure that the capital market remains attractive to foreign direct investors, the Malaysian government continues to improve its economic policies to accommodate the dynamic global economic climates.

Our main analysis indicates that underperformance anomaly does exist in the Malaysian market, and the magnitude is higher in the ACE market. We further find the magnitudes of long-run underperformance to change when different methods are employed. Our calendar-time analysis consistently shows evidence of underperformance in Malaysian IPOs in both the whole market and segmented markets.

The remainder of this article is structured as follows. The following section discusses the framework of IPO long-run abnormal performance. The third section describes the data and methodology employed in the current study. The fourth and fifth sections present empirical findings using event- and calendar-time methods, respectively. And the sixth section concludes the article.

2. Theoretical Framework

2.1 Segmentations of Bursa Malaysia

Since the deregulation of the Malaysian capital market in 1996, it has experienced tremendous development. Previously known as the Kuala Lumpur Stock Exchange (KLSE), the Malaysian capital market was renamed Bursa Malaysia in 2004. Securities are listed in two distinct markets, the main and the ACE markets. In this study, we conduct comparative analysis of 36-month performance of the main and ACE markets IPOs.

The main and ACE markets are different with respect to market characteristics and listing requirements. Panels A and B of Table 1 summarises the key characteristics and listing requirements for each market, respectively.

Table 1.
Characteristics of Malaysian Stock Markets

Characteristics Main Market ACE Market

Panel A: Key Characteristics of Malaysian Stock Market

Establishment 1964 2002 (Previously known as the MESDAQ market)

Size (as at 31 December 2014) MYR 1,633 billion
(US$467 billion) MYR 9.7 billion
(US$2.8 billion)

Number of listed securities (as at 31 December 2014) 799 (88.2%) 107 (11.8%)

Composition Mixed Mixed, but mainly consists of technology companies

Panel B: Key Listing Requirements in Bursa Malaysia

Firm age One financial year prior to submission for listing application No minimum age requirement

Profit after tax (PAT) requirement 3-year uninterrupted PAT, minimum aggregate of MYR 20 million (US$5.72 million)
Minimum PAT of MYR 6 million (US$1.72 million) for the most recent financial year No minimum eligibility requirements

Market capitalisation MYR 500 million (US$143 million) No minimum eligibility requirements

Characteristics	Main Market	ACE Market
Panel A: Key Characteristics of Malaysian Stock Market
Establishment	1964	2002 (Previously known as the MESDAQ market)
Size (as at 31 December 2014)	MYR 1,633 billion (US$467 billion)	MYR 9.7 billion (US$2.8 billion)
Number of listed securities (as at 31 December 2014)	799 (88.2%)	107 (11.8%)
Composition	Mixed	Mixed, but mainly consists of technology companies
Panel B: Key Listing Requirements in Bursa Malaysia
Firm age	One financial year prior to submission for listing application	No minimum age requirement
Profit after tax (PAT) requirement	3-year uninterrupted PAT, minimum aggregate of MYR 20 million (US$5.72 million) Minimum PAT of MYR 6 million (US$1.72 million) for the most recent financial year	No minimum eligibility requirements
Market capitalisation	MYR 500 million (US$143 million)	No minimum eligibility requirements

Source: Adopted from Bursa Malaysia official website (http://www.bursamalaysia.com)

The different characteristics and listing requirements between the main and ACE markets indicate that stocks listed in the respective markets carry different risk profiles. Due to higher leniency in the listing requirements for ACE market, the stocks listed on this market tend to exhibit higher risk profile. Accordingly, investors would expect higher returns to adequately compensate for the associated level of risks borne by them.

2.2 The Framework of IPO Long-run Abnormal Returns

One of the main anomalies of IPOs is the long-run underperformance, that occurs when IPOs exhibit poorer performance in comparison to their benchmarks. Overall, the majority of the studies provide empirical evidence of underperformance and that the level of abnormal returns is sensitive to the methodologies employed in determining the IPO performances.

The early finding of Ritter (1991) triggers global attention on this strand of research. More recent studies in the US market consistently document long-run underperformance for the US IPOs (Eckbo & Norli, 2005; Gao & Jain, 2011; Brau, Couch, & Sutton, 2012), thus supporting Ritter (1991). Similar studies conducted across the global market, such as in the UK (Khurshed, Mudambi, & Goergen, 1999; Espenlaub, Gregory, & Tonks, 2000), Spain (Álvarez & González, 2005), Germany (Bessler & Thies, 2006), Japan (Kirkulak, 2008) and Mauritius (Agathee, Sannassee, & Brooks, 2014), all report poor long-run abnormal performance of IPOs.

Despite the vast evidence of underperformance, few studies have reported overperformance in some markets such as in Australia (da Silva Rosa, Velayuthen, & Walter, 2003), Indonesia (Emasari & Tamara, 2010) and in France and Italy (Vismara et al., 2012). The mixed evidence may be attributed to market characteristics, time period of study or methodological factor. The literature suggests that the magnitude, direction and statistical significance of IPOs long-run abnormal returns are sensitive to the methods employed in measuring the returns (Kothari & Warner, 1997; Fama, 1998). A change in method, benchmark or weighting measurement may cause the abnormal performance to change. For example, Chorruk and Worthington (2010), Emasari and Tamara (2010), Erdogan (2010), Thomadakis, Nounis, and Gounopoulos (2012) and Agathee et al. (2014) document that CAR vary from BHAR even when the same benchmark is used. On the other hand, large variations in abnormal returns across different benchmarks are documented by Espenlaub et al. (2000), Drobetz, Kammermann, and Walchli (2005), Gregory et al. (2010), Gao and Jain (2011) and Brau et al. (2012). Further, equal-weighting of abnormal returns may produce different abnormal returns from value-weighting, as documented by Levis (1993), Kooli and Suret (2004), Locke and Gupta (2009), Erdogan (2010) and Agathee et al. (2014).

The event-time approach carries some limitations. For example, Fama (1998) and Mitchell and Stafford (2000) highlight that event-time approach suffers from cross-sectional correlations among individual firms, hence they suggest the use of calendar-time approach to mitigate such potential misspecification. It is important to note that the FF3F model (1993) is commonly employed due to the notion that firm size and BTM effects play a major role in explaining stock returns. Previous studies also document that the long-run underperformance documented using event-time approach disappears when calendar-time approach is used (e.g., Levis, 1993; Erdogan, 2010; Moshirian, Ng, & Wu, 2010; Su & Bangassa, 2011). In contrast, Mazouz, Saadouni, and Yin (2008) find consistent evidence of underperformance using the calendar-time approach.

A more recent strand of research focuses on comparative analysis between two groups of IPOs, segmented by firm characteristics related to ex ante uncertainties such as listing boards. For example, Gregory et al. (2010) and Vismara et al. (2012) conduct comparative analysis on IPOs listed in the main markets and alternative markets in the UK and European countries (France, Germany and Italy), respectively. Gregory et al. (2010) report that IPOs listed in the UK’s alternative investment market (AIM) perform poorer than the MM IPOs in 36-month period using equally weighted BHAR. Vismara et al. (2012) report similar findings in France and Italy, indicating that larger IPOs perform better than smaller IPOs. Conversely, in the German market, AM IPOs perform better than the MM IPOs.

In the context of emerging markets, Komenkul et al. (2012) adopt a similar approach by comparing 36-month abnormal performance of Thai IPOs listed in the main and alternative markets, known as the SET and the MAI, respectively. They demonstrate that MAI IPOs outperform their SET counterparts, which is consistent with the findings on the German market (Vismara et al. (2012).

In short, the debates in the literature on IPOs long-run abnormal performance revolve around whether underperformance exists in the focal market(s) around the study period and how different methodologies cause variations in the abnormal performance level. Some studies (Komenkul, Sherif, & Xu, 2016) offer further insights with comparative analysis between two or more groups of IPOs. Comparative analysis between two listing boards is still non-existent in the Malaysian market. Previous studies that conduct comparative analysis focus on privatisation versus non-privatisation IPOs (Paudyal, Saadouni, & Briston, 1998) and growth versus value IPOs (Corhay, Teo, & Tourani Rad, 2002). Other studies tend to focus either on the entire market or on one listing board (Jelic, Saadouni, & Briston, 2001; Ahmad-Zaluki et al., 2007; Ahmad-Zaluki & Kect, 2012). Overall, in the Malaysian context, studies that are robust in terms of methodology are still scant. By far, the most methodologically robust study in the Malaysian context is that of Ahmad-Zaluki et al. (2007), whereby they employ both event- and calendar-time approaches and use different benchmarks and weighting methods.

Previous studies in Malaysia (Corhay et al., 2002; Ahmad-Zaluki et al., 2007; Ahmad-Zaluki & Kect, 2012) show variations in results when using different methods, indicating that the abnormal returns of Malaysian IPOs are sensitive to the IPOs methodology as in other markets. The majority of the literature shows no evidence of underperformance in the Malaysian market. In fact, Paudyal et al. (1998), Jelic et al. (2001), Corhay et al. (2002) and Ahmad-Zaluki et al. (2007) report overperformance in Malaysian MM IPOs. However, a recent study by Ahmad-Zaluki and Kect (2012) show evidence of long-run underperformance in the MESDAQ market (currently known as the ACE market). It is worthy to mention that the majority of the existing studies in the context of Malaysia cover the period pre-2000 or prior to the 1997 Asian financial crisis. Hence, our study is one of the earliest to employ more recent data.

In their pioneer study focusing on the MESDAQ market, Ahmad-Zaluki and Kect (2012) highlight the possible influence of listing board on IPO long-run abnormal returns. However, the difference in long-run behaviours of IPOs listed in the main market and the ACE market has yet to be documented in the literature. Hence, it is imperative to conduct a comparative analysis on the long-run abnormal performance of IPOs listed in the two markets. Accordingly, our study fills the gap in the literature by attempting to determine whether the IPOs listed in the main and ACE markets portray different long-run behaviours.

We further contribute to the literature by extending the work of Ahmad-Zaluki et al. (2007) by constructing a new benchmark portfolio based on size and BTM, thus mimicking the small-minus-big (SML) and high-minus-low (HML) factors of the FF3F (1993) model. We examine how the Malaysian IPOs perform in comparison to firms with similar size and BTM characteristics. Indeed, the reference portfolio construction in the Malaysian market using this approach is novel to this study. In addition, we study the industry effect on long-run abnormal return of IPOs. For calendar-time approach, we employ three asset pricing models. We first employ the FF3F (1993) model to consider the impact of size and BTM values in long-run abnormal returns. Next, we employ the MOM (1997) which considers the impact of security returns in the previous period on the IPO returns and finally, the LCAPM that considers the impact of stock liquidity on IPO long-run abnormal performance. We further contribute to the literature by incorporating the three asset pricing models in the calendar-time analysis. Our study, therefore, is the first to employ MOM and LCAPM in the Malaysian context.

Finally, as mentioned earlier, we focus on a novel data set using IPOs listed from January 2000 to December 2011. Since the full recovery from the Asian financial crisis of 1998, the Malaysian economy has grown dramatically. Subsequently, this study provides a more current insight of the development of the Malaysian IPO market in particular and the Malaysian capital market in general.

To summarise, the primary objective of this study is to test the hypothesis that in the long-run, Malaysian IPOs show negative abnormal performance. Second, this study is conducted to determine whether IPOs behave differently when listed in markets with distinct characteristics. Third, we attempt to determine whether the IPO long-run abnormal returns in Malaysia are sensitive to the methods employed in calculating the returns.

3. Data and Methodology

3.1 Description and Sources of Data

The data used in this study are IPOs listed in Bursa Malaysia from January 2000 to December 2011. We adopt January 2000 because the Malaysian market had just recovered from the 1998 Asian financial crisis. The exclusion of IPOs listed during the crisis prevents inclusion of abnormally high or low performance during the period. A total of 476 IPOs was listed in Bursa Malaysia during the whole study period.

We exclude 28 IPOs that belong in the financial services industry (including banking, investment houses and insurance companies), real estate and real estate investment trusts IPOs from the sample due to differences in financial reporting requirements. This procedure is consistent with Ahmad-Zaluki et al. (2007), allowing direct comparison of results. The filtration resulted in a final sample size of 448 IPOs, representing 94.12 per cent of the total Malaysian IPOs listed during the study period. Out of this final sample, 296 (66.07%) and 152 (33.93%) IPOs are listed on the main and ACE markets, respectively. Two IPO firms that were delisted within less than 36 months are also excluded to avoid survivorship bias.

Table 2 represents the sample distribution by year and industry, segregated by listing boards. As indicated in Panel A, the maximum number of IPO sample is n = 72 in year 2005, while the minimum is n = 14 in 2009. The highest and lowest number of sample from the main market is 43 (2002) and 11 (2009), respectively. From the ACE market, the highest is 46, appearing in 2005, while the lowest is 3, as seen in 2007 and 2009.

Table 2.
Distribution of Malaysian IPOs Listed between 2000 and 2011

Panel A: IPO Sample Distribution by Year and Listing Board

Entire Sample % of Total Main Market IPOs % of Total ACE Market IPOs % of Total

2000 38 8.48 38 12.88 n/a n/a

2001 20 4.46 20 6.78 n/a n/a

2002 49 10.94 43 14.58 6 3.92

2003 56 12.50 37 12.54 19 12.42

2004 69 15.40 39 13.22 30 19.61

2005 72 16.07 27 8.81 46 30.07

2006 35 7.81 14 4.75 20 13.73

2007 21 4.69 18 6.10 3 1.9

2008 22 4.91 14 4.75 8 5.23

2009 14 3.13 11 3.73 3 1.96

2010 27 6.03 21 7.12 6 3.92

2011 25 5.58 14 4.75 11 7.19

Total 448 296 153

Panel B: IPO Sample Distribution by Industry and Listing Board

Entire Sample
% of Total
Main Market IPOs
% of Total
ACE Market IPOs
% of Total

Construction 7 1.56 7 2.36 0 0.00

Consumer Products 86 19.20 80 27.03 6 3.95

Industrial 138 30.80 110 37.16 28 18.42

Heavy Machinery 21 4.69 5 1.69 16 10.53

Plantation 9 2.01 9 3.04 0 0.00

Property 12 2.68 12 4.05 0 0.00

Technology 93 20.76 13 4.39 80 52.63

Trading and Services 82 18.30 60 20.27 22 14.47

Total 448 296 152

Panel A: IPO Sample Distribution by Year and Listing Board
2000	38	8.48	38	12.88	n/a	n/a
2001	20	4.46	20	6.78	n/a	n/a
2002	49	10.94	43	14.58	6	3.92
2003	56	12.50	37	12.54	19	12.42
2004	69	15.40	39	13.22	30	19.61
2005	72	16.07	27	8.81	46	30.07
2006	35	7.81	14	4.75	20	13.73
2007	21	4.69	18	6.10	3	1.9
2008	22	4.91	14	4.75	8	5.23
2009	14	3.13	11	3.73	3	1.96
2010	27	6.03	21	7.12	6	3.92
2011	25	5.58	14	4.75	11	7.19
Total	448		296		153
Panel B: IPO Sample Distribution by Industry and Listing Board
	Entire Sample	% of Total	Main Market IPOs	% of Total	ACE Market IPOs	% of Total
Construction	7	1.56	7	2.36	0	0.00
Consumer Products	86	19.20	80	27.03	6	3.95
Industrial	138	30.80	110	37.16	28	18.42
Heavy Machinery	21	4.69	5	1.69	16	10.53
Plantation	9	2.01	9	3.04	0	0.00
Property	12	2.68	12	4.05	0	0.00
Technology	93	20.76	13	4.39	80	52.63
Trading and Services	82	18.30	60	20.27	22	14.47
Total	448		296		152

Source: Authors’ own, based on information sourced from Bursa Malaysia official website & authors' calculations.

As shown in Panel B of Table 2, IPO samples cover eight sector groupings, namely, construction, consumer products, industrial, heavy machinery,¹ plantation, property, technology and trading and services. When IPO samples are segregated by sector, the maximum number of sample is from the industrial sector (n = 138), while the minimum is associated with the construction sector (n = 7). The majority of the main market sample is associated with the industrial sector (n = 110) and the lowest is from the heavy machinery sector (n = 5). In contrast, the majority of the firms listed in the ACE market is from the technology sector (n = 80) with no firms from the construction, plantation and property sectors.

The data are obtained from a number of sources. IPO offer prices are obtained from the Information Department of Bursa Malaysia. Share prices, market-to-book equity values, market capitalisations and the prices for both the market and industry indices are obtained from DataStream.

3.2 Methodology

3.2.1 Abnormal Return Model

We investigate the long-run performance of IPO stocks from the first month to the thirty-sixth month post-listing, excluding the initial returns due to the abnormally high returns that are often found in the first trading day IPOs. The monthly raw returns of stock i, R_i, is calculated using the closing price of stock i at the end of the last trading day of month t, P_it.² The benchmark-adjusted abnormal return (AR) of IPO stock i, excluding the initial returns, is the difference between monthly raw return (R_it) and a monthly benchmark return (R_mt), then AR is identified as follows:

A R_{i t} = R_{i t} - A R_{m t} ​

(1)

We use the FTSE Bursa Malaysia Kuala Lumpur Composite Index (FBM KLCI) to benchmark the MM IPOs, while AM IPOs are benchmarked against the FTSE Bursa Malaysia ACE Index (FBM ACE). The literature extensively documents that abnormal returns are sensitive to the benchmark adopted in the analysis (Barber & Lyon, 1997; Kothari & Warner, 2004). The lack of consensus on the most accurate benchmark emphasises the importance of testing multiple models to control for potential misspecification and to compare the sensitivity of outcomes. Therefore, in this study, we use reference matching portfolio and industry indices returns as alternative benchmarks.

3.2.2 Cumulative Abnormal Returns

According to Ritter (1991) and Fama and French (1993), CAR is a traditional performance measure and is the summation of the mean benchmark-adjusted ARs during the 36-month aftermarket time period. It involves monthly portfolio rebalancing to achieve monthly equal weighting. For the benchmark-adjusted returns, AR_t, is the equally weighted arithmetic mean:

A R_{i, t} = \frac{1}{n} \sum_{i = 1}^{s} A R_{i, t} ​

(2)

The CAR is subsequently calculated as follows:

C A R_{q, s} = \sum_{t = q}^{s} ω_{i} A R_{t} ​

(3)

where ω is the equal (value) weighting of ARs.³

To test if CARs are significantly different from zero, we use the conventional t-statistic that is identified as follows:

C A R_{t, m o n t h} = \frac{\overset{─}{C A R_{i, t}}}{σ (C A R_{i, t} / \sqrt{n})} ​

(4)

where σ is the standard deviation (SD) of the ARs, and n is the number of IPOs in the month t. 

3.2.3 Buy-and-hold Abnormal Returns

The BHAR approach does not involve monthly rebalancing of portfolios. It assumes that the securities are held from the purchase date up to the selling date, reflecting the actual experience of investors. In this study, we assume 36-month holding period starting from the day after the listing day. Following Ritter (1991), we exclude listing day due to the abnormally high initial returns usually found in the first listing day. BHAR is used to mitigate the potential upward bias in CAR due to the accumulation of the monthly abnormal returns, as suggested by Barber and Lyon (1997) and Kothari and Warner (2006). Following Loughran and Ritter (1995), BHAR for IPO stock i is identified as the geometrically compounded return on equity in time t and calculated as follows:

B H R_{i, T} = [\prod_{t = s t a r t}^{\min (T, d e l i s t)} (1 + r_{i t}) - 1] \times 100 % ​

(5)

where r_i is the IPO stock returns in month t, start is the first listing period (month), and min (T,delist) is the earlier of the 36-months window or the final month of the listed trading.

The mean BHAR for IPO stocks and the benchmarks are calculated as follows:

\overset{─}{B H R_{T}} = \sum_{i = 1}^{n} ω_{i} B H R_{i T}

(6)

Consequently, the benchmark-adjusted BHAR for IPO stock i for holding-period T, BHAR_i is identified as follows:

B H A R_{i T} = [\prod_{t = s t a r t}^{\min (T, d e l i s t)} (1 + r_{i t}) - 1] - [\prod_{t = s t a r t}^{\min (T, d e l i s t)} (1 + r_{m t}) - 1]

(7)

where r_it is the IPO stock IPO returns at month t and r_mt is the benchmark monthly return at time t. An overperformance (or underperformance) over the benchmark is indicated by a positive (or negative) BHAR value.

As long-run abnormal returns suffer from potential skewness bias (Lyon, Barber, & Tsai, 1999), we employ Johnson (1978) bootstrapped skewness-adjusted t-statistics to examine whether the ARs are significantly different from zero, using the following:

t_{s a} = \sqrt{n} (S + \frac{1}{3} \overset{˘}{γ} S^{2} + \frac{1}{6 n} \overset{˘}{γ}) ​

(8)

and

\overset{˘}{γ} = \frac{\sum_{i = 1}^{n} {(B H A R_{i, t} - \overset{─}{B H A R_{t}})}^{3}}{n σ {(B H A R_{t})}^{3}} ​

(9)

3.2.3 Wealth Relatives

Wealth relatives (WR) is the ratio of the end-of-period wealth from holding a portfolio of IPOs to the end-of-period wealth from holding a benchmark portfolio (Ritter, 1991). It serves as an indicator of overall long-run relative performance and is calculated as follows:

W R = \frac{1 + \overset{─}{B H R_{t, 36}}}{1 + \overset{─}{B H R_{m, 36}}}

(10)

where BHR_i,36 is the 36-month BHAR of IPO i and BHR_m,36 is the 36-month BHAR of the respective benchmarks. A wealth relative value higher than 1.00 indicates overperformance, while a value less than 1.00 indicates underperformance.

3.2.4 Alternative Benchmark 1: Size/BTM Reference Portfolio

Following Komenkul et al. (2016), the size/BTM matching portfolios are constructed using the value-weighted average returns on six portfolios derived from the FF3F model: big-high (BH), big-medium (BM), big-low (BL), small-high (SH), small-medium (SM) and small-low (SL). We follow Fama and French (1993) procedure in constructing portfolios for the size⁴ and BTM equity.

3.2.5 Alternative Benchmark 2: Industry Indices

In addition to listing board effect, we further study the industry impact on IPO long-run abnormal returns. The IPOs are segregated into eight sectors as per Bursa Malaysia classifications, as illustrated in Table 1. The industry-adjusted abnormal return (AR) of IPO stock i is the difference of its monthly raw return (R_it) and a monthly industry index return (R_st), excluding the initial returns. Therefore, the industry-adjusted AR is identified as follows:

A R_{i t} = R_{i t} - R_{s t} ​

(11)

where R_st is the monthly industry index returns of the sector in which the IPOs are first listed in. The sector indices are as per FTSE Bursa Malaysia industry indices for each of the eight industries that are listed in Table 1.

3.2.6 Calendar-Time Approach

Event-time method suffers from cross-sectional correlations among individual securities (Fama, 1998; Mitchell & Stafford, 2000; Gao & Jain, 2011; Sherif, Komenkul, & Xu, 2016; Sherif & Erkol, 2017). Therefore, the statistical significance of mean abnormal returns may be overstated. Kothari and Warner (2006) argue that most return models employed in the event-time approach for calculating abnormal stock performance appear misspecified over different periods. Hence, to mitigate the potential cross-sectional correlations and misspecification issues, previous researchers suggest the use of calendar-time approach (Fama & French, 1996; Gompers & Lerner, 2003; Ahmad-Zaluki et al., 2007; Agathee et al., 2014). Here, we employ the FF3F model (1993), the Carhart (1997) momentum model (MOM) and the LCAPM. For calendar-time approach, the preceding 36-months excess returns of IPO is regressed against the factors over the whole study period to obtain the Jensen’s α, a value that indicates the magnitude and direction of abnormal performance of the IPOs.

3.2.6.1 FF3F Model

Fama and French (1993) report that firm size and BTM values are significant in explaining stock returns. Similarly, Drew and Veeraraghavan (2002) find that the stock return variation in Malaysia is affected by size and BTM. Following Espenlaub et al. (2000) and Komenkul et al. (2016), we employ FF3F model as a benchmark, in which the AR is estimated as follows:

R_{i t} - R_{f t} = α + β_{i} (R_{m} - R_{f}) + β_{S M B} S M B_{t} + β_{H M L} H M L_{t} + ε_{i t} ​

(12)

where R_mt is the market return in month t calculated using the FBM KLCI; R_it is the IPO return in month t; R_ft is the 1-month Malaysian government security return in month t; SMB is the value-weighted return on a portfolio of SML firms and HML is the value-weighted return on a portfolio of high BTM firms minus the value-weighted return on firms with low BTM. The betas are estimated by regressing the preceding 36-month IPOi excess average returns (monthly) on the average market excess returns (monthly), size factor returns and BTM for the 36-month estimation period.

The SMB and HML portfolios in Equation (13) are constructed using Fama and French (1993) methodology by considering each event in month t and sorting the stocks listed in Bursa Malaysia by market capitalisation and BTM value. First, the stocks are sorted into small (S) and big (B) according to their market capitalisation, divided using the median market capitalisation at end of June every year. Next, the stocks are divided into three BTM groups based on the top 30 per cent, middle 40 per cent and bottom 30 per cent, representing high (H), medium (M) and low (L) BTM values, respectively. The BTM value is the reciprocal of market-to-book value at the end of December each year. Finally, six portfolios are constructed, SL, SM, SH, BL, BM and BH. Then, the SMB_t value is derived from the difference between the average returns of small and big firms. The HML_t value is the average returns of IPOs with high BTM ratio minus low BTM ratio. Hence the following:

S M B_{t} = \frac{(S L + S M + S H)}{3} - \frac{(B L + B M + B H)}{3} ​

(13)

H M L_{t} = \frac{(S H + S L)}{2} - \frac{(B H + B L)}{2} ​

(14)

3.2.6.2 Carhart Momentum Model

Thomadakis et al. (2012) estimated the Carhart 4-factor momentum (MOM) model (1997) as a benchmark to calculate abnormal return of IPO in Greece and reported that stock returns momentum is significant in explaining stock performance. Securities returns in the previous period are found to have impact on returns (Carhart, 1997). Subsequently, we use the model in our study to assess its impact on abnormal returns of Malaysian IPOs. The MOM model is then defined as follows:

\begin{matrix} R_{i t} - R_{f t} = α + B_{i} (R_{m} - R_{f}) + β_{S M B} S M B_{t} + β_{H M L} H M L_{t} \\ + β_{M O M} M O M_{t} + ε_{i t} \end{matrix}

(15)

where MOM is the value-weighted average return of the top 30 per cent securities with the maximum returns in the preceding year, and bottom 30 per cent securities with the lowest securities in the preceding year. The MOM value is derived from the excess returns of high-and-low momentum securities.

3.2.6.3 Liquidity Capital Asset Pricing Model (LCAPM)

In addition to FF3F and MOM models, following Chen and Sherif (2016), we employ the LCAPM model that considers the liquidity risk factor of a security, and it is presented as follows:

R_{i t} - R_{f t} = α + B_{i} (R_{m} - R_{f}) + β_{L I Q} L I Q_{t} + ε_{i t}

(16)

where βLIQLIQ_i is the liquidity risk factor of stock i (Amihud, 2002), and it is calculated as follows:

L I Q_{i t} = \frac{1}{D_{i y}} \sum_{t = 1}^{D_{i y}} \frac{| r_{i t} |}{D v o l_{i t}}

(17)

where LIQ_iy is the measure of IPO firm i estimated in month t (Amihud, 2002); D_it is the number of non-zero trading days in month t, |r_it| is the absolute value of returns of stock i in month t; and Dvol_it is Malaysian Ringgit (MYR) trading volume for stock i in month t.

4. Empirical Findings

4.1 Descriptive Statistics

We begin our analysis with the summary statistics. Table 3 exhibits the descriptive statistical data, more specifically, the mean, standard deviation (SD), minimum (Min) and maximum (Max) of the market value, the overall proceeds and the initial returns of Malaysian IPOs. Panel A in Table 3 presents the mean size of the IPO firms, represented by the market capitalisation in Malaysian Ringgit (MYR). The mean market capitalisation for main market (MYR 659.73 million) IPO firms is notably larger than their ACE market counterparts (MYR 71.85 million). The average size of IPO firms in the main market segmentation is higher than the average size of all firms listed in Bursa Malaysia (MYR 460.27 million), implying that some of the main market firms are pronouncedly larger than the ACE market firms. The size of the MM IPO firms is also more dispersed than the ACE market, as implied by the standard deviation of 3,351.70 and 58.57 for the respective markets. In the main market, the size difference between the largest and smallest IPO firms is extremely large that is MYR 40.32 billion for the largest and only MYR 44 million for the smallest. The difference between the maximum and minimum market values in the ACE market is much smaller: the highest market capitalisation is MYR 322 million, while the smallest is MYR 16 million.

Table 3.
Descriptive Statistics of Market Value, Gross Proceeds and Initial Returns of Malaysian IPOs

Mean Std. Dev. Minimum Maximum

Panel A: Market Value of IPOs (MYR million)

All IPOs (n = 448)
MM (n = 296)
AM (n = 152) 460.27
659.73
71.85 2,737.28
3,351.70
58.57 16.00
44.00
16.00 40,320.00
40,320.00
322.00

Panel B: Gross Proceeds of IPOs (MYR million)

All IPOs MM AM 128.51
186.99
14.61 815.70
999.01
10.46 0.05
0.05
1.10 12,499.20
12,499.20
80.80

Panel C: Initial Returns of IPOs by Markets

All IPOs MM AM 0.27
0.23
0.36 0.48
0.37
0.63 −0.71
−0.39
−0.71 2.64
1.94
2.64

Panel D: Initial Returns of IPOs by Industry

Construction 0.27 0.36 −0.01 1.05

Consumer products 0.27 0.44 −0.67 1.80

Industrial 0.20 0.40 −0.71 2.64

Heavy machinery 0.32 0.37 −0.26 0.93

Plantation 0.16 0.19 −0.06 0.60

Properties 0.04 0.22 −0.29 0.42

Technology 0.35 0.61 −0.51 2.46

Trading and services 0.35 0.53 −0.52 2.62

	Mean	Std. Dev.	Minimum	Maximum
Panel A: Market Value of IPOs (MYR million)
All IPOs (n = 448) MM (n = 296) AM (n = 152)	460.27 659.73 71.85	2,737.28 3,351.70 58.57	16.00 44.00 16.00	40,320.00 40,320.00 322.00
Panel B: Gross Proceeds of IPOs (MYR million)
All IPOs MM AM	128.51 186.99 14.61	815.70 999.01 10.46	0.05 0.05 1.10	12,499.20 12,499.20 80.80
Panel C: Initial Returns of IPOs by Markets
All IPOs MM AM	0.27 0.23 0.36	0.48 0.37 0.63	−0.71 −0.39 −0.71	2.64 1.94 2.64
Panel D: Initial Returns of IPOs by Industry
Construction	0.27	0.36	−0.01	1.05
Consumer products	0.27	0.44	−0.67	1.80
Industrial	0.20	0.40	−0.71	2.64
Heavy machinery	0.32	0.37	−0.26	0.93
Plantation	0.16	0.19	−0.06	0.60
Properties	0.04	0.22	−0.29	0.42
Technology	0.35	0.61	−0.51	2.46
Trading and services	0.35	0.53	−0.52	2.62

Source: Bank Negara Malaysia.

Note: MYR 1 is equivalent to US$0.2860 as at 31 December 2014.

Panel B in Table 3 reports the statistics of offer size and indicates that on average the offer size of Malaysian IPOs is MYR 128.51 million. The main market firms have larger average issuance size (MYR 186.99 million) as compared to their ACE market counterparts (MYR 14.61 million). The dispersion of issuance size in the main market is higher than the ACE market, as depicted by the standard deviation of 999.01 and 19.46 for the respective markets⁵.

Panel C in Table 3 displays the initial returns of IPOs, and the average initial return of Malaysian IPOs is 27 per cent. The AM IPOs are more underpriced than the MM IPOs at 36 per cent and 23 per cent respectively, implying that the global phenomenon of IPO underpricing also exists in the Malaysian markets. It is worth noting that the listing regulation allows the ACE market to list younger firms, even those without prior operational and financial track record. Hence, the information asymmetry between investors and IPO firms is higher in the ACE market than in the main market, leading to a higher underpricing scale. More importantly, the underpricing reported for the period of our study is lower than those recorded in previous studies in Malaysia (e.g., Dawson (1987): [166.7%]; Yong (1991): [167.4%]; Loughran, Ritter, and Rydqvist (1994): [80.3%]; Paudyal et al. (1998): [61.8%]; Jelic et al. (2001): [99.1%]; and according to Ahmad-Zaluki et al. (2007): [95.2%]). This implies a downward trend in the underpricing of Malaysian IPOs, reflecting the increased efficiency of the Malaysian market.

Panel D in Table 3 further illustrates the initial returns by sector groupings. As can be seen from the table, events in each industry have distinct impacts on IPOs, and the impact of sector characteristics on IPOs returns varies between sectors. For sector groupings, we found that IPO underpricing is mainly dominated by the characteristics associated with the properties sector (4%), followed by technology and trading and services sectors (35%). This is justified as technology firms are generally younger and smaller, thus suffered from higher ex ante uncertainty. In addition, the majority of those firms are listed in the ACE market, which also partially explain the higher underpricing found in the market.

4.2 Event-Time Approach

4.2.1 Cumulative Abnormal Returns

Now we move on to examine the performance of Malaysian IPOs in the whole economy, the MM and ACE markets, as well as sector groupings using the market index and reference portfolios. While Panels A and B in Table 4 report the 3-year equally and value-weighted CARs of Malaysian IPOs for the whole market, Panel C presents the results of the test of differences in CARs for the two markets.

Table 4.
Three-Year Cumulative Abnormal Returns

Market Index
Reference Portfolio

CAR (%) Para. t-stat Para. t-stat Non-para. s-stat CAR (%) Para. t-stat Para. t-stat Non-para. s-stat

Panel A: Equally Weighted Cumulative Abnormal Returns (EWCAR)

All IPOs
MM
AM −2.86
−14.16
19.23 −0.69
−3.05**
2.40* −0.72
−3.72**
2.83* −1.92*
−4.11***
2.16* 1.70
−0.90
6.90 0.40
−0.19
0.80 0.69
−0.18
0.43 −0.45
−1.09
0.59

Panel B: Value-weighted Cumulative Abnormal Returns (VWCAR)

All IPOs MM AM −24.20
−20.44
−47.16 −23.05*
−18.94*
−14.90* −13.74*
−11.65*
−8.59* −6.01*
−9.43*
−7.90* −15.10
−10.66
−34.64 −23.49*
−18.4*
−7.262* −13.26*
−12.06*
−7.094* −10.23*
−9.72*
−5.32

Panel C: Main Market Versus ACE Market IPOs

Para. t-stat Non-para. z-stat Para. t-stat Non-para. z-stat

EWCAR −3.86* 3.96* −0.855 0.99

VWCAR −4.32* 4.02* −5.33 4.52

	Market Index	Reference Portfolio
Panel A: Equally Weighted Cumulative Abnormal Returns (EWCAR)
All IPOs MM AM	−2.86 −14.16 19.23	−0.69 −3.05** 2.40*	−0.72 −3.72** 2.83*	−1.92* −4.11*** 2.16*	1.70 −0.90 6.90	0.40 −0.19 0.80	0.69 −0.18 0.43	−0.45 −1.09 0.59
Panel B: Value-weighted Cumulative Abnormal Returns (VWCAR)
All IPOs MM AM	−24.20 −20.44 −47.16	−23.05* −18.94* −14.90***	−13.74* −11.65* −8.59***	−6.01* −9.43* −7.90***	−15.10 −10.66 −34.64	−23.49* −18.4* −7.262***	−13.26* −12.06* −7.094***	−10.23* −9.72* −5.32**
Panel C: Main Market Versus ACE Market IPOs
	Para. t-stat	Non-para. z-stat				Para. t-stat	Non-para. z-stat
EWCAR	−3.86***	3.96***				−0.855	0.99
VWCAR	−4.32***	4.02***				−5.33	4.52

Source: Authors’ own empirical findings.

Note: 1. The abnormal returns are calculated up to a 36-month period post-listing, excluding the initial return. Market indices and size/BTM matching portfolio returns are used as benchmarks. 2. The parametric conventional t-statistics (Para. t-stat) and crude dependence test (Para. Adj. t-stat) are the 2-tailed results of the null hypothesis that the means are equal to zero. The non-parametric Wilcoxon Signed Rank test (Non-para. s-stat) is used to test the null hypothesis that the median abnormal return in zero. Panel C indicate the difference in mean and median between MM and AM IPOs, based on the parametric independent t-test and the non-parametric Mann–Whitney U test. ***, ** and * indicate statistical significance at the 1%, 5% and 10%, respectively.

For equally weighted cumulative abnormal return (EWCAR) using market index, we found evidence of significant underperformance (−14.16%) in the MM IPOs. In contrast, the ACE market IPOs over- performance (19.0%) is less significant. For the reference portfolios, we found a marginal underperformance evidence (−0.9%) with MM IPOs and overperformance (6.9%) with AM IPOs.

Turning our attention to the value-weighting scheme (Panel B), we found significant evidence of underperformance (p < 0.001) being associated with an intensified level of underperformance. This implies that larger firms tend to perform worse than smaller firms in the long run. Our findings are in contrast with previous findings in the Malaysian market. With regard to MM IPOs, our findings contradict with those reported by Jelic et al. (2001) and Ahmad-Zaluki et al. (2007) for the Malaysian market. However, the findings for AM IPOs are in line with Ahmad-Zaluki and Kect (2012) that report high underperformance in the MESDAQ market. Based on the initial analysis, we found that the abnormal returns are sensitive to benchmark and weighting scheme employed in calculating stock performances, supporting the majority of the literature.

To determine whether the CAR results of MM and AM IPOs are significantly different, using both 2-sample t-test (parametric) and Mann-Whitney test (non-parametric), our finding as seen in Panel C of Table 4 suggests that the CAR of MM and AM IPOs are significantly different when using the market index, implying that in the long run, the Malaysian IPOs perform lower than all non-IPO stocks listed in Bursa Malaysia taken together. However, we found an insignificant evidence supporting the difference in CARs when using the reference portfolio benchmark, indicating that when compared to only non-IPO firms with similar size and market-to-book value characteristics, the IPOs perform indifferently.

To further examine the IPO performance in the 36-month period, we supplement our results with trend analysis of 3-year CARs. As illustrated in Figure 1, when using equal weighting, we found that AM IPOs outperform both benchmarks throughout the 36-month period. The long-term positive EWCARs are driven by the high proportion of firms that consistently generate monthly abnormal return of more than 100 per cent throughout the 36-month period. However, MM IPOs generate negative abnormal returns beginning in month 19 and 32 for market benchmark and reference portfolio, respectively. For value-weighting estimates, we found that both MM and AM IPOs underperform and are mainly driven by large IPO firms. In particular, the abnormal return of AM IPOs significantly drops when the value-weighting scheme is adopted. The high underperformance is driven by the constant negative monthly abnormal returns generated by the three largest IPOs in the AM that represent 52 per cent of the overall market capitalisation of AM IPOs. Figure 1(c) and (d) exhibits that the return of AM IPOs are more volatile than their MM counterparts, suggesting that when size factor is taken into consideration, smaller firms tend to be more volatile and riskier.

Figure 1.

Source: Graphs built based on authors' own empirical findings.

4.2.2 Buy-and-hold Abnormal Returns

Next is the abnormal return estimates using the buy-and-hold approach. We report the equally weighted buy-and-hold abnormal returns (EWBHAR) and the value-weighted buy-and-hold abnormal returns (VWBHAR) portfolios in Panels A and B of Table 5, respectively. Starting with the equally weighted portfolio, we found insignificant results across all markets and benchmarks, with the exception of the MM, where a significant underperformance is evidenced at the 10 per cent level. In contrast, although insignificant, we note that AM IPOs overperform the market index.

The results are found to be sensitive to the benchmark used. With reference portfolios, EWBHAR in MM IPOs turn to be overperforming (2.53%), while AM IPOs turn to be underperforming (12.9%). For the VWBHAR reported in Panel B of Table 5, there are clear evidence supporting underperformance across all markets and benchmarks at a 1 per cent level of significance. Since BHAR reflects the actual investor experience, the findings suggest that IPOs are not a promising investment option in the long run.

Table 5.
Three-year Buy-and-hold Abnormal Returns

Market Index
Reference Portfolio

BHAR (%) Para. t-stat Non- para. Adj t-stat Non- para. s-stat Wealth Relative
(WR) BHAR (%) Para. t-stat Non- para. Adj t-stat Non- para. s-stat Wealth Relative
(WR)

Panel A: Equally Weighted Buy-and-hold Abnormal Returns (EWBHAR)

All IPOs
MM
AM −6.55
−12.85
5.77 −1.23
−1.93*
0.48 −1.67
−1.87
0.36 −7.25*
−6.61
−3.28 0.94
0.9
1.06 −2.69
2.53
−12.90 −0.03
0.38
0.80 0.02
−0.42
0.93 −5.21*
−3.28
−4.21*** 0.98
1.02
0.89

Panel B: Value-weighted Buy-and-hold Abnormal Returns (VWBHAR)

All IPOs
MM
AM −35.77
−35.31
−37.74 −10.27*
−9.02*
−4.83* −12.73*
−10.57*
−5.43* −5.01*
−8.73*
−9.80* 0.74
0.71
0.76 −31.27
−21.80
−71.16 −8.17*
−5.17*
−8.97* −918*
−6.06*
7.09* −10.23*
−9.72*
−5.32 1.1
0.81
0.64

Panel C: Main Market Versus ACE Market IPOs

t-stat z-stat t-stat z-stat

EWBHAR
VWBHAR 1.57
1.43 2.23*
2.38* −0.855
−5.33 0.99
4.52

	Market Index	Reference Portfolio
Panel A: Equally Weighted Buy-and-hold Abnormal Returns (EWBHAR)
All IPOs MM AM	−6.55 −12.85 5.77	−1.23 −1.93* 0.48	−1.67 −1.87 0.36	−7.25* −6.61 −3.28	0.94 0.9 1.06	−2.69 2.53 −12.90	−0.03 0.38 0.80	0.02 −0.42 0.93	−5.21* −3.28 −4.21***	0.98 1.02 0.89
Panel B: Value-weighted Buy-and-hold Abnormal Returns (VWBHAR)
All IPOs MM AM	−35.77 −35.31 −37.74	−10.27* −9.02* −4.83***	−12.73* −10.57* −5.43***	−5.01* −8.73* −9.80***	0.74 0.71 0.76	−31.27 −21.80 −71.16	−8.17* −5.17* −8.97***	−918* −6.06* 7.09***	−10.23* −9.72* −5.32**	1.1 0.81 0.64
Panel C: Main Market Versus ACE Market IPOs
	t-stat	z-stat				t-stat	z-stat
EWBHAR VWBHAR	1.57 1.43	2.23* 2.38*				−0.855 −5.33	0.99 4.52

Source: Authors’ own empirical findings.

Note: 1. The parametric conventional t-statistics (Para. t-stat) and the non-parametric skewness-adjusted bootstrapped t-statistic (Non-para. Adj. t-stat.) are the 2-tailed results of the null hypothesis that the means are equal to zero. The non-parametric Wilcoxon Signed Rank test (Non-para. s-stat) is used to test the null hypothesis that the median abnormal return in zero. Panel C indicates the difference in mean and median between MM and AM IPOs, based on the parametric independent t-test and the non-parametric Mann-Whitney U test. ***, ** and * indicate statistical significance at the 1%, 5% and 10%, respectively. 2. Refer to Table 3 for explanation on tests of differences.

When the VWBHAR of the two markets are compared, we found that the MM IPOs perform better than AM IPOs. These results are in line with Gregory et al. (2010) and Vismara et al. (2012) but contradict with Komenkul et al. (2012). In agreement with the findings reported for CAR, it is worth noting that the negative BHARs are magnified in the value-weighted portfolios. Overall, our EWBHAR results are in line with those reported by Paudyal et al. (1998) and Ahmad-Zaluki and Kect (2012) but in contrast with those found by Ahmad-Zaluki et al. (2007). However, and with regard to VWBHAR, our findings are supportive to those reported by Ahmad-Zaluki et al. (2007). Importantly and in line with our previous findings, the medians of MM and AM IPOs are significantly different as indicated by the non-parametric Mann-Whitney test.

As a further check to the movement of BHARs in the 3-year after- market, the results are illustrated graphically in Figure 2. For the entire sample, our findings imply that investors’ over-optimism exist on the future prospect of the Malaysian IPOs. The optimistic investors initially overvalue the IPOs and then, as more information are available, they begin to revalue the IPOs pessimistically. Thus, the IPO returns are driven downwards, which leads to the negative long-run abnormal returns.

Figure 2.

Source: Graphs built based on authors’ own empirical findings.

For the value weighting, interestingly, we found that AM IPOs substantially underperform the reference portfolio by −71.16 per cent. The results are dominated by the largest IPO firms in the AM market that unreasonably underperform (−51.11%) in 36th month. The results are clearly a size-effect pattern and not an outlier. For the IPO firms, the raw return in the month is only −3.2 per cent, proportionally smaller than the worst performance of −42.9 per cent in the same month. However, due to the fact that these firms represent 27 per cent of the total market capitalisation in AM portfolio, the BHAR is amplified.

Table 6.

Three-year Abnormal Returns Based on Sector Groupings

Panel A: Equally Weighted Portfolios
	Cumulative Abnormal Returns				Buy-and-hold Abnormal Returns
Industry	CAR (%)	Para. t-stat	Para. Adj. t-stat	Non-para. s-stat	BHAR (%)	Para. t-stat	Para. Adj. t-stat	Non-para. s-stat	Wealth Relative (WR)
Construction	19.63	0.70	0.94	0.85	35.67	0.92	0.05	0.68	1.31
Consumer Products	−46.61	−6.89***	−6.72**	−4.75***	−52.95	−1.70*	−1.55	−6.10***	0.96
Industrial	−9.44	−1.27	−1.07	−1.51	−1.06	−0.33	−0.53	−3.42***	0.98
Heavy Machinery	−23.79	−1.32	−1.65	−0.94	−8.49	−1.04	−1.43	−3.35***	0.92
Plantation	2.18	0.07	0.16	0.77	−54.26	0.12	0.02	−0.89	1.04
Property	16.52	0.60	0.22	0.16	0.43	−0.09	−0.57	−0.31	0.98
Technology	50.45	5.27***	4.37**	0.92	18.57	0.03	0.94	−4.46***	1
Trading and Services	21.93	2.19*	2.22*	1.45	20.13	0.32	0.33	−0.05	1.02
Panel B: Value-weighted Portfolios
Industry	Cumulative Abnormal Returns				Buy-and-hold Abnormal Returns
Industry	CAR (%)	Para. t-stat	Para. Adj. t-stat	Non-para. s-stat	BHAR (%)	Para. t-stat	Para. Adj. t-stat	Non-para. s-stat	Wealth Relative (WR)
Construction	35.67	2.55*	0.36	0.49	26.55	2.61*	2.71*	0.48	1.23
Consumer products	−4.00	−13.58***	17.31***	15.51***	−59.95	−0.17	−0.17	−7.70	0.99
Industrial	−21.40	−11.20***	−4.57***	9.24***	−26.78	−5.28***	−6.35***	−4.32***	0.99
Heavy machinery	14.30	2.70*	2.56*	2.84*	−42.60	−6.15***	−6.89	−3.32*	0.89
Plantation	2.33	0.77	0.39	0.41	−0.69	0.45	0.61	0.79	1.43
Property	12.86	3.33**	1.98*	1.45*	19.70	–1.41	−1.47	−0.37	0.82
Technology	50.46	10.52***	5.56***	6.52***	5.04	−2.82**	−3.01**	−5.54**	0.99
Trading and services	−12.15	−8.36***	−6.25***	−7.35***	−20.02	2.69**	2.75**	4.35**	1.03

Source: Authors’ own empirical findings.

Note: Refer to Table 3 for explanation on calculations of abnormal returns and Tables 3 and 4 for explanation on calculations of tests of differences for CAR and BHAR, respectively.

4.3 Industry Effect on IPO Long-run Abnormal Returns

We next provide more robustness analysis by investigating the industry effect. We divide the samples into eight sector groupings based on the Bursa Malaysia sector classification. The results of industry-specific CAR and BHAR are shown in Table 6 and in Figure 3. Turning our attention first to the 3-year equally weighted portfolio, it is evident that the maximum positive abnormal return is achieved by the IPOs in technology sector, the EWCAR is 50.45 per cent (t-stat = 5.27). IPO firms classified in the trading and services sector produce a 3-year EWCAR of 21.93 per cent (t-stat = 2.19).

Figure 3.

Source: Graphs built based on authors' own empirical findings.

These key findings indicate that investors may gain significant cumulative abnormal return if they buy technology IPOs on the first trading day at the closing price, and hold them for a 36-month period. This justifies why the Malaysian government has taken initiatives to stimulate the growth of the technology sector. For example, in 1996, the Multimedia Super Corridor was established to attract local and foreign entrepreneurs to set up technology companies in the country. In the regional IPO market, the Malaysian technology IPOs play a competitive role as such IPO firms tend to be the largest in terms of market capitalisation.

Our EWBHAR findings further justify that the abnormal returns are sensitive to the methods employed. For example, for the technology sector, while EWCAR is upward trending, EWBHAR shows a downward trend. However, the choice of weighting method does not seem to alter the results in this case, where technology IPOs show overperformance when both equal- and value-weightings are adopted. Likewise, Figure 3 clearly exhibits that Value-weighted Cumulative Abnormal Returns (VWCAR) of IPOs in the technology sector increase between months 27 and 36, although we can also see some volatility. However, the volatility in abnormal return mirrors the price volatility of the largest IPO in the technology portfolio. The consumer products (−13.58%, t-stat = 13.58), industrial (−21.40%, t-stat = −11.20) and trading and services (−12.15%, t-stat = −8.36) sectors exhibit a significant underperformance evidence in the long run. The consumer products and industrial sectors consistently exhibit underperformance regardless of the methods employed. In particular, for the consumer products sector, the underperformance is statistically significant with EWCAR, EWBHAR and VWCAR methods. For the industrial sector, the results are significant only with VWCAR and VWBHAR. The industry analysis supports our early observations that long-run abnormal returns of IPOs are sensitive to the methodology employed to calculate stocks performance. Our initial findings suggest that the underperformance phenomena exist in both the main and ACE markets. Strikingly, based on the industry analysis, the underperformance is not widespread to all industries, but only in some sectors. Again, the results persistently show that the abnormal returns are method-sensitive. Overall, our industry-based analysis shed the light that long-run abnormal performance return is not only related to listing market but also partially driven by sector groupings. The sector impact is more apparent when a higher number of stocks are associated with a particular sector.

Table 7.

Calendar-time Abnormal Returns Using Equally Weighted Portfolio

	α	β_Rm-Rf	β_SMB	β_HML	β_MOM	β_LIQ	Adj. R²
Panel A: All IPOs
FF3F model	−0.004	0.565***	0.057	−0.065
FF3F model	(−0.008)	(−0.104)	(−0.213)	(−0.228)
MOM model	−0.003	0.575***	0.096	−0.063	−0.068		0.225
MOM model	(−0.008)	(−0.105)	(−0.211)	(−0.226)	(−0.099)
LCAPM Model	−0.002**	0.150*				−0.001**
LCAPM Model	(−0.009)	(−0.126)				(−0.002)	0.222
Panel B: Main Market IPOs
FF3F model	−0.005	0.587***	0.055	−0.046			0.239
FF3F model	(−0.008)	(−0.101)	(−0.213)	(−0.231)
MOM model	−0.005	0.595***	0.090	−0.045	−0.061		0.240
MOM model	(−0.008)	(−0.103)	(−0.213)	(−0.230)	(−0.107)
LCAPM model	−0.002**	0.158				−0.001**	0.301
LCAPM model	−0.007	−0.124				(−0.002)
Panel C: ACE Market IPOs
FF3F model	0.000	0.401**	0.058	−0.013			0.084
FF3F model	(−0.009)	(−0.165)	(−0.265)	(−0.279)
MOM model	0.000	0.406**	0.090	−0.012	−0.056		0.085
MOM model	(−0.009)	(−0.166)	(−0.265)	(−0.279)	(−0.070)
LCAPM model	0.002*	0.204				0.001**	0.085
LCAPM model	(−0.008)	(−0.168)				−0.002

Source: Authors’ own empirical findings.

Notes: This table presents the regression results of calendar-time market-adjusted monthly abnormal returns using the following models:

1. FF3F model: R_it − R_ft = α_i + β_i(R_m − R_ft) + β_SMBSMBt + β_HMHM_Lt + ε_it where R_it is the return on IPO stock i in month t, R_mt is the return on the market in event month t calculated using the FTSE Bursa Malaysia Kuala Lumpur Composite Index (FBM KLCI), R_ft is the 1-month Malaysian Government Security return in month t, SMB is the value-weighted return on a portfolio of small-minus-big firms, and HML is the value-weighted return a portfolio of high book-to-market (BTM) firms minus the value-weighted return on firms with low BTM; 2. Carhart (1997) 4-factor model, with momentum (MOM) factor added to the FF3F equation, representing the difference between returns of high-and-low momentum securities in the previous year; and (3) the capital asset pricing model (CAPM) with controlling liquidity factor (LCAPM). The t-statistics (shown in brackets) are calculated using the time-series standard deviation of the mean monthly abnormal returns. The statistical significance is generated after White heteroskedasticity adjustments. ***, ** and * indicate statistical significance at the 1, 5 and 10 per cent, respectively.

Table 8.

Calendar-time Abnormal Returns Using Value-weighted Portfolio

	α	β_Rm _{− Rf}	β_SMB	β_HML	β_MOM	β_LIQ	Adj. R²
Panel A: All IPOs
FF3F model	(0.000)	0.49200***	0.001	0.001			0.135
	(−0.00005)	(−0.13300)	(−0.00119)	(−0.00154)
MOM model	(0.000)	0.49900***	0.001	0.001	(0.000)		0.137
	(−0.00005)	(−0.13700)	(−0.00120)	(−0.00153)	(−0.00069)
LCAPM model	(0.013)	0.043				0.001
	(−0.01000)	(0.199)				(−0.01050)
Panel B: Main Market IPOs
FF3F model	(0.000)	0.53500***	0.001	0.001			0.145
	(−0.00005)	(−0.14800)	(−0.00127)	(−0.00163)
MOM model	(0.000)	0.540***	0.001	0.001	(0.000)		0.146
	(−0.00005)	(−0.15400)	(−0.00130)	(−0.00162)	(−0.00084)
LCAPM model	(0.016)	0.077				0.001	0.164
	(−0.00780)	(0.137)				(−0.00170)
Panel C: ACE Market IPOs
FF3F model	−0.003***	0.503***	0.068*	(0.030)			0.592
	(−0.001)	(−0.055)	(−0.037)	(−0.033)
MOM model	−0.003***	0.502***	0.067*	(0.030)	0.003
	(−0.001)	(−0.056)	(−0.039)	(−0.033)	(−0.018)		0.592
LCAPM model	(0.001)	0.154				−0.000**
	(−0.358)	(−0.606)				(−0.949)

Source: Authors’ own empirical findings.

Note: See Table 6 for detailed description.

4.4 Long-run Abnormal Return Using Calendar-time Approach

With event-time approach, long-run abnormal returns may suffer from cross-sectional relations of IPOs (Brown & Warner, 1980). To mitigate the issue, we employ the calendar-time portfolio approach as robustness test, using the FF3F model (Fama & French, 1993), the MOM (1997), and the LCAPM (Amihud, 2002).

The dependent variable, R_i−R_m, is the monthly excess returns on IPO portfolios between January 2000 and December 2011. For the entire sample and the MM IPOs, the sample size covers a total of 143 months from March 2003 to December 2014.⁶ Thus, the sample size for the market covers 116 months, beginning from May 2005. The maximum number of firms in the IPO trading in the aggregate sample is 200, occurring in May 2009. The minimum number of observation is two, which occurred in February and March 2003. For the main (ACE) market samples, the maximum number of firms in the IPO portfolio is 127 (105), occurring in March 2008 (November 2009). The minimum is 2 (3) which occurred in February and March 2003 (May and July 2005).

The value-weighted and equally weighted excess returns of the IPOs returns are regressed on risk premium (R_m − R_f), small-minus-big (SMB), high-minus-low (HML), momentum (MOM) and liquidity (LIQ) variables using the ordinary least squares (OLS) regression. We first perform the regression on the entire sample and then repeat the procedure for the MM and ACE market IPOs, separately. It is important to note that the intercept α from the regressions indicates the monthly risk-adjusted performance of the IPOs.

Starting with the equally weighted portfolio using the three-asset pricing models, our findings indicate that underperformance exists in MM IPOs as presented in Panel A of Table 7. Notably, the abnormal return is only significant (p < 0.01) with LCAPM but not the case with the FF3F and the MOM models. The negative intercept of −0.002 in the LCAPM model indicates that MM IPOs underperform at −0.2 per cent monthly, which turns to a −6.95 per cent 3-year abnormal return.⁷ Our results are in line with previous studies (e.g. Ahmad-Zaluki et al., 2007; Moshirian et al., 2010; Su & Bangassa, 2011; Agathee et al., 2014) for AM IPOs. Moreover, as seen in Panel B of Table 7, all our models exhibit positive intercept, indicating overperformance in the IPOs. In agreement with the findings associated with MM IPOs, the LCAPM model shows weak significance at 10 per cent level. Based on the LCAPM model, AM IPOs overperform at 0.2 per cent monthly, or 7.45 per cent in 3-year-time horizons.

As a final robustness test, we repeat the estimates and regressions using the value-weighted IPO portfolio returns (Table 8). The results for MM IPOs consistently show underperformance, as can be seen in Panel B of Table 8. Subsequently, for AM IPOs, we also report negative intercepts for all models. The results, as shown in Panel C of Table 8, suggest that AM IPOs also underperform in the long run.

Our calendar-time regressions provide stronger support to our event-time results that MM IPOs significantly underperform in the long run. Due to the larger proportion of MM IPOs to AM IPOs in terms of size and number of IPOs, the abnormal return in the whole sample is heavily influenced by the behaviour of MM IPOs. Hence, in aggregate, the Malaysian IPOs underperform in the period of 36-month aftermarket.

To conclude, our analysis using calendar-time show strong evidence of underperformance in MM and AM IPOs and also at the overall economy level. The underperformance is higher in AM, which is in line with Ritter’s (1991) hypothesis that higher underpricing will lead to higher underperformance. The statistical significance is particularly stronger ( p < 0.001) with value-weighted portfolio, indicating that size effect plays a significant role in driving the abnormal returns. As the underperformance tends to be higher in value-weighted portfolio, this implies that larger IPOs tend to perform worse in the long run. Overall, the results provide answers to our research questions on whether underperformance exists in the Malaysian market and whether the main and ACE markets portray different long-run abnormal performance.

5. Conclusion

Most prior literature claims that stocks of firms that go public under- perform their peers over a period of 2–5 years following their issue date. This long-term abnormal return anomaly is a debatable and puzzled issue that contradicts all forms of the efficient market hypothesis. The main purpose of this article is to investigate the long-term stock performance of 448 Malaysian IPOs listed between 2000 and 2011 in the main and ACE alternative markets. We investigate the 3-year stock returns of the IPOs using CAR, BHAR and WR. We employ related returns for the market indices, size/BTM matched reference portfolio and eight industry indices as benchmarks and utilise more robust statistical tests. We further adopt the calendar-time approach to test the existence of long-run abnormal returns using the market model with additional liquidity factor, as well as the Fama–French and Carhart models.

Based on our analysis, we draw four important conclusions. First, in line with the majority of the literature, the anomaly of long-run underperformance is found to be exist in the Malaysian market. In addition, we find distinguishable long-run performance between IPOs during the pre-and post-2000 period. Unlike the IPOs listed in the post-2000 period, we find Malaysian IPOs generally overperform in the period of 36-month aftermarket, indicating that publicly available information is reflected in the IPOs performance and the market is relatively efficient. Second, there is evidence supportive of the differences in the long-run performance of the main and ACE market IPOs. Further, we find higher tendency of ACE market IPOs to underperform, and their investors are more optimistic about the future prospect of the IPOs compared to the main market investors. However, given the characteristics of the main market that generally lists more mature and financially stable firms, the overall IPOs performance show a lower level of risk and hence, reflect a lower rate of expected returns by investors. This supports the hypothesis that IPOs with better initial returns are the worst long-run performers. Furthermore, although underperformance exists in the Malaysian market as a whole, the anomaly is not widespread across all sector groupings. We find the consumer products and industrial sectors to underperform in the long run, while the construction, property and technology sectors overperform. There is also significant variation in the level of abnormal returns across sector groupings. Finally, our findings are robust to a wide range of sensitivity checks including parametric and non-parametric tests.

Overall, our article illustrates the long-term stock performance of Malaysian IPOs in the main and ACE alternative stock markets at economy and sector levels. Therefore, our study offers insights to policy- makers interested in pricing the future and new security offerings. For issuing firms, our results provide them with rooms for price adjustments by reducing information asymmetry in general and also between investors. Our findings are also helpful for regulators overseeing other emerging markets beyond Malaysia.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Footnotes

Notes

References

Agathee

, Sannassee

, & Brooks

(2014). The long-run performance of IPOs: The case of the stock exchange of Mauritius. Applied Financial Economics, 24(17), 1123–1145.

Ahmad-Zaluki

, & Kect

(2012). The investment performance of MESDAQ market initial public offerings (IPOs). Asian Academy of Management Journal of Accounting and Finance, 8(1), 1–23.

Ahmad-Zaluki

, Campbell

, & Goodacre

(2007). The long run share price performance of Malaysian initial public offerings (IPOs). Journal of Business Finance and Accounting, 34(1–2), 78–110.

Álvarez

, & González

(2005). Signalling and the long-run performance of Spanish initial public offerings (IPOs). Journal of Business Finance and Accounting, 32(1–2), 325–350.

Amihud

(2002). Illiquidity and stock returns: Cross-section and time-series effects. Journal of Financial Markets, 5(1), 31–56.

Barber

, & Lyon

(1997). Detecting long-run abnormal stock returns: The empirical power and specification of test statistics. Journal of Financial Economics, 43(3), 341–372.

Bessler

, & Thies

(2006). Initial public offerings, subsequent seasoned equity offerings, and long-run performance: Evidence from IPOs in Germany. Journal of Entrepreneurial Finance, 11(3), 1–37.

Brau

, Couch

, & Sutton

(2012). The desire to acquire and IPO long-run underperformance. Journal of Financial and Quantitative Analysis, 47(3), 493–510.

Brown

, & Warner

(1980). Measuring security price performance. Journal of Financial Economics, 8(3), 205–258.

10.

Carhart

(1997). On persistence in mutual fund performance. The Journal of Finance, 52(1), 57–82.

11.

Chen

, & Sherif

(2016) Illiquidity premium and expected stock returns in the UK: A new approach. Physica A: Statistical Mechanics and Its Applications, 458(1), 52–66.

12.

Chorruk

, & Worthington

(2010). New evidence on the pricing and performance of initial public offerings in Thailand, 1997–2008. Emerging Markets Review, 11(3), 285–299.

13.

Corhay

, Teo

, & Tourani Rad

(2002). The long run performance of Malaysian initial public offerings (IPO): Value and growth effects. Managerial Finance, 28(2), 52–65.

14.

Da Silva Rosa

, Velayuthen

, & Walter

(2003). The sharemarket performance of Australian venture capital-backed and non-venture capital-backed IPOs. Pacific-Basin Finance Journal, 11(2), 197–218.

15.

Dawson

(1987). Secondary stock market performance of initial public offers, Hong Kong, Singapore and Malaysia: 1978–1984. Journal of Business Finance and Accounting, 14(1), 65–76.

16.

Drew

, & Veeraraghavan

(2002). A closer look at the size and value premium in emerging markets: Evidence from the Kuala Lumpur Stock Exchange. Asian Economic Journal, 16(4), 337–351.

17.

Drobetz

, Kammermann

, & Walchli

(2005). Long-term performance of initial public offerings: The evidence for Switzerland. Schmalenbach Business Review, 57(3), 253–275.

18.

Eckbo

, & Norli

Ø.

(2005). Liquidity risk, leverage and long-run IPO returns. Journal of Corporate Finance, 11(1–2), 1–35.

19.

Emasari

, & Tamara

(2010). Investigation towards the long-run performance of initial public offerings: Evidence from Indonesian Capital Market. Journal of Applied Finance and Accounting, 2(2), 100–125.

20.

Erdogan

(2010). The long-run performance of initial public offerings: The case of Turkey. European Journal of Economics, Finance and Administrative Sciences, 26, 57–64.

21.

Espenlaub

, Gregory

, & Tonks

(2000). Re-assessing the long-term underperformance of UK initial public offerings. European Financial Management, 6(3), 319–342.

22.

Fama

(1998). Market efficiency, long-term returns, and behavioral finance. Journal of Financial Economics, 49(3), 283–306.

23.

Fama

, & French

(1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1), 3–56.

24.

Fama

, & French

(1996). Multifactor explanations of asset pricing anomalies. The Journal of Finance, 51(1), 55–84.

25.

Gao

, & Jain

(2011). Founder CEO management and the long-run investment performance of IPO firms. Journal of Banking and Finance, 35(7), 1669–1682.

26.

Gompers

, & Lerner

(2003). The really long-run performance of initial public offerings: The pre-NASDAQ evidence. The Journal of Finance, 58(4), 1355–1392.

27.

Gregory

, Guermat

, & Al-Shawawreh

(2010). UK IPOs: Long run returns, behavioural timing and pseudo timing. Journal of Business Finance and Accounting, 37(5–6), 612–647.

28.

Jelic

, Saadouni

, & Briston

(2001). Performance of Malaysian IPOs: Underwriters reputation and management earnings forecasts. Pacific-Basin Finance Journal, 9(5), 457–486.

29.

Johnson

(1978). Modified t tests and confidence intervals for asymmetrical populations. Journal of the American Statistical Association, 73(363), 536–544.

30.

Khurshed

, Mudambi

, & Goergen

(1999). On the long-run performance of IPOs. Reading: ISMA Centre, University of Reading.

31.

Kirkulak

(2008). The initial and long-run returns of Japanese venture capital-backed and non-venture capital-backed IPOs. International Journal of Managerial Finance, 4(2), 112–135.

32.

Komenkul

, Brzeszczynski

, & Sherif

(2012). Long-run performance of initial public offers (IPOs): Evidence from Thailand (Heriot-Watt University, Working Paper). Edinburgh: Heriot-Watt University.

33.

Komenkul

, Sherif

, & Xu

(2016). IPOs’ signalling effects for speculative stock detection: Evidence from the stock exchange of Thailand. Applied Economics, 49(31), 3067–3085. doi: 10.1080/00036846.2016.1254338

34.

Kooli

, & Suret

J.-M.

(2004). The aftermarket performance of initial public offerings in Canada. Journal of Multinational Financial Management, 14(1), 47–66.

35.

Kothari

, & Warner

(1997). Measuring long-horizon security price performance. Journal of Financial Economics, 43(3), 301–339.

36.

Kothari

, & Warner

(2004). The econometrics of event studies. Available at SSRN 608601.

37.

Kothari

, & Warner

(2006). Econometrics of event studies. In Handbook of empirical corporate finance (Vol. 1, pp. 3–36).

38.

Levis

(1993). The long-run performance of initial public offerings: The UK experience 1980–1988. Financial Management, 22(1), 28–41.

39.

Locke

, & Gupta

(2009). The return to initial public offerings: A Sino-Indian comparison. Venture Capital, 11(3), 255–277.

40.

Loughran

, & Ritter

(1995). The new issues puzzle. The Journal of Finance, 50(1), 23–51.

41.

Loughran

, Ritter

, & Rydqvist

(1994). Initial public offerings: International insights. Pacific-Basin Finance Journal, 2(2), 165–199.

42.

Lyon

, Barber

, & Tsai

(1999). Improved methods for tests of long-run abnormal stock returns. The Journal of Finance, 54(1), 165–201.

43.

Mazouz

, Saadouni

, & Yin

(2008). The long-term performance of Hong Kong share-only and unit initial public offerings (IPOs). Journal of Multinational Financial Management, 18(3), 209–228.

44.

Miller

, & Reilly

(1987). An examination of mispricing, returns, and uncertainty for initial public offerings. Financial Management, 16(2), 33–38.

45.

Mitchell

, & Stafford

(2000). Managerial decisions and long-term stock price performance. The Journal of Business, 73(3), 287–329.

46.

Moshirian

, Ng

, & Wu

(2010). Model specification and IPO performance: New insights from Asia. Research in International Business and Finance, 24(1), 62–74.

47.

Paudyal

, Saadouni

, & Briston

(1998). Privatisation initial public offerings in Malaysia: Initial premium and long-term performance. Pacific-Basin Finance Journal, 6(5), 427–451.

48.

Ritter

J. R.

(1991). The long-run performance of initial public offerings. The Journal of Finance, 46(1), 3–27.

49.

Saade

(2015). Investor sentiment and the underperformance of techno-logy firm’s initial public offerings. Research in International Business and Finance, 34(2), 205–232.

50.

Schwab

(Ed.). (2013). The global competitiveness report 2013–2014. World Economic Forum. Retrieved from http://www3.weforum.org/docs/WEF_GlobalCompetitivenessReport_2013-14.pdf

51.

Securities Commission Malaysia (2011). Capital market masterplan 2. Retrieved from https://www.sc.com.my/wp-content/uploads/eng/html/cmp2/cmp2_final.pdf

52.

Sherif

, & Erkol

(2017). Sukuk and conventional bonds: Shareholder wealth perspective. Journal of Islamic Accounting and Business Research, 8(4), 347–374. Retrieved from https://doi.org/10.1108/JIABR-09-2016-0105

53.

Sherif

, Komenkul

, & Xu

(2016). Prospectus disclosure and the stock market performance of initial public offerings (IPOs): The case of Thailand. Investment Management and Financial Innovations, 13(4), 146–159. doi: 10.21511/imfi.13(4-1).2016.02

54.

, & Bangassa

(2011). The impact of underwriter reputation on initial returns and long-run performance of Chinese IPOs. Journal of International Financial Markets, Institutions and Money, 21(5), 760–791.

55.

The World Bank (2015). Gross domestic product growth rate. Retrieved from https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=MY-SG-TH

56.

Thomadakis

, Nounis

, & Gounopoulos

(2012). Long-term performance of Greek IPOs. European Financial Management, 18(1), 117–141.

57.

Vismara

, Paleari

, & Ritter

(2012). Europe’s second markets for small companies. European Financial Management, 18(3), 352–388.

58.

Yong

(1991). Performance of new issues of securities in Malaysia. The Malaysian Accountant, 3–6.