The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/1743-9132.htm

The effect of underwriter’s Underpricing

of Malaysian
market share, spread and IPOs
management earnings forecasts
bias and accuracy on 351
underpricing of Malaysian Received 1 December 2014

IPOs
Revised 23 February 2015
Accepted 31 March 2015

Mohammed Abdullah Ammer

Faculty of Commerce and Economics, Amran University, Amran, Yemen, and
Nurwati A. Ahmad-Zaluki
Finance Department, Universiti Utara Malaysia, Kedah, Malaysia

Abstract
Purpose – Underpricing is one of the most important anomalies associated with initial public
offerings (IPOs). The purpose of this paper is twofold; first, it examines the impact of underwriter ’s
market share and spread on the underpricing of IPOs; and second, it investigates the effect of
management earnings forecasts bias and accuracy on the underpricing of IPOs.
Design/methodology/approach – A sample of 190 Malaysian IPOs listing on the main market of
Bursa Malaysia from January 1, 2002 to February 29, 2012 was used and collected data were
analyzed through univariate analysis and pooled ordinary least squares regression.
Findings – The empirical evidence shows that IPOs underwritten by underwriters as having high
market share and charging low underwriting spread experience higher level of underpricing.
The paper also finds that IPOs issued more biased earnings forecasts are related with severe
underpricing. Finally, this paper reveals that the more accurate the earnings forecasts are, the more
minimized will be the asymmetric information and hence, the less will be IPO underpricing.
Practical implications – The paper has some implications for policy makers, investors and
underwriters. First, this study offers some insights for policy makers who are responsible for
Malaysian financial markets current reforms. Further, knowing the importance of the economic
outcomes of the earnings forecasts on underpricing for policy makers, they may adopt the findings in
their discussion of costs of litigation and potential modifications in the requirements of issuing
earnings forecasts. For the investors, findings may improve their understanding of equity valuation
and for the underwriters, it would assist them in identifying underwriting cost.
Originality/value – This paper is considered the first study to extend IPO literature by investigating
the relationships between underwriter’s market share, underwriter’s spread, earnings forecasts bias,
earnings forecasts accuracy and IPO underpricing in an emerging country, such as Malaysia.
Keywords Malaysia, IPOs, Earnings forecasts bias and accuracy,
Underwriter’s market share and spread
Paper type Research paper

1. Introduction
The puzzle of initial public offerings (IPO) underpricing is an interesting research
field that has caught academic scholars’ and researchers’ attention alike for the
International Journal of Managerial
past 40 years[1]. In fact, the IPO underpricing seems to have reached a position of Finance
universal phenomenon (Loughran et al., 1994; Ritter, 1984; Ritter and Welch, 2002; Vol. 12 No. 3, 2016
pp. 351-371
Chambers and Dimson, 2009; Boulton et al., 2011). A widespread acknowledged © Emerald Group Publishing Limited
1743-9132
notion is that IPO underpricing goes against market efficiency and may adversely DOI 10.1108/IJMF-12-2014-0187

impact companies attempting to raise capital in order to expand. This perception led
to a major portion of
IJMF literature to address this financial anomaly. For instance, Jenkinson and Ljungqvist
12,3 (2001) and Ritter and Welch (2002) reviewed several theories (i.e. asymmetric
information, winner’s curse and signaling) and reached to the conclusion that no
single theory is capable of clarifying underpricing. It is noteworthy that the studies
dedicated to South East Asia provided inconclusive findings. In the context of
Malaysia, IPO studies reported the existence of underpricing in the Malaysian IPO
352 market (e.g. Dawson, 1987; Yong and Isa, 2003; Prasad et al., 2006; Uddin, 2008;
Murugesu and Santhapparaj, 2009; Ahmad-Zaluki and Kect, 2012).
This paper is interested in shedding light on the association between underpricing
and underwriter’s market share and spread in addition to the bias and accuracy of
management earnings forecast[2]. Compounding the motivation is the fact that the
potential relationship between these variables and IPO underpricing has largely been
untouched. Hence, the present paper explains IPO underpricing wherein the issuers
are linked to superior quality underwriters along with the stress on the quality of
disclosed earnings forecasts within the IPO prospectuses. It is expected that offerings
that are of greater quality are underwritten by higher quality investment banks, and
that the latter possesses greater market share in light of IPO proceeds garnered
(Megginson and Weiss, 1991). Nevertheless, this issue is still open for debate (Huang,
2012). Moreover, underwriter spread reflects the risk bearing function of the
underwriter and interacts with IPO underpricing. It may therefore be contended that
underwriter spread should play a key role in clarifying the puzzle of IPO pricing if the
spread variable represents uncertainty concerning new issues pricing. Earnings
forecasts are among the major information that should be included in the IPO
prospectus. Past studies reveal that the accounting information, like earnings and book
value of equity hold the key to IPO pricing (Keasey and McGuinness, 1991; Klein,
1996; Beatty et al., 2002). In other words, forecast studies can deepen our
understanding on the IPO underpricing phenomenon.
The present study provides additional evidence regarding financial performance of
Malaysian IPO market. In particular, it contributes to existing literature in three ways;
first, it is the pioneering study to directly examine underwriting market share and IPO
underpricing in the context of Malaysian IPO. The Malaysian underwriting industry is
known for its high concentration, with the top underwriters as AmInvestment Bank
Berhad, CIMB Investment Bank Berhad, RHB Investment Bank Berhad and MIMB
Investment Bank Berhad. These underwriters constituted the most IPOs in Malaysia
in the years from 2002 to 2012 (in terms of number as well as value). It is worth
noting that Malaysia possesses an oligopolistic underwriting market that is driven by
these four main actors. Second, although underwriter spread constitutes the top
portion of the direct cost of new securities issuance, prior research dedicated to
underwriter spread in the context of Malaysia has not yet been performed. Hence, in
the Malaysian IPO, the effect of underwriter’s market share and spread on IPO
underpricing remains an open empirical question and our study seeks to fill these
gaps.
Third, the earnings forecasts role in signaling the market value of IPOs and in
clarifying initial returns (IR) of stock has not been fully explored in the literature of
accounting and finance (e.g. Clarkson et al., 1992; How, 1996; Firth, 1998; Jog and
McConomy, 2003). The present study fills such a void left by prior studies. In this
study, the relative underpricing throughout the sample companies is tested to provide
evidence of the proposition that earnings forecasts decrease asymmetric information
and ex ante uncertainty (Rock, 1986; Beatty and Ritter, 1986). Furthermore,
according to Ljungqvist (2007), information has first-order effects on underpricing
and hence, it is logical to wonder whether or not the underpricing variation may be
encouraged by
the quality information differences that are provided to investors. The present study’s
contribution to literature lies in its supply of evidence on the economic consequences Underpricing
of the incorporation of management earnings forecasts within the IPO prospectus. of Malaysian
The findings show that on average, the underpricing level in the Malaysian IPO IPOs
market is 21.22 percent. Further, this study shows that underwriters’ market share is
positively but insignificantly related to IPO underpricing. With regards to
underwriter’s spread, although it is one of the methods for pricing the IPO company’s
risk, the spread is found to be negatively but insignificantly linked with underpricing. 353
The association between earnings forecasts error and IPO underpricing is also noted.
The study found that IPO companies having biased earnings forecasts are more
underpriced compared to those with unbiased and cautious earnings forecasts. Finally,
our result indicates that the accuracy of management earnings forecasts do decrease
the IPO underpricing.
The rest of the study is organized as follows; Section 2 provides the prior literature
and hypotheses concerning the IPO underpricing determinants. Section 3 discusses
the data selection process, data sources, empirical methods and Sections 4 and 5
provide the results and additional analysis. Finally, the paper is summarized in
Section 6.

2. Literature review and hypothesis development

2.1 Underwriter’s market share
Activities related to corporate finance, such as securities issuance, offer great revenues
for underwriting banks. Thus, underwriters are in great competition for new businesses
of underwriting. This behavior is specifically true in the IPO market as underwriting
fees as a percentage of proceeds raised, are more significant for IPOs compared to
seasoned equity or debt offerings (Dunbar, 2000). The underwriter may have to face the
loss of future revenues of new issues owing to its association with the overpriced
offering. Hence, the underwriting bank’s market share will be reduced in the future as
investors may be hesitant in trusting its future issues pricing. Further, the management
of future IPOs are likely to refrain from using a bank that is associated with significant
underpricing. In fact, their reputation is damaged every time they generate inaccurate
information, which leads to inaccurate pricing (Chemmanur and Fulghieri, 1994). These
market share effects are expected to be greater in established banks as their reputation
is more at risk.
Moreover, underpricing is quite pricey to the issuing company and hence, low-risk
companies try to relay their low-risk features to the market by employing an
underwriter with a good reputation (Carter and Manaster, 1990). The selection of such
an underwriter could be considered as a signal where high-quality underwriters will
be chosen by companies with good news and information (Titman and Trueman,
1986). The reputation of the underwriter can minimize the adverse selection and
moral hazard issues and in effect, decrease the ex ante uncertainty linked with IPO
(Cooney et al., 2001). Similarly, according to empirical evidence (e.g. Carter and
Manaster, 1990; Carter et al., 1998), underwriters are more inclined to underwrite
issues of low-dispersion companies, and low-dispersion companies are often linked to
less underpricing. Other researchers (Kenourgios et al., 2007; Jones and Swaleheen,
2010) reported that the reputation of the underwriter significantly influence the IPO
underpricing level. Evidence was presented by Megginson and Weiss (1991) to the
fact that lower underpricing is related to the underwriter’s relative market shares. To
this end, prior literature and arguments suggest that underpricing should be inversely
associated to the IPO’s underwriter’s market share. Hence, the present study
hypothesized that:

H1. Underwriter’s market share are negatively associated with IPO underpricing.
IJMF 2.2 Underwriter’s spread
12,3 As mentioned before, underwriters have a key role in the process of IPO as they are
advisors to the issuing company, they set offer prices and they sell the new issues.
In return, they are compensated in the IPO process in terms of gross spread. Thus,
the underwriting spread is the revenue that is generated by the underwriters when they
underwrite the new issues. The underwriting services price has great implications for
354 investors. For example, underwriting fees may be associated to the offer price and the
underpricing level (Chen et al., 2006). The industry of underwriting should therefore
be characterized by high competition with low profit margins (Bartling and Park,
2009).
The revenue of underwriters are obtained from the spread, which is composed of
two elements, namely, fixed and variable. The former is often linked with the offer
size while the variable is linked with price uncertainty of the new issue (Chen and
Mohan, 2002). This is explained by the fact that underwriters face loss if the IPO sells
at a price lower than the preset offering price. Underwriters are however prevented
from selling the issue at a higher price with the increase of market price. Thus,
underwriter spread comprises compensation for the risk bearing function. In other
words, underwriters tend to underprice the new issues so that the IPO will be
completely subscribed and increase in price directly after the offer (McDonald and
Fisher, 1972). That is, the proportion of spread is directly associated to the risk
for each new issue. The perception is that higher uncertainty about the future
equilibrium price encourages underwriters to underprice the new issues and therefore
charge specific level of spread that would be sufficient to bear the full cost of risk.
Thus, the variables underpricing and spread complement each other and the
association between them hinges on the issuing company’s attributes and the power of
negotiation, level of competition in the market and the underwriter’s strategies of
pricing. The variable of spread is included in this study as the underwriter spread
may be limited by regulation or competition or both in a way that different
underpricing levels are required for full cost of risk bearing absorption. The spread
encourages the investment bank to select risky prices. If the spread is considerably
low, banks steer clear of failing IPO and set a corresponding low, risk-free price, and
at this price level, there would be great level of underpricing. Chen and Mohan (2002)
claimed that spread is significantly and positively associated with IPO underpricing,
which leads us to the following hypothesis:
H2. Underwriter’s spread are positively associated with IPO underpricing.

2.3 Earnings forecasts bias

The earnings forecasts, provided by the management in the prospectuses of Malaysian
IPOs, offer considerable information for potential investors in Malaysia. The lack of
prior record of forecasts accuracy will lead investors to assess earnings forecasts and
to determine which of them are overly optimistic and which are pessimistic. Upon
successful differentiation of pessimistic forecasts from optimistic forecasts, investors
then have the ability to distinguish the premium levels that are related to the accuracy
of forecasts ( Jelic et al., 2001). As such, less underpricing would be expected for
conservative (unbiased) forecasts. However, if investors are unable to evaluate the
bias in forecasts, they will be inclined to pay a premium for IPOs disclosing
optimistic forecast, and this may lead to higher underpricing for such companies.
Investors’ ability in analyzing earnings forecasts at the time of their disclosure should
be noted in this context. If investors only depend on these forecasts without being
completely
aware of the bias level, stock prices may show overly optimistic earnings forecasts.
Overly optimistic forecasts will consequently lead to higher information asymmetry. Underpricing
Hence, it is hypothesized that the higher the management earnings forecasts error, the of Malaysian
greater will be the information asymmetry, and the greater will be the underpricing. IPOs
We therefore propose a positive association between forecast bias and the
underpricing. We specify the following hypothesis:
H3. Earnings forecasts bias is positively associated with IPO underpricing. 355
2.4 Earnings forecasts accuracy
IPOs are found to be less underpriced in countries where existing public companies
generate greater quality earnings information (Boulton et al., 2011). Additionally,
large underpricing in the immediate after market can signal great forecast error. Some
managers become evidently incapable of hiding their inability to offer accurate
earnings forecasts so they underprice issues to attract potential investors. If the issue
price is considered as a function of the earnings forecasts and underpricing is a fixed
percentage, then any noted differences in the IR may be due to the investor’s
perceptions of the accuracy of earnings forecasts. The forecasts accuracy-IR
relationship may also be explained by Benveniste and Spindt’s (1989) model. The
model postulates that the IR is a positive uncertainty function; stated differently, the
greater the forecasts error, the higher will be the uncertainty and underpricing.
However, accurate earnings forecasts will accordingly result in less information
asymmetry. Therefore, it is suggested that the greater the management earnings
forecasts accuracy, the lesser will be the information asymmetry, and the lesser will
be the underpricing. Based on the above, the following hypothesis is proposed:
H4. Earnings forecasts accuracy is negatively associated with IPO underpricing.

3. Data and research method

3.1 Data
This study’s data cover the publicly available information, gathered from the IPO
prospectuses and annual reports of the listed companies in Bursa Malaysia. The
study’s sample includes companies that were incorporated on the main market of
Bursa Malaysia for the duration 2002-2012. As of February 29, 2012, 265 financial
and non- financial IPOs are listed on the main market. Financial companies, close-end
fund and real estate investment trust companies are excluded following the studies of
Karamanou and Vafeas (2005) and Ahmad-Zaluki and Wan-Hussin (2010).
Accordingly, three companies from the finance sector, one from close-end fund and
fourteen from the real estate investment trust were not included. Two infrastructure
project companies were also excluded as proposed by Ahmad-Zaluki et al. (2007).
These industries are not included as their regulatory requirements are different from
the sample companies. Moreover, IPOs that did not issue management earnings
forecasts in the prospectus are omitted from the sample. Accordingly 54 IPO
companies without earnings forecasts and 21 with different regulatory regimes were
excluded which resulted in the final sample of 190 IPO companies.

3.2 Research methods

As with prior IPO studies dedicated to performance, underpricing is proxied by the
IR
of IPO companies. IR is measured as the difference between the closing price on the
IJMF initial day of trading and the initial offering price, expressed as a percentage of the
12,3 initial offering price. A positive value indicates an underpriced security. The IR on
the first trading day is therefore calculated as follows:
Σ. Σ
IR i;t ¼ Pi;1 – Pi;0Σ=Pi;0 ~ 100% (1)
356 where IRi,t is the IR of company i; Pi,1 is the closing price of company i at the end of the
first trading day; Pi,0 is the offer price of company i.
We test our four hypotheses regarding the impact of underwriter’s reputation and
spread and the impact of earnings forecasts bias and accuracy on the IPO underpricing
using both of univarate analysis and multivariate regression analysis. In the
multivariate regression analysis we use IR as dependent variable and underwriter’s
reputation, underwriter’s spread, earnings forecasts bias, earnings forecasts accuracy
as independent variables. We also include a number of control variables such as
family ownership, management ownership, auditor’s quality, size, age, forecasts
horizon and leverage. These control variables were generally employed in prior IPO
researches as proxies for the apparent IPOs risk. They are expected to capture the
potential impact of ownership, auditor’s prestige, forecasts horizon and company’s
size, age and leverage on underpricing of the IPO companies. Our pooled ordinary
least squares (OLS) regression model is presented as follows:
IRi;t ¼ b0 þb1 UND_MSi;t þb2 UND_SPRi;t þb3 EF_BIASi;t
þb4 EF — ACCURACYi;t þb5 FOWNi;t þb6 MOWNi;t þb7 AUDi;t
þb8 CSI Z E i;t þ b9 CAGE i;t þ b10 FHORI Z ON i;t þ b11 LEV i;t þe (2)
where UND_MSi,t (underwriter’s market share) is the ringgit value for all shares
underwritten by the underwriter divided by the ringgit value of all underwritten IPOs
in the sample; UND_SPRi,t (underwriter’s spread) the ratio of the gross fees charged
by the underwriter including underwriting commission, management fees, sales
commissions and other fees charged by the underwriter to the gross proceeds of
the issue; EF_BIASi,t (forecasts bias) the difference between actual earnings and the
earnings forecasts scaled by absolute earnings forecasts; EF_ACCURACYi,t (forecasts
accuracy) the absolute difference between the actual earnings and the earnings
forecasts divided by the absolute earnings forecasts; FOWNi,t (family ownership) the
percentage of shares retained by the family directors on the IPO board of directors;
MOWNi,t (management ownership) the percentage of shares retained by the executive
directors on the IPO board of directors; AUDi,t (auditor’s quality) a binary variable
equal to “1” if the auditor of IPO company is Big 4 (PricewaterhouseCoopers,
Deloitte, KPMG, Ernst and Young) and “0” otherwise, CSIZEi,t (company size) the
natural logarithm of total assets, at the date of IPO prospectus; CAGEi,t (company age)
the natural logarithm of (1+ number of years between the date of establishment and
the date of IPO); FHORIZONi,t (forecasts horizon) the number of months from the
management forecast date (prospectus date) to end of the period for which the
forecast is made; LEVi,t (leverage) the the percentage of total liabilities over the total
assets.

4. Analysis
4.1 Descriptive statistics
The descriptive statistics of IR of Malaysian IPOs for the years 2002-2012 are
presented in Table I. Panel A displays an average underpricing of 21.22 percent
with standard
Panel
2012 A: descriptive statistics of the underpricing of 190 Malaysian IPOs between 2002
Underpricing
Dependent
Mean Median SD Min. Max. t-test Kolmogorov- of Malaysian
variable Smirnov test IPOs
t-value p-value z-value p-value
IR 21.22 12.92 34.04 −33.33 245.45 8.590 0.000 1.937 0.001
Panel B: distribution of IR
n Percentage 357
IR o0% 29 15.26
0% oIR W 98 51.58
25%
25% oIR W 50% 36 18.95
50% oIR W 75% 20 10.53
75% oIR W 2 1.05
100%
IR W 100% 5 2.63
Total 190 100
Panel C: IPO erpricing (yea rly)
und n Mean Median SD
Year
2002 43 14.52 7.69 25.98
2003 36 41.38 34.52 44.89
2004 41 26.57 20.00 39.41
2005 26 6.14 0.00 17.26
2006 14 11.25 12.14 20.79
2007 19 24.69 17.65 26.07
2008 10 1.12 3.19 19.27
2009 – – – –
2010 – – – –
Notes: This table exhibits the summary statistics of the initial return ( IR) as proxy of
the underpricing for the 190 Malaysian IPOs. Average raw initial returns are computed on the basis
of equally weighted. IR (closing
¼ price − offer price)/offer price) × 100 percent. Distribution of
raw initial return is presented in Panel B. In Panel C, most IPO companies did not issue earnings
forecasts in 2009, 2010, 2011 (the only IPOs that issued earnings forecast were from real estate Table I.
investment trust sector, which is excluded from our sample). p-values show the level of significance Descriptive statistics
for the two-tailed t-test and Kolmogorov-Smirnov test of mean and median, respectively, at the of the IPO
1 percent level of significance underpricing

deviation of 34.04 percent. Contrary to prior studies dedicated to Malaysian IPOs, the
present study reports a significantly lower average amount of underpricing at the level
of 1 percent through t-test and Kolmogorov-Smirnov test. Specifically, underpricing
varies from −33.33 percent (the lowest return displayed by Sin Chew Media
Corporation Berhad on October 18, 2004), to 245.45 percent (the highest return
displayed by PMB Technology Berhad on November 5, 2003). These findings show
that the average level of IPO underpricing is lower compared to the percentage
revealed by prior authors (e.g. Dawson, 1987 at 166.7 percent; Jelic et al., 2001 at 99
percent; Yong and Isa, 2003 at 94.91 percent; Murugesu and Santhapparaj, 2009 at 81
percent). In general, IPO underpricing decreased from the last period until 2012.
This lower average underpricing indicates that some IPO issues in Malaysia have
been resolved and become more efficient through the new method liberalized by the
Securities Commission in 1996, namely, market-based pricing. This mechanism
placed the total responsibility on issuers and advisers and allowed them to set IPOs
price.
IJMF Panel B displays that majority of the IPO companies, numbering 98 IPOs reported
12,3 underpricing at 0-25 percent. The samples performance throughout the years and their
IPOs underpricing are also presented in Panel C, with no overpricing in any year.
Panel C indicates that the greatest level of underpricing at the Malaysian IPO market
was experienced in 2003 with a mean and standard deviation of 41.38 and 44.89
percent, respectively. On the other hand, the lowest degree of underpricing was
358 experienced in 2008 with a mean and standard deviation of 1.12 and 19.27 percent,
respectively.
The descriptive statistics of IPO samples are also presented in Table II. The table
indicates that the mean underwriter market share is 11 percent, a minimum of 0.07
percent by Kenanga Investment Bank Berhad and a maximum of 28.35 percent by
CIMB Investment Bank Berhad. In total, 13 Malaysian investment banks played a key
role in underwriting in the period under study. However, only five of these
underwriters have played a leading role of over 15 new issues (53, 51, 18, 18 and 16)
[3]. These banks were deemed to have the highest market share, with the total market
share of the top five underwriters to be over 54.44 percent. This indicates that the
Malaysian underwriting industry is quite concentrated. From the list, CIMB
Investment Bank Berhad and RHB Investment Bank Berhad are both leading the
industry with 28.35 and
16.14 percent market shares, respectively. Meanwhile, the mean underwriter spread is
4.82 percent and the median is 2.01 percent. As for the maximum underwriting
spread, it is charged by CIMB Investment Bank Berhad at 15.93 percent on the issues
of DXN Holdings Berhad. The mean of underwriting spread in Malaysia is quite
similar to those in the Korean IPO market (5.93 percent in Ahn et al., 2007) as well as
in Australian IPO market (3.7 percent in How and Yeo, 2001).

Mean Median SD Min Max

UND_MS (%) 11 6.03 11.07 0.07 28.35
UND_SPR (%) 4.82 2.01 1.84 0.12 15.93
EF_BIAS (%) 3.36 1.12 54.50 −176.04 525.45
EF-ACCURACY (%) 24.94 9.37 48.54 0.13 525.45
FOWN (%) 15.68 5.52 19.37 0.00 66.13
MOWN (%) 11.76 4.15 15.58 0.00 66.94
AUD (%) 54.74 – – – –
CSIZE (RM million) 350.11 101.19 1,402.58 35.12 17,073.86
CAGE (years) 5.53 2.25 6.95 0.17 32.67
FHORIZON (months) 7.71 7.00 3.01 3.00 14.00
LEV (%) 51.04 47.83 23.89 3.86 150.48
Notes: This table shows the descriptive statistics of the IPO companies used in this study. UND_MS
is the ringgit value for all shares underwritten by the underwriter divided by the ringgit value of all
IPOs in the sample. UND_SPR is the ratio of the gross fees charged by the underwriter including
underwriting commission, management fees, sales commissions and other fees charged by the
underwriter to the gross proceeds of the issue. Earnings forecasts bias (EF_BIAS)¼ (actual
earnings−forecast earnings)/| forecast earnings|. Earnings forecasts accuracy (EF-ACCURACY) ¼
| (actual earnings − forecast earnings)|/| forecast earnings|. FOWN is the percentage of shares
retained by the family members on the IPO board of directors. MOWN is the percentage of shares
Table II. retained by the executive directors on the IPO board of directors. AUD is a dummy variable equal to
Descriptive statistics “1” if auditor is Big 4 and “0” otherwise. CSIZE is the ln of total assets, at the date of IPO prospectus.
for 190 Malaysian CAGE is the ln (1+ number of years between incorporation date and the IPO date). FHORIZON is the
IPOs between 2002 number of months between the management forecast date (prospectus date) to end of the period for
and 2012 which the forecast has been conducted. LEV is total liabilities over the total assets
The findings of the mean of forecasts bias is 3.36 percent with the positive sign
indicating that actual earnings are normally greater than the forecasted earnings in the Underpricing
prospectuses, and hence, the forecasts are pessimistic. This mean is significantly of Malaysian
lower than the mean reported in prior studies conducted in Malaysia (i.e. 9.34 percent IPOs
in Mohamad et al., 1994; 33.37 percent in Jelic et al., 1998), but akin to Hussin et
al.’s (2004) study that reported 2.10 percent. This shows that general reduction in the
forecast bias of Malaysian IPOs is to be expected over time. Moreover, the forecasts
accuracy mean for the samples is 24.94 percent, indicating that the earnings 359
forecasts accuracy is greater compared to the accuracy reported in prior studies in
Malaysia (i.e. Mohamad et al., 1994 reported 28 percent), in Thailand (Lonkani and
Firth, 2005 reported
35.67 percent) and in Australia (Firth et al., 2012 reported 34.49 percent), and lower
compared to other countries – in Hong Kong (McGuinness, 2005 reported 7.26
percent), in New Zealand (Firth, 1997 reported 11.10 percent) and similar to Ahmad
Zaluki and Wan-Hussin’s (2010) study that reported 23.76 percent.
Furthermore, a percentage constituting 15.68 percent with maximum of
66.13 percent and a minimum of 0 percent is shares owned by the family members
on the IPO companies’ boards. The executive directors retain an average percentage
of shares equating to 11.76 percent of the whole sample. Around 54.74 percent
constituting 104 IPOs have been audited by the Big 4. The IPO companies size
greatly differs, with the largest company, namely, Plus Express Ways Berhad, having
a total asset amounting to RM17073.86 million. The least size is TAFI Industries
Berhad with
35.12 million. The average operating history of the IPO sample is about 5.53 years,
minimum of 0.17 year and maximum of 32.67 years. In addition, the average
forecasts horizon is approximately 7.71 months. With regards to the leverage, the
mean of liabilities to total asset is 51.04 percent.

4.2 Univariate analysis

The univariate analysis for the whole sample on the basis of subgroups of each
experimental variable is listed in Table III. The hypotheses are tested by developing
two portfolios based on high and low underwriter’s market share, underwriter’s spread
and accuracy of earnings forecasts. As for the bias in earnings forecasts, two groups
are developed based on optimistic and pessimistic forecasts.
The results presented in Panel A of Table III indicate that underwriters with great
market share are inclined to increase initial underpricing, which is contrary to the
reported results of underwriter’s role (e.g. Beatty and Ritter, 1986; Carter et al., 1998;
Ahmad-Zaluki and Kect, 2012). The results of the parametric t-test and non-
parametric Mann-Whitney test show significant differences at the level of 1 percent
between average returns of IPOs underwritten with high market share and IPOs
underwritten by their less market share counterparts. This result is inconsistent with
our hypothesis one, suggesting that underwriter market share is negatively related
with underpricing. The result of underwriter’s spread, depending on the distribution of
sample, is remarkable. The group sample of high spread reported underpricing of
17.44 percent while the low spread group reported underpricing of 24.55 percent. The
difference between them shows a statistical significance at the level of 5 percent on
the basis of non-parametric Mann- Whitney test. In other words, the Malaysian IPO
companies that paid high underwriting spread are more likely to be less underpriced –
a fact inconsistent with the findings of Chen and Mohan (2002) who reported that
companies paying high spread to underwriters are related to greater underpricing
compared to those who paid low underwriting spread. On this basis, the finding does
not support our hypothesis two.
 Tabl
e III.
36 IJMF
Univ 0 12,3
ariate
analy
sis

Panel A: level of underpricing based on the distribution of underwriter’s market share, underwriter’s spread and earnings forecasts accuracy
Grouping variables Median (cut-off point) n Mean Median SD t-test Mann-Whitney test
t-value p-value z-value p-value
UND_MS ⩾6.03 High market share 120 16.66 16.14 10.33 16.229 0.000 −11.733 0.000
o6.03 Low market share 70 1.29 1.11 0.77
UND_SPR ⩾2.01 High spread 95 17.44 8.33 34.78 −1.437 0.152 −1.992 0.046
o2.01 Low spread 95 24.55 16.98 33.35
EF-ACCURACY ⩾9.37 Low accuracy 95 20.08 18.75 31.89 0.662 0.509 −1.472 0.141
o9.37 High accuracy 95 23.35 7.69 35.98
Panel B: level of underpricing based on the distribution of biased earnings forecasts
Sign n Mean Median SD t-test Mann-Whitney test
t-value p-value z-value p-value
EF_BIAS W0 Pessimistic 105 21.97 17.19 28.497 −0.337 0.737 −1.689 0.091
o0 Optimistic 85 20.29 6.25 40.008
Total 190
Notes: This table shows the results of univariate analysis of Malaysian IPO underpricing in relation with the main four experimental variables ( UND_MS,
UND_SPR, EF_BIAS and EF-ACCURACY). We used a median value as the cut-off point to distinguish between the high and low UND_MS, UND_SPR and
EF-ACCURACY. For the EF_BIAS, we use “0” value to differentiate between the optimistic and pessimistic earnings forecasts. UND_SPR is the ratio of the
gross fees charged by the underwriter including underwriting commission, management fees, sales commissions and other fees charged by the underwriter
¼ (actual earnings − forecast earnings)/| forecast earnings|. Earnings forecasts accuracy
to the gross proceeds of the issue. Earnings forecasts bias (EF_BIAS)
(EF-ACCURACY)¼| (actual earnings − forecast earnings)|/| forecast earnings|. Results significantly different from zero at the 10, 5 and 1 percent levels,
respectively (two-tailed)
Based on the results, companies having accurate earnings forecasts are more
underpriced compared to those with inaccurate ones. The difference is however Underpricing
insignificant. Specifically, the riskiest IPOs related with low accuracy (with EF- of Malaysian
ACCURACY greater than 9.37 percent) display a lower underpricing level. This does IPOs
not support Gounopoulos (2011) and also rejects hypothesis four of this study. In
Panel B of Table III, IPOs with pessimistic earnings forecasts are linked with higher
degrees of underpricing in the immediate after market. The difference between
optimistic and pessimistic companies in terms of underpricing level is statistically 361
significant at a 10 percent level according to the non-parametric Mann-Whitney test.
Despite the fact that the optimism degree is not known at the time of IPO,
the underpricing is lower for the optimistic forecasts, even though the forecasts turned
out to be biased. Furthermore, this is evidence that Malaysian market has no
mechanisms to recognize IPOs with optimistic forecasts. This result is also
inconsistent with our hypothesis three, but in line with the results of Firth (1998),
Chen et al. (2001), Jelic et al. (2001) and Jog and McConomy (2003).

4.3 Pairwise correlation

The Pearson correlation coefficients matrix for the entire variables in this study is
listed in Table IV. We also present the Spearman correlation above the diagonal in an
effort to mitigate the effect of outliers. The market share of underwriters is
significantly and positively associated with the reputation of the auditor and the size
of the IPO company. However, it is negatively associated with forecast bias, forecast
accuracy, family ownership, management ownership and leverage. Of particular
interest is that a negative relationship is found between the underwriter’s market share
and spread, indicating that underwriters with high market share may need smaller
spread for the underwriting contract. This is supported by Carter and Dark’s (1992)
and Chen and Mohan’s (2002) argument that reputable underwriters have a tendency
to mitigate IPO spread. On a closer examination of the underwriter spread-other
variables relationship, the spread is highlighted to be significantly and negatively
associated with the company size. Furthermore, the underwriter spread is negatively
but insignificantly associated with forecast bias and accuracy. Forecast errors along
with absolute forecast error are significantly and positively associated with family
ownership and management ownership. On a final note, forecast error is found to be
negatively and significantly associated with the IPO company’s leverage
percentage. Further, from Table IV we can validate the nonexistence of
multicolinearity between the independent variables since no correlation coefficient
goes further than 0.8.

4.4 Pooled OLS regression of IR and independent variables

For hypotheses testing, Equation (2) is tested by employing pooled OLS multiple
regression. More specifically, the multivariate regression analysis was run to examine
the relations among the level of IPO IR, market share of underwriter, spread of
underwriter, earnings forecasts bias and earnings forecasts accuracy, along with
additional control variables. Specifically, we present two regression model results
in Table V. Columns 3 and 4 report the results of the first regression model (Model
1), which consider all experimental variables without any control variables, and
columns 5 and 6 show the results of the second regression model (Model 2), which
consider all experimental variables, and other control variables. The results from the
regression of Model 2 with control variables are, to some extent, similar to those from
the regression
 Table IV. Correlation 36 IJMF
coefficients between
variables (Pearson 2 12,3
(Spearman) correlation
coefficients are above
(below) the diagonal)

1 2 3 4 5 6 7 8 9 10 11
1. UND_MS 1 −0.099 −0.014 −0.130* −0.057 −0.069 0.131* 0.240** 0.052 0.031 0.008
2. UND_SPR −0.043 1 0.058 −0.031 0.071 0.047 −0.088 −0.186* −0.050 0.056 0.047
3. EF_BIAS −0.062 −0.015 1 −0.030 −0.006 −0.018 −0.022 0.046 −0.070 −0.089 −0.052
4. EF-ACCURACY −0.107 −0.058 0.647*** 1 0.043 0.006 −0.006 0.054 −0.032 0.071 0.034
5. FOWN −0.081 0.011 0.121* 0.128* 1 0.567*** −0.092 −0.308*** −0.011 0.026 −0.022
6. MOWN −0.090 −0.029 0.126* 0.126* 0.579*** 1 −0.133* −0.336*** −0.012 0.046 0.057
7. AUD 0.128* 0.003 −0.094 −0.077 −0.097 −0.132* 1 0.231*** 0.123* −0.044 −0.111
8. CSIZE 0.265*** −0.139** −0.061 −0.010 −0.294*** −0.300*** 0.255*** 1 0.127 0.057 0.222**
9. CAGE 0.029 0.011 −0.038 −0.041 −0.045 −0.086 0.126* 0.145** 1 0.083 −0.044
10. FHORIZON 0.038 0.025 −0.036 0.100 −0.001 0.058 −0.041 0.023 0.077 1 −0.143*
11. LEV −0.038 0.003 −0.149** −0.077 −0.010 0.027 −0.128* 0.107 −0.062 −0.164** 1
Notes: This table presents the correlations among variables used in the pooled OLS regression. UND_MS is the ringgit value for all shares underwritten by
the underwriter divided by the ringgit value of all IPOs in the sample. UND_SPR is the ratio of the gross fees charged by the underwriter including
underwriting commission, management fees, sales commissions and other fees charged by the underwriter to the gross proceeds of the issue. Earnings
¼ (EF_BIAS) (actual earnings − forecast earnings)/| forecast earnings|. Earnings forecasts accuracy (EF-ACCURACY) | (actual earnings −
forecasts bias
forecast earnings)|/
| forecast earnings|. FOWN is the percentage of shares retained by the family members on the IPO board of directors. MOWN is the percentage of shares
retained by the executive directors on the IPO board of directors. AUD is a dummy variable equal to “1” if auditor is Big 4 and “0” otherwise. CSIZE is the
ln of total assets, at the date of IPO prospectus. CAGE is the ln (1+ number of years between incorporation date and the IPO date). FHORIZON is the
number of months between the management forecast date (prospectus date) to end of the period for which the forecast has been conducted. LEV is total
liabilities over the total assets.
*,**,***Significant at the 10, 5 and 1 percent levels, respectively (two-tailed)
 Model 1 Model 2
Underpricing
Experimental variables only All variables of Malaysian
Expected sign Coeff. t-stat Coeff. t-stat IPOs
Experimental variables
UND_MS − 0.048 0.39 0.125 0.99
UND_SPR + −1.381 −1.91** −0.924 −0.95
EF_BIAS + 0.079 2.21** 0.099 2.86** 363
EF-ACCURACY − −0.053 −1.30 −0.075 −1.96*
Control variables
FOWN − −0.024 −0.29
MOWN − −0.144 −1.38
AUD − 0.916 0.32
CSIZE − −4.639 −2.94**
CAGE − 3.535 2.21*
FHORIZON + 1.506 3.23**
LEV + −6.313 −1.04
Constant 16.337 5.33*** 57.143 3.02**
n 164 164
Adjusted R2 2.41% 12.69%
F-value 2.40** 3.150***
Notes: This table reports pooled OLS regression results explaining the dependent variable (IR).
UND_MS is the ringgit value for all shares underwritten by the underwriter divided by the ringgit
value of all IPOs in the sample. UND_SPR is the ratio of the gross fees charged by the underwriter
including underwriting commission, management fees, sales commissions and other fees charged by
the underwriter to the gross proceeds of the issue. Earnings forecasts bias (EF_BIAS) (actual
earnings − forecast earnings)/| forecast earnings|. Earnings forecasts accuracy (EF-ACCURACY) ¼
| (actual earnings−forecast earnings)|/| forecast earnings|. FOWN is the percentage of shares retained¼
by the family members on the IPO board of directors. MOWN is the percentage of shares retained by
the executive directors on the IPO board of directors. AUD is a dummy variable equal to “1” if
auditor is Big 4 and “0” otherwise. CSIZE is the ln of total assets, at the date of IPO prospectus.
CAGE is the ln (1+ number of years between incorporation date and the IPO date). FHORIZON is
the number of months between the management forecast date (prospectus date) to end of the period
for which the forecast has been conducted. LEV is total liabilities over the total assets. According to
Belsley et al. (1980), outliers are identified when the absolute values of the observations’ studentized
residuals are two or greater. To confirm and delete the outliers, this study conducts the studentized
residuals test. Therefore, 26 IPO companies have been removed from the sample. *,**,***Significant Table V.
at the 10, 5 and 1 percent levels, respectively Determinants of IPO
underpricing

of the Model 1 without control variables. The only difference is that the underwriter’s
spread (UND_SPR) is found to be significantly related with IPO underpricing in
Model 1 while not in Model 2. Further, earnings forecasts accuracy (EF-ACCURACY)
is found to be significantly associated with IPO underpricing in Model 2 while not in
Model 1. However, the significance level and adjusted R-squared of Model 2 is higher
than those reported for Model 1, suggesting that it is important to include the control
variables in our pooled regressions.
Focussing on the results of Model 2, the results presented that in contrast to the
first hypothesis, the underwriter’s market share (UND_MS) is positively and
insignificantly associated with underpricing. This result goes against the result
reached by prior studies that underwriter’s reputation is negatively and significantly
related to IR (e.g. Carter and Manaster, 1990; Carter et al., 1998). However, this
result is in line with some previous studies (e.g. Beatty and Welch, 1996;
Loughran and Ritter, 2004).
IJMF Specifically, Beatty and Welch (1996) reported a positive association between
12,3 underpricing and underwriter reputation and suggested that the underwriter behavior
change may be attributed to economic situations. This result also follows Kirkulak
and Davis’s (2005) study that involved data from IPOs in Japan and suggested that
reputable underwriters are associated with a greater degree of underpricing. Similarly,
Bates and Dunbar (2002) showed that positive association between market share of
364 reputable underwriters and underpricing indicates a structural change in the US IPO
market owing to the high-tech boom. On the other hand, Loughran and Ritter (2004)
showed inconclusive findings – they showed a negative relationship between 1980
and 1989 and a positive one in 1990-1998, 1999-2000 and 2001-2003. This reversal is
attributed in part to the changes in the objective function of the issuer. Based on their
hypothesis, issuers in the 1990s were more concerned with analyst research coverage
rather than underpricing and they had the tendency to accept higher underpricing from
an underwriter who provides higher analyst coverage. According to Logue et al.
(2002), pre-market underwriting activities rather than the reputation of the
underwriter plays a part on IPO returns. Their findings supported no significant
negative relationship between reputation of underwriter and IPO underpricing.
More importantly, the present study’s result of underwriter’s spread (UND_MS), is
not similar to the results in the IPO underpricing literature. In this study, spread is
found to be negatively but insignificantly related to IR. This goes against our second
hypothesis that postulates that deeper underpricing is often followed by higher
underwriter spread and is inconsistent with Chen and Mohan’s (2002) finding that the
more the spread, the more will be the underpricing. In this study, earnings forecasts
error (EF_BIAS) is positively and significantly associated with IR. The positive
coefficient of EF_BIAS supports our third hypothesis that forecast error proxies
ex ante uncertainty and is positively associated with the IPO IR. The result indicates
that share price heightens in reaction to the negative forecasts errors and it reduces in
reaction to positive ones on the date of announcement. This supports Henry et al.’s
(2002) research, which was focussed on Australian IPOs. In addition, the result does
not support the notion that investors are able to accurately predict forecast bias and
incorporate it to the pricing of IPOs[4]. The finding regarding earnings forecasts
accuracy (EF-ACCURACY) shows a negative and significant association with IPO
underpricing. This explains that the higher the forecast accuracy of management,
the lower will be the underpricing. It supports our fourth hypothesis stating that the
higher the forecast accuracy, the lower will be the asymmetric information, hence,
less IPO underpricing.
In terms of control variables, we found that family ownership (FOWN) is
negatively related with IR. The result is consistent with Shi et al. (2012), in that IPO
companies having more family directors are less underpriced. We also report that the
more management ownership (MOWN) there is, the less there will be the IPO
underpricing. On the other hand, the auditor’s quality (AUD) has the opposite sign.
However, the result is statistically insignificant. As the case with prior studies, the
result of CSIZE in this study revealed that the lowest underpriced IPOs are those
wherein most of the incorporated companies are large companies. On the basis of the
theoretical argument, big companies are expected to be well established and have a
great experience. They have lower information asymmetry compared to their smaller
counterparts. Additionally, a positive relationship was found between underpricing
and IPO company age (CAGE ). In other words, the longer the operating history of the
IPO, the deeper will be their underpricing. The significant positive association
between
company age and IR goes against the ex ante uncertainty model (Beatty and Ritter,
1986). Based on the signaling theory, age proxies ex ante risk and in this case Underpricing
investors are privy to information of old companies and hence, uncertainty is lower. of Malaysian
The forecast horizon (FHORIZON) variable is positively and highly significant in IPOs
relation with IR. The delay in issuing subscription and the prolonged stock exchange
listing shows that long forecast horizon could affect IPO valuation negatively owing
to the increased noise and uncertainty. This risk is mitigated by the heavy
underpricing in new issues. The leverage (LEV) is negatively related with IPO 365
underpricing and this result is consistent with that of Sun and Liu’s (2009).

5. Robustness analysis
5.1 Sensitivity analysis on underpricing
In the above analysis, underpricing is defined as (IRi,t ¼ ((Pi,1 – Pi,0)/Pi,0) × 100 percent).
In this section, robustness of our results is checked and accordingly, we present the
results using market adjusted initial return (MAIR) as the proxy of IPO underpricing.
The distinction between this proxy and IR was more evident in the 1990s when the
listings are conducted way after the offering. In large time lag periods, several
changes in market dynamics could take place and the IR may be influenced by them.
Therefore, the raw IR is adjusted for market variances. Owing to the size and markets
variation of the sample companies, the KLSE index is employed to calculate MAIR:
Σ. Σ
MAIR i;t ¼ IR i;t— I i;1–I i;0 =I
Σ i;0 (3)

where MAIRi,t is the market adjusted IR of company i; IRi,t is the IR of company i; Ii,1
is the KLSE composite index at the end of the first trading day of company i; Ii,0 is the
KLSE composite index at the end of the approval day of company i. We find that the
mean of MAIR is about 19.73 percent which is less than IR (21.22 percent) found
earlier. The regression analysis (I) reported in Table VI shows that the results of the
experimental variables are similar to those of initial regression using IR as IPO
underpricing proxy in Table V.

5.2 Alternative measurements for some experimental variables

In the results of regression (II) presented in Table VI, IR is considered as the
dependent variable but with different measurement for the UND_MS, EF_BIAS and
EF-ACCURACY. The UND_MS is measured as a dummy variable equal to “1”, if
the underwriter is among the four most reputable banks; AmInvestment Bank, CIMB
Investment Bank Berhad, Alliance Investment Bank Berhad and MIMB Investment
Bank Berhad, otherwise equal to “0.” Similar to Jog and McConomy’s (2003) study, in
this part, this study measured EF_BIAS as a dummy variable equal to “1” if the
forecast is optimistic, and “0” if not. For the EF-ACCURACY, we used squared
forecast error metric as the squared of difference between actual earnings and the
earnings forecasts scaled by earnings forecasts. The results of regression (II) slightly
differs from the initial one in Table V. Here, the EF_BIAS is significant at the level of
10 percent and EF-ACCURACY is insignificant.

6. Conclusion
The main purpose of this study is to examine the effect of attributes of underwriters
and management earnings forecasts on the underpricing of Malaysian IPOs.
Accordingly, four issues are addressed; the role played by the underwriter market
 IJMF Regression I Regression II
12,3 Coeff. t-stat Coeff. t-stat
Experimental variables
UND_MS 0.107 0.87 4.651 1.55
UND_SPR −1.352 −1.43 −1.076 −1.11
EF_BIAS 0.103 3.03** 6.231 2.30*
366 EF-ACCURACY −0.081 −2.18* −0.001 −0.32
Control variables
FOWN 0.007 0.09 −0.005 −0.06
MOWN −0.148 −1.45 −0.1802 −1.68*
AUD 1.497 0.55 1.115 0.39
CSIZE −3.584 −2.36* −5.214 −3.34***
CAGE 2.748 1.78* 3.583 2.22*
FHORIZON 1.470 3.24*** 1.115 2.44**
LEV −2.654 −0.47 −11.023 −1.85*
Constant 43.962 2.42** 63.083 3.36***
n 166 164
Adjusted R2 10.6% 11.71%
F-value 2.78** 2.97***
Notes: This table reports the robustness check of the pooled OLS regression results of equation 2
using MAIR as a dependent variable in regression I and different measurements for experimental
variables in regression II. UND_MS is the ringgit value for all shares underwritten by the underwriter
divided by the ringgit value of all IPOs in the sample. UND_SPR is the ratio of the gross fees charged
by the underwriter including underwriting commission, management fees, sales commissions and
other fees charged by the underwriter to the gross proceeds of the issue. Earnings forecasts bias
(EF_BIAS) ¼(actual earnings − forecast earnings)/| forecast earnings|. Earnings forecasts accuracy
(EF-ACCURACY)¼| (actual earnings − forecast earnings)|/| forecast earnings|. FOWN is the percen-
Table VI. tage of shares retained by the family members on the IPO board of directors. MOWN is the
Determinants of IPO percentage of shares retained by the executive directors on the IPO board of directors. AUD is a
underpricing using dummy variable equal to “1” if auditor is Big 4 and “0” otherwise. CSIZE is the ln of total assets, at
MAIR as a the date of IPO prospectus. CAGE is the ln (1+ number of years between incorporation date and the
dependent variable IPO date). FHORIZON is the number of months between the management forecast date (prospectus
in regression I and date) to end of the period for which the forecast has been conducted. LEV is total liabilities over the
different total assets. According to Belsley et al. (1980), outliers are identified when the absolute values of the
measurements for observations’ studentized residuals are two or greater. To confirm and delete the outliers, this study
experimental conducts the studentized residuals test. Therefore, 24 and 26 IPO companies have been
variables in removed from the samples used in regressions I and II, respectively. *,**,***Significant at the 10, 5
regression II and 1 percent levels, respectively

share on IPO underpricing, the impact of charged fees by the underwriter on IPO
underpricing, the impact of optimistic/pessimistic earnings forecast on underpricing
and finally, the issue of the earnings forecast accuracy level and its relationship with
underpricing. The primary contribution of this study is the examination of the
addressed issues based on Malaysian data.
The sample comprised of 190 IPOs issued in the period 2002-2012. The results
showed a significant decrease in the IPO underpricing of Malaysian companies over
time. This study employed pooled regression equation to model IR. The results show
that IR, on average, are higher for IPO companies having highly relative market share
and less underwriting spread. Further, the management earnings forecast risk-
underpricing relationship is supported. According to the results, a positive and
statistically significant coefficient exists between earnings forecasts bias and
underpricing, which supports the proposition that investors are incapable of predicting
the sign of the earnings forecasts error and employ this prediction in share pricing. Underpricing
Nevertheless, giving conservative forecasts as opposed to optimistic ones does not of Malaysian
help in mitigating the level of underpricing although the benefit of such forecast may IPOs
not be evident in the short-term. For earnings forecasts accuracy, it does decrease the
IPO underpricing. In other words, providing accurate forecasts is beneficial to
forecasters as they are able to be related with less underpricing level.
Our paper has implications to regulators, investors as well as underwriters. Its 367
analysis of the roles of various players (i.e. issuers, underwriters, and investors) and the
related institutional arrangements upon the market may offer some information to policy
makers who are responsible for Malaysian financial markets current reforms. Further, the
examination of the economic outcome of the disclosing of earnings forecasts on
underpricing is important for policy makers and different stock exchanges in their
discussion of costs of litigation and potential modifications in the requirements of
issuing earnings forecasts. For the investors, results may improve their insights of equity
valuation. With regards to the underwriters, it would assist them in identifying
underwriting cost. It is believed that studying the underwriting spread will assist in better
comprehending other issues that are linked to the IPO costs in the Malaysian market.
The findings of this research provides an opportunity for further research to
investigate the effect of underwriter’s market share and spread, and the management
earnings forecasts bias and accuracy, on the IPO performance in the long-term. Future
studies may also evaluate whether or not management earnings forecasts bias and
accuracy influence the potential of companies to carry out subsequent seasoned equity
offerings in the future. Another extension of this research would be to examine the
determinants and characteristics of underwriting market share in-depth along
the spread in the context of the IPO market in Malaysia. Finally, as a limitation of our
study, caution should be applied when generalizing the results of the study since it is
restricted to Malaysian IPO environment and may not be applicable to other
environments such as those that have strong protection for investors.

Notes
1. In this paper, we use underpricing and raw initial returns interchangeably.
2. In this paper, we use underwriter and investment bank interchangeably.
3. The banks are Amlnvestment Bank Berhad, CIMB Investment Bank Berhad, MIMB
Investment Bank Berhad, RHB Investment Bank Berhad and Maybank Investment Bank.
4. When the shares are first listed on the stock exchange, investors are expected to add a
higher premium to the offering price if they know that the forecasted earnings are
excessively pessimistic. This suggests that the price is based on the earnings forecasts. On
the other hand, if investors know that earnings forecasts are excessively optimistic,
subsequently the price of listing can be lower than offering price.

References
Ahmad-Zaluki, N.A. and Kect, L.B. (2012), “The investment performance of MESDAQ market
initial public offerings (IPOs)”, Asian Academy of Management Journal of Accounting
and Finance, Vol. 8 No. 1, pp. 1-23.
Ahmad-Zaluki, N.A. and Wan-Hussin, W.N. (2010), “Corporate governance and earnings
forecasts accuracy”, Asian Review of Accounting, Vol. 18 No. 1, pp. 50-67.
IJMF Ahmad-Zaluki, N.A., Campbell, K. and Goodacre, A. (2007), “The long run share price
12,3 performance of Malaysian initial public offerings (IPOs)”, Journal of Business Finance
& Accounting, Vol. 34 Nos 1-2, pp. 78-110.
Ahn, O.H., Kim, J. and Son, P. (2007), “The pricing of underwriting services in the Korean IPO
market: characteristics and determinants”, Asia-Pacific Journal of Financial Studies, Vol.
36 No. 5, pp. 731-764.
368 Bartling, B.R. and Park, A. (2009), “What determines the level of IPO gross spreads?
Underwriter profits and the cost of going public”, International Review of Economics &
Finance, Vol. 18 No. 1, pp. 81-109.
Bates, T. and Dunbar, C.G. (2002), “Investment bank reputation, market power, and the pricing
and performance of IPOs”, working paper, University of Western Ontario, London and
Ontario.
Beatty, A.L., Ke, B. and Petroni, K.B. (2002), “Earnings management to avoid earnings
declines across publicly and privately held banks”, Accounting Review, Vol. 77 No. 3,
pp. 547-570.
Beatty, R.P. and Ritter, J.R. (1986), “Investment banking, reputation, and the underpricing of
initial public offerings”, Journal of Financial Economics, Vol. 15 No. 1, pp. 213-232.
Beatty, R.P. and Welch, I. (1996), “Issuer expenses and legal liability in initial public offerings”,
Journal of Law and Economics, Vol. 39 No. 2, pp. 545-602.
Belsley, D.A., Kuh, E. and Welsch, R.E. (1980), Regression Diagnostics: Identifying Influential
Data and Sources of Collinearity, John Wiley & Sons, New York, NY.
Benveniste, L.M. and Spindt, P.A. (1989), “How investment bankers determine the offer price
and allocation of new issues”, Journal of Financial Economics, Vol. 24 No. 2, pp. 343-
361.
Boulton, T.J., Smart, S.B. and Zutter, C. (2011), “Earnings quality and international IPO
underpricing”, Accounting Review, Vol. 86 No. 2, pp. 483-505.
Carter, R.B. and Dark, F.H. (1992), “An empirical examination of investment banking
reputation measures”, Financial Review, Vol. 27 No. 3, pp. 355-374.
Carter, R.B. and Manaster, S. (1990), “Initial public offerings and underwriter reputation”,
Journal of Finance, Vol. 45 No. 4, pp. 1045-1067.
Carter, R.B., Dark, F.H. and Singh, A. (1998), “Underwriter reputation, initial returns, and the
long- run performance of IPO stocks”, Journal of Finance, Vol. 35 No. 1, pp. 285-311.
Chambers, D. and Dimson, E. (2009), “IPO underpricing over the very long run”, Journal of
Finance, Vol. 64 No. 3, pp. 1407-1443.
Chemmanur, T.J. and Fulghieri, P. (1994), “Investment bank reputation, information
production, and financial intermediation”, Journal of Finance, Vol. 49 No. 1, pp. 57-79.
Chen, C.R. and Mohan, N.J. (2002), “Underwriter spread, underwriter reputation, and IPO
underpricing: a simultaneous equation analysis”, Journal of Business Finance &
Accounting, Vol. 29 Nos 3-4, pp. 521-540.
Chen, G., Firth, M. and Krishnan, G.V. (2001), “Earnings forecast errors in IPO prospectuses
and their associations with initial stock returns”, Journal of Multinational Financial
Management, Vol. 11 No. 2, pp. 225-240.
Chen, H.C., Fok, R.C. and Wang, Y.J. (2006), “Why do underwriters charge low underwriting
fees for initial public offerings in Taiwan?”, Journal of Business Finance & Accounting,
Vol. 33 Nos 7-8, pp. 979-1005.
Clarkson, P.M., Dontoh, A., Richardson, G. and Sefcik, S. (1992), “The voluntary inclusion of
earnings forecasts in IPO prospectuses”, Contemporary Accounting Research, Vol. 8 No.
2, pp. 601-616.
Cooney, J., Singh, A. and Carter, R.B. (2001), “IPO initial returns and underwriter reputation:
has the inverse relation flipped in the 1990s”, working paper, University of Kentucky, Underpricing
Lexington, KY. of Malaysian
Dawson, S.M. (1987), “Secondary stock market performance of initial public offers, Hong IPOs
Kong, Singapore and Malaysia: 1978-1984”, Journal of Business Finance & Accounting,
Vol. 14 No. 1, pp. 65-76.
Dunbar, C.G. (2000), “Factors affecting investment bank initial public offering market share”,
Journal of Financial Economics, Vol. 55 No. 1, pp. 3-41. 369
Firth, M. (1997), “An analysis of the stock market performance of new issues in New Zealand”,
Pacific-Basin Finance Journal, Vol. 5 No. 1, pp. 63-85.
Firth, M. (1998), “IPO profit forecasts and their role in signalling firm value and explaining
post- listing returns”, Applied Financial Economics, Vol. 8 No. 1, pp. 29-39.
Firth, M., Gounopoulos, D. and Pulm, J. (2012), “IFRS adoption and management earnings
forecasts of Australian IPOs”, working paper, University of Barcelona, Barcelona.
Gounopoulos, D. (2011), “Associations between management forecast accuracy and pricing
of IPOs in Athens stock exchange”, Multinational Finance Journal, Vol. 15 Nos 3-4,
pp. 235-272.
Henry, D., Ahmed, K. and Riddell, A. (2002), “The effect of IPO prospectus earnings
forecast errors on shareholder returns”, Journal of Corporate Communications, Vol. 4,
pp. 1-27.
How, J.C. (1996), “Voluntary forecast disclosure and the underpricing of IPOs”, Accounting
Research Journal, Vol. 9 No. 1, pp. 17-28.
How, J.C. and Yeo, J.J. (2001), “The impact of forecast disclosure and accuracy on equity
pricing: the IPO perspective”, Journal of Accounting, Auditing & Finance, Vol. 16 No. 4,
pp. 401-425.
Huang, W. (2012), “Underwriter reputation, IPO underpricing and underwriting spreads:
evidence from Chinese stocks market”, available at: http://papers.ssrn.com/sol3/pap
(accessed November 12, 2013).
Hussin, B.M., Sulong, F. and Osman, S. (2004), “The effect of forecast earnings errors in IPO
prospectus on shareholders return”, paper presented at the International Business
Management Conference, Kuala Lumpur, available at:
http://dspace.uniten.edu.my/xmlui/ handle/123456789/600 (accessed November 15,
2013).
Jelic, R., Saadouni, B. and Briston, R. (1998), “The accuracy of earnings forecasts in IPO
prospectuses on the Kuala Lumpur Stock Exchange”, Accounting and Business Research,
Vol. 29 No. 1, pp. 57-72.
Jelic, R., Saadouni, B. and Briston, R. (2001), “Performance of Malaysian IPOs: underwriters
reputation and management earnings forecasts”, Pacific-Basin Finance Journal, Vol. 9
No. 5, pp. 457-486.
Jenkinson, T. and Ljungqvist, A. (2001), Going Public: The Theory and Evidence on How
Companies Raise Equity Finance, Clarendon Press, Oxford.
Jog, V. and McConomy, B.J. (2003), “Voluntary disclosure of management earnings forecasts
in IPO prospectuses”, Journal of Business Finance & Accounting, Vol. 30 Nos 1-2, pp.
125-168.
Jones, T.L. and Swaleheen, M. (2010), “Endogenous examination of underwriter reputation and
IPO returns”, Managerial Finance, Vol. 36 No. 4, pp. 284-293.
Karamanou, I. and Vafeas, N. (2005), “The association between corporate boards, audit
committees, and management earnings forecasts: an empirical analysis”, Journal of
Accounting Research, Vol. 43 No. 3, pp. 453-486.
IJMF Keasey, K. and McGuinness, P. (1991), “Prospectus earnings forecasts and the pricing of new
12,3 issues on the unlisted securities market”, Accounting and Business Research, Vol. 21 No.
82, pp. 133-145.
Kenourgios, D.F., Papathanasiou, S. and Melas, E.D. (2007), “Initial performance of Greek
IPOs, underwriter’s reputation and oversubscription”, Managerial Finance, Vol. 33 No.
5, pp. 332-343.
370 Kirkulak, B. and Davis, C. (2005), “Underwriter reputation and underpricing: evidence from the
Japanese IPO market”, Pacific-Basin Finance Journal, Vol. 13 No. 4, pp. 451-470.
Klein, A. (1996), “Can investors use the prospectus to price initial public offerings?”, Journal
of Financial Statement Analysis, Vol. 2 No. 1, pp. 23-39.
Ljungqvist, A. (2007), “IPO underpricing”, in Eckbo, B.E. (Ed.), Handbook of Corporate
Finance: Empirical Corporate Finance, Elsevier, North Holland, pp. 375-422.
Logue, D., Rogalski, R.J., Seward, J.K. and Foster-Johnson, L. (2002), “What is special about
the roles of underwriter reputation and market activities in initial public offerings?”,
Journal of Business, Vol. 75 No. 2, pp. 213-243.
Lonkani, R. and Firth, M. (2005), “The accuracy of IPO earnings forecasts in Thailand and their
relationships with stock market valuation”, Accounting and Business Research, Vol. 35
No. 3, pp. 269-286.
Loughran, T. and Ritter, J.R. (2004), “Why has IPO underpricing changed over time?”,
Financial Management, Vol. 33 No. 3, pp. 5-37.
Loughran, T., Ritter, J.R. and Rydqvist, K. (1994), “Initial public offerings: international insights”,
Pacific-Basin Finance Journal, Vol. 2 No. 2, pp. 165-199.
McDonald, J.G. and Fisher, A.K. (1972), “New-issue stock price behavior”, Journal of Finance,
Vol. 27 No. 1, pp. 97-102.
McGuinness, P.B. (2005), “A re-examination of the disclosure of discretionary earnings
forecasts: the case of initial public offerings in Hong Kong”, Applied Financial
Economics, Vol. 15 No. 16, pp. 1107-1123.
Megginson, W.L. and Weiss, K.A. (1991), “Venture capitalist certification in initial public
offerings”, Journal of Finance, Vol. 46 No. 3, pp. 879-903.
Mohamad, S., Nassir, A., Kuing, T.K. and Ariff, M. (1994), “The accuracy of profit forecasts of
Malaysian IPOs”, Capital Market Review, Vol. 2 No. 2, pp. 46-69.
Murugesu, J. and Santhapparaj, S. (2009), “Valuation errors and the initial price efficiency of
the Malaysian IPO market”, IUP Journal of Applied Finance, Vol. 15 No. 10, pp. 19-38.
Prasad, D., Vozikis, G.S. and Ariff, M. (2006), “Government public policy, regulatory
intervention, and their impact on IPO underpricing: the case of Malaysian IPOs”, Journal
of Small Business Management, Vol. 44 No. 1, pp. 81-98.
Ritter, J.R. (1984), “The ‘hot issue’ market of 1980”, Journal of Business, Vol. 57 No. 2, pp.
215-240. Ritter, J.R. and Welch, I. (2002), “A review of IPO activity, pricing, and allocations”,
Journal of
Finance, Vol. 57 No. 4, pp. 1795-1828.
Rock, K. (1986), “Why new issues are underpriced”, Journal of Financial Economics, Vol. 15
No. 1, pp. 187-212.
Shi, C., Pukthuanthong, K. and Walker, T. (2012), “Does disclosure regulation work? Evidence
from international IPO markets”, Contemporary Accounting Research, Vol. 30 No. 1,
pp. 356-387.
Sun, J. and Liu, G. (2009), “The impact of the CSRC Regulation No. 12-1996 on the credibility
of Chinese IPO earnings forecasts”, Global Finance Journal, Vol. 20 No. 2, pp. 165-179.
Titman, S. and Trueman, B. (1986), “Information quality and the valuation of new issues”,
Journal of Accounting and Economics, Vol. 8 No. 2, pp. 159-172. Underpricing
Uddin, M. (2008), “An empirical examination of intended and unintended IPO underpricing in of Malaysian
Singapore and Malaysia”, International Research Journal of Finance and Economics, IPOs
Vol. 20, pp. 55-74.
Yong, O. and Isa, Z. (2003), “Initial performance of new issues of shares in Malaysia”, Applied
Economics, Vol. 35 No. 8, pp. 919-930.
371
About the authors
Mohammed Abdullah Ammer has a PhD in Finance from the Othman Yeop Abdullah Graduate
School of Business, College of Business, Universiti Utara Malaysia, Malaysia, an MSc in
Finance from the Othman Yeop Abdullah Graduate School of Business, College of Business,
Universiti Utara Malaysia, Malaysia and a BBA from the Sana’a University, Yemen. His areas
of expertise include initial public offerings (IPOs), corporate finance, investment, financial
disclosure and management earnings forecasts. Mohammed Abdullah Ammer is the
corresponding author and can be contacted at: mohd7776@yahoo.com
Nurwati A. Ahmad-Zaluki is an Associate Professor at the Universiti Utara Malaysia.
She received her PhD in Accounting and Finance at the University of Stirling, Scotland, in
2006. She has published in leading accounting and finance journals, including the Journal of
Business Finance & Accounting, International Journal of Accounting and Asian Review of
Accounting. Her articles also appear in professional journals such as Accountants Today and
Banker’s Journal.

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com