Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
Initial Public Offering Valuation Dynamics;
Evidence from Pakistani Capital Market
By
Abdul Rasheed
CIIT/FA11-PMS-005/ISB
PhD Thesis
In
Management Sciences
COMSATS University Islamabad
Islamabad - Pakistan
Spring, 2018
II
COMSATS University Islamabad
Initial Public Offering Valuation Dynamics;
Evidence from Pakistani Capital Market
A Thesis Presented to
COMSATS University Islamabad
In partial fulfillment
of the requirement for the degree of
PhD (Management Sciences)
By
Abdul Rasheed
CIIT/FA11-PMS-005/ISB
Spring, 2018
III
Initial Public Offering Valuation Dynamics;
Evidence from Pakistani Capital Market
A Post Graduate Thesis submitted to the Department of Management Sciences
as partial fulfilment of the requirement for the award of Degree of Ph.D. in
Management Sciences.
Supervisor
Dr. Muhammad Khalid Sohail
Assistant Professor, Department of Management Sciences
COMSATS University Islamabad (CUI)
(Islamabad Campus)
Name Registration Number
Abdul Rasheed CIIT/FA11-PMS-005/ISB
IV
Certificate of Approval
This is to certify that the research work presented in this thesis, entitled “Initial Public
Offering Valuation Dynamics; Evidence from Pakistani Capital Market” was
conducted by Mr. Abdul Rasheed, CIIT/FA11-PMS-005/ISB, under the supervision
of Dr. Muhammad Khalid Sohail. No part of this thesis has been submitted anywhere
else for any other degree. This thesis is submitted to the Department of Management
Sciences, COMSATS University Islamabad, in the partial fulfillment of the
requirement for the degree of Doctor of Philosophy in the field of Management
Sciences.
Student Name: Abdul Rasheed Signature: __________________
Examinations Committee:
Signature:___________________________
External Examiner 1:
Prof. Dr. Usman Mustafa
Director
Project Evaluation and Training Division,
Pakistan Institute of Development
Economics (PIDE),
Quid-i-Azam University Campus,
Islamabad
Signature:__________________________.
External Examiner 2:
Dr. Ashfaq Ahmad
Associate Professor
Hailey College of Commerce,
University of the Punjab, Lahore
.
Dr. Muhammad Khalid Sohail
Supervisor
Department of Management Sciences,
COMSATS University Islamabad,
Islamabad
.
Dr. Khalid Riaz
Dean,
Faculty of Business Administration,
COMSATS University Islamabad
.
.
V
Author’s Declaration
I, Abdul Rasheed, CIIT/FA11-PMS-005/ISB hereby state that my PhD thesis titled
“Initial Public Offering Valuation Dynamics; Evidence from Pakistani Capital
Market” is my own work and has not been submitted previously by me for taking any
degree from this university i.e. COMSATS University Islamabad or anywhere else in
the country / world.
At any time if my statement is found to be incorrect even after my graduate the
University has the right to withdraw my PhD degree.
Date: ___
________________________
Abdul Rasheed
CIIT/FA11-PMS-005/ISB
VI
Plagiarism Undertaking
I, solemnly declare that research work presented in the thesis title “Initial Public
Offering Valuation Dynamics; Evidence from Pakistani Capital Market” is solely my
work with no significant contribution from any other person. Small contribution / help
wherever taken has been duly acknowledged and that complete thesis has been written
by me.
I understand the zero tolerance policy of the HEC and COMSATS University
Islamabad towards plagiarism. Therefore, I as an author of the above titled thesis
declare that no portion of my thesis has been plagiarized and any material used as
reference is properly referred/cited.
I understand that if I am found guilty of any formal plagiarism in the above titled
thesis even after award of PhD degree, the University reserves the rights to withdraw /
revoke my PhD degree and that HEC and the University has the right to publish my
name on the HEC / University Website on which names of students are placed who
submitted plagiarized thesis.
Date: ______
________________________
Abdul Rasheed
CIIT/FA11-PMS-005/ISB
VII
Certificate
It is certified that Abdul Rasheed, CIIT/FA11-PMS-005/ISB has carried out all the
work related to this thesis under my supervision at the Department of Management
Sciences, COMSATS University Islamabad, Islamabad and the work fulfills the
requirement for award of PhD degree.
Date:
Supervisor:
____________________________ Dr. Muhammad Khalid Sohail
Assistant Professor
Department of Management Sciences
COMSATS University Islamabad,
Islamabad
Head of Department:
__________________________________
Dr. Aneel Salman, Assistant Professor
Department of Management Sciences,
COMSATS University Islamabad,
Islamabad
VIII
DEDICATION
I dedicate my work to my beloved Mother
IX
ACKNOWLEDGEMENTS
First and foremost, I am grateful to ALLAH Almighty who bestowed upon me the
knowledge, skills, courage and strength to complete this thesis. I pay my sincere
gratitude from the core of my heart to Holy Prophet Hazrat Muhammad (Peace Be
Upon Him) who is the sole reason for creation of this universe and is the role model
for the whole mankind.
I wish to pay my sincerest gratitude to my supervisor, Dr. Muhammad Khalid Sohail,
Assistant Professor, Department of Management Science, COMSATS University
Islamabad, who imparted his knowledge, broad experience and positive vision to me.
He has always been a source of inspiration for me since I started my research work. I
would never forget his unflinching readiness to help me out whenever I was in a
trouble, no matter how trivial the trouble was.
I am obliged to Professor Dr. Khalid Riaz, Dean, Faculty of Business Administration
for their guidance and help in research work as well as in administrative matters
whenever I seek. Likewise, I am also grateful to Dr. Aneel Salman, Head of the
Department of Management Sciences and Dr. Zahid Iqbal, former Head of the
department of Management Sciences for their great help in administrative matters.
I owe my heartily gratitude to my sweet mother, Mrs. Zohra (Late) and my lovely
wife, Mahmoona Rasheed, as without their love, cooperation and continuous
encouragement this research work was not possible. They cooperated in every respect
throughout my PhD program right from the beginning till the end. I would like to pay
special thanks to my nephew, Jasim Sarwar for his never ending encouragement and
moral support throughout the PhD program.
.
Abdul Rasheed
CIIT/FA11-PMS-005/ISB
X
ABSTRACT
Initial Public Offering Valuation Dynamics; Evidence from
Pakistani Capital Market
The thesis examines the valuation dynamics of Pakistani Initial Public Offerings
(IPOs): their practices, motivations and implications. This study examines the pre-IPO
valuation dynamics and the post-IPO price performance paradigms using 88 IPOs
floated from 2000 to 2016 on the Pakistan Stock Exchange. The main objectives of
this study includes: (1) to provide insights of preferred valuation methods when
valuing IPOs, (2) to compare the bias and accuracy attached to each valuation
methods, (3) to provide the usefulness of prospectus information on the initial
valuations, the underpricing and the long-run underperformance, and finally, (4) to
validate the long-run underperformance using calendar-time approaches.
The binary logit model, the signed predictcion errors (SPE) and the absolute
prediction errors (APE) were used to explain the choice, bias and accuracy attached to
each valuation methods respectively. The accounting-based valuation model was used
to estimate the impact of fundamental, risk and signaling factors on the post-IPO
performance. The capital asset pricing model (CAPM), Fama-French three- (FF3F)
and five-factor (FF5F) models were used as robust measures to affirm the long-run
underperformance anomaly.
The findings document that the Pakistani underwriters repeatedly used dividend
discount model (DDM), discounted cash flow (DCF) model and the comparable
multiples valuation methods when valuing IPOs. The findings of SPE reveal that the
DDM and DCF methods seem to be unbiased value estimators than the comparable
multiples. The findings of APE document that the DCF produce more valuation
accuracy than the other valuation methods.
The average underpricing of 32.85% was observed in the Pakistani primary market.
This research extends the underpricing analysis in various aspects such as: (1) the
level of underpricing was negatively related to the firm size, (2) the underpricing of
IPOs issued in the hot-issue market was significantly higher than the IPOs issued in
the cold-issue market, (3) the underpricing of IPOs issued through bookbuilding was
lower than the IPOs issued through the fixed price auction, (4) the underpricing of
privatization IPOs was higher than the underpricing of non-privatization IPOs, (5) the
underpricing of survivor IPOs was higher than the underpricing of non-survivor IPOs,
XI
(6) the IPOs offered in the Oil & Gas and Chemicals sectors produce more
underpricing than the other sectors. The finding of initial excess returns (IER)
regression analysis reveals that the earnings, financial leverage, efficiency risk, firm
beta and the underwriter reputation were the key determinants to explain variation in
the level of underpricing.
In the long-run returns (LRR) analysis, the buy and hold abnormal returns (BHAR)
produce negative returns of -23.52% and -65.22% in year 3 and year 5 respectively.
On the similar pattern, the cumulative abnormal returns (CAR) produce negative
returns of -24.62% and -29.37% in year 3 and year 5 respectively. This study extend
the long-run performance analysis in various aspects such as: (1) the IPOs issued in
hot-issue market produce more negative returns than the IPOs issued during cold-
issue market, (2) the Automobile & Electrical Goods sector IPOs produce worst
negative returns, while the Modaraba & Foods sector IPOs outperform the market in
the long run, (3) the privatization IPOs outperform the market in the long run than the
non-privatization IPOs. The finding of LRR regression analysis reveals that the book
value of shareholder’s equity, earnings, capital availability risk, firm beta, underwriter
reputation, the percentage of shares offered and initial excess returns were the
significant determinants that explain the variation in the long-run returns.In the
Calendar-time approach, the negative values of intercepts of CAPM, FF3F & FF5F
validate the negative performance in the long run. The market risk premium was the
most significant determinant in all asset pricing models, while HML-value factor (in
equally-weighted FF5F) and CMA-investment factor (in value-weighted FF5F) were
also significant determinants in the Fama-French five-factor models.
This study is one of the few studies in IPO valuation literature that is being
accomplished in a growing and transforming from loose regulated capital market to
synchronize the state of affairs and first in Pakistan to investigate the explanatory
power of prospectus information on IPO valuation dynamics.
Keywords: Pakistan Stock Exchange, Initial Public Offerings, Valuation Methods,
Asset Pricing Models, Event- & Calendar-time Approaches
XII
TABLE OF CONTENTS 1. Introduction ................................................................................................... 1
1.1 Background of the study ........................................................................... 2
1.2 Overview of IPOs in the Pakistan Stock Exchange .................................. 4
1.2.1 The Pakistan Stock Exchange ........................................................... 4
1.2.2 The History of Karachi Stock Exchange ........................................... 5
1.2.3 The Performance of Karachi Stock Exchange .................................. 6
1.2.4 The History of Initial Public Offerings in Pakistan ........................... 8
1.2.5 The Listing Procedure of IPOs in Pakistan Stock Exchange ............ 9
1.3 The Issues Related to IPO Valuations and Pricing ................................. 12
1.4 The Issues Related with IPO aftermarket Price Performance ................. 13
1.5 Problem Statement .................................................................................. 15
1.6 Research Objectives and Questions ........................................................ 16
1.6.1 Research Objectives ........................................................................ 16
1.6.2 Research Questions.......................................................................... 16
1.7 Significance of the Study ........................................................................ 17
1.8 Structure of Dissertation ......................................................................... 19
2. Literature Review ........................................................................................ 20
2.1 IPO Valuation ......................................................................................... 21
2.1.1 Post-IPO Valuation Methods based studies .................................... 21
2.1.2 Pre IPO Valuation Methods based studies ...................................... 28
2.2 The IPO Underpricing Phenomenon ....................................................... 33
2.2.1 The winner’s Curse Hypothesis ....................................................... 34
2.2.2 The Prestigious Underwriter Hypothesis......................................... 35
2.2.3 The Signaling Hypothesis ................................................................ 37
2.2.4 The Lawsuit Avoidance Hypothesis ................................................ 40
2.2.5 The Prospect theory ......................................................................... 43
2.2.6 An International Empirical Evidence .............................................. 44
2.2.7 Underpricing in Pakistan ................................................................. 49
2.3 The IPO Long-run performance .............................................................. 52
2.3.1 Fad Hypothesis ................................................................................ 53
XIII
2.3.2 Heterogeneous expectations hypothesis .......................................... 54
2.3.3 Agency Hypothesis .......................................................................... 55
2.3.4 Signaling Hypothesis ....................................................................... 56
2.3.5 An International Empirical Evidence .............................................. 57
2.3.6 Difficulties with Long-term Returns Measurement ........................ 63
2.3.7 Long-run IPOs Performance in Pakistan ......................................... 65
3. Research Methodology ................................................................................ 68
3.1 Data and Sample Description .................................................................. 69
3.2 Theoretical Background .......................................................................... 72
3.3 The Choice, Bias and Accuracy of Valuation Methods .......................... 73
3.3.1 Enlightening the choice of valuation methods ................................ 73
3.3.2 Bias and Accuracy of Valuation Methods ....................................... 78
3.4 The Basic Valuation Model .................................................................... 83
3.4.1 The IPO Valuation Model ............................................................... 85
3.4.2 Hypotheses about the Valuation Model .......................................... 92
3.5 The Initial Excess Return Model ............................................................ 94
3.5.1 Hypotheses about the Initial Excess Returns Model ....................... 97
3.6 The Long-run Performance Model ......................................................... 99
3.6.1 Hypotheses about the Long Run Returns Model ........................... 103
3.6.2 CAPM, Fama-French Three- and Five-Factor Models ................. 106
3.6.2.1 FF 3-Factor Variables construction ................................................. 107
3.6.2.2 FF 5-Factor Variables construction ................................................. 108
4. Results and Discussion .............................................................................. 110
4.1 The Choice, Bias and Accuracy of Valuation Methods ........................ 111
4.1.1 Explaining the Choice of Valuation Methods ............................... 111
4.1.2 Explaining the Bias of Valuation Methods ................................... 126
4.1.3 Explaining the Accuracy of Valuation Methods ........................... 131
4.2 The IPOs Initial Prices (Valuation) Analysis........................................ 138
4.2.1 Descriptive Statistics ..................................................................... 140
4.2.2 The Univariate Analysis ................................................................ 143
4.2.3 Accounting Based Valuation Models Analysis ............................. 146
XIV
4.3 The IPOs Initial Excess Returns Analysis ............................................ 155
4.3.1 Descriptive Statistics of IER ......................................................... 156
4.3.2 The Univariate Analysis ................................................................ 163
4.3.3 The Analysis of Initial Excess Returns Models ............................ 165
4.3.4 The Sensitivity Analysis ................................................................ 170
4.4 The IPOs Long-run Returns Analysis ................................................... 172
4.4.1 Descriptive Statistics of LRR ........................................................ 173
4.4.2 The Univariate Analysis ................................................................ 183
4.4.3 The Analysis of Long-run Returns (LRR) Models ....................... 186
4.5 IPOs Long-run Performance Using Calendar-Time Approach ............ 193
4.5.1 Descriptive Statistics ..................................................................... 193
4.5.2 The Univariate Analysis ................................................................ 194
4.5.3 The Analysis of Asset Pricing Models .......................................... 196
5. Summary and Conclusion ......................................................................... 200
5.1 The Analysis of Pre-IPO Valuation Dynamics ..................................... 202
5.2 The Analysis of Post-IPO Price Performance ....................................... 203
5.3 Policy Implications of the Study ........................................................... 206
5.3.1 The Pakistan Stock Exchange ....................................................... 207
5.3.2 The Investment Banks/Underwriters ............................................. 207
5.3.3 The Unlisted/Potential IPO-issuing Firms..................................... 208
5.3.4 The Investment Community .......................................................... 209
5.4 Limitations of the Study ........................................................................ 209
5.5 Suggestions for Future Research .......................................................... 210
References ...................................................................................................... 212
Appendix ........................................................................................................ 242
XV
LIST OF TABLES___________________________________________
Table 1. 1: The Performance of KSE-100 Index during 1995-2016 .............................. 7
Table 1. 2: Shares Offered and Funds Raised during 1995-2016 .................................. 9
Table 1. 3: Step by Step IPO Listing Procedure in Pakistan Stock Exchange ............. 11
Table 2. 1: Post IPO valuation studies under different comparable benchmarks ........ 25
Table 2. 2: IPOs Initial Excess Returns Studies From International Literature ........... 47
Table 2. 3: IPOs Initial Excess Returns Studies From Pakistani Literature ................ 50
Table 2. 4: IPOs Long-run Performance Studies From International Literature ......... 60
Table 2. 5: IPOs Long-run performance Studies using the Event-Time Approach ..... 61
Table 2. 6: IPOs Long-run performance Studies using Calendar-Time Approach...... 65
Table 2. 7: IPOs Long-run Performance Studies From Pakistani Literature ............... 66
Table 3. 1: Sample Selection Criteria and Description ................................................ 70
Table 3. 2: Sector-wise IPO firms in the sample ......................................................... 71
Table 3. 3: Operational definitions of variables used in Equation (1), (6) and (7) ...... 82
Table 3. 4: Operational definitions of variables used in IPO valuation model ............ 91
Table 4. 1: Descriptive Statistics of IPO Firm’s characteristics ................................ 112
Table 4. 2: Correlation Matrix of Variables used in Binary Logit & Cross-sectional
Models........................................................................................................................ 115
Table 4. 3: The Summary of Valuation Methods disclosed in prospectus ................ 119
Table 4. 4: Results of Binary Logit of Preferred Valuation Methods ........................ 121
Table 4.5: Analysis of Signed Prediction Errors (at 1st Day Closing Prices) ........... 127
Table 4. 6: Analysis of Signed Prediction Errors (at IPO Offer Prices) .................... 128
Table 4. 7: Cross-sectional Regressions of Bias of Valuation Methods .................... 130
Table 4. 8: Analysis of Absolute Prediction Errors (at 1st Day Closing Prices) ....... 132
Table 4. 9: Analysis of Absolute Prediction Errors (at IPO Offer Prices) ................. 133
Table 4. 10: The Analysis of Value Relevancy Through Regressions ...................... 135
Table 4. 11: Cross-sectional Regressions of Accuracy of Valuation Methods.......... 137
Table 4. 12: Descriptive Statistics of Variables used in Performance Models .......... 140
Table 4. 13: Correlation Matrix of Variables used in Valuation and Aftermarket
Performance Models .................................................................................................. 144
Table 4. 14: Empirical Findings of Basic Valuation Models..................................... 147
Table 4. 15: Cross-sectional Analysis of Valuation Models using Full Sample ....... 150
XVI
Table 4. 16: Cross-sectional Analysis of Valuation Models using non-PIPO Sample
.................................................................................................................................... 153
Table 4. 17: Descriptive Statistics of Initial Excess Returns Analysis ...................... 156
Table 4. 18: Descriptive Statistics of IERs in different Issue Proceeds ..................... 157
Table 4. 19: Descriptive Statistics of IERs in Hot- & Cold-issue Periods ................ 159
Table 4. 20: Descriptive Statistics of IERs of various sub-samples .......................... 160
Table 4. 21: Year-wise Initial Excess Returns Analysis ............................................ 161
Table 4. 22: Sector-wise Initial Excess Returns Analysis ......................................... 162
Table 4. 23: Correlation Matrix of Variables used in Initial Excess Returns Analysis
.................................................................................................................................... 164
Table 4. 24: Regression Analysis of IER Models using Full IPO Sample ................ 166
Table 4. 25: Regression Analysis of IER Models using Non-PIPO Sample ............. 171
Table 4. 26: Descriptive Statistics of equally-weighted Long-run Returns ............... 173
Table 4. 27: Descriptive Statistics of value-weighted Long-run Returns .................. 173
Table 4. 28: Year-wise Long-run Returns Analysis using BHAR and CAR............. 174
Table 4. 29: Year-wise Long-run Returns Analysis using BHAR ............................. 175
Table 4. 30: Year-wise Long-run Returns Analysis using CAR ............................... 176
Table 4. 31: Sector-wise Long-run Returns Analysis using BHAR .......................... 177
Table 4. 32: Sector-wise Long-run Returns Analysis using CAR ............................. 177
Table 4. 33: Privatization and Non-Privatization IPOs Long-run Returns Analysis . 179
Table 4. 34: Firm’s Size-wise Long-run Returns Analysis using BHAR .................. 180
Table 4. 35: Firm’s Size-wise Long-run Returns Analysis using CAR ..................... 180
Table 4. 36: Initial Returns-wise Long-run Returns Analysis using BHAR ............. 181
Table 4. 37: Initial Returns-wise Long-run Returns Analysis using CAR ................ 181
Table 4. 38: Financial and Non-Financial IPOs Long-run Returns Analysis ............ 182
Table 4. 39: Correlation Matrix of Variables used in LRR Models .......................... 184
Table 4. 40: Regression Analysis of LRR Models using Full Sample ...................... 187
Table 4. 41: Regression Analysis of LRR Models using Non-PIPO Sample ............ 191
Table 4. 42: Descriptive Statistics of Variables Used in Asset Pricing Models ........ 194
Table 4. 43: Correlation Matrix of Variables Used in Asset Pricing Models ............ 195
Table 4. 44: Regression Analysis of Capital Asset Pricing Models .......................... 196
Table 4. 45: Regression Analysis of Fama-French Three Factor (FF3F) Models ..... 197
Table 4. 46: Regression Analysis of Fama-French Five Factor (FF5F) Models ....... 198
Table A. 1: Characteristics of IPO Firms used in Sample ......................................... 243
XVII
Table A. 2: The List of Non-Survivor IPO Firms During Sample Period ................. 245
Table A. 3: The List of State Owned Enterprize (SOE) IPO Firms .......................... 245
Table A. 4: Firms Post-IPO Cash and Stock Dividends (%) Announcements History
.................................................................................................................................... 246
Table A. 5: Cross-sectional Regressions of Bias of Valuation Methods ................... 249
Table A. 6: Value Relevance Regressions (at IPO Offer prices) ............................... 250
Table A. 7: Cross-sectional Regressions of Accuracy of Valuation Methods (IPO
Offer Prices) ............................................................................................................... 251
Table A. 8: Empirical findings of basic valuation models......................................... 252
Table A. 9: Cross-sectional Analysis of Full Sample Valuation Models .................. 253
Table A. 10: Equally and Value-weighted Monthly Returns using BHAR & CAR . 254
Table A. 11: Correlation Matrix of Variables used in LRR Models ......................... 259
Table A. 12: Regression Analysis of LRR Models using Full Sample ..................... 261
Table A. 13: Regression Analysis of LRR Models using Non-PIPO Sample ........... 262
Table A. 14: Descriptive Statistics of Variables Used in Value-weighted Analysis . 263
Table A. 15: Correlation matrix of variable used in value-weighted analysis........... 264
Table A. 16: Year-wise New-Listings and De-Listings in PSX ................................ 265
XVIII
LIST OF FIGURES__________________________________________
Figure 1. 1: Regulatory and Operational Structure of Pakistan Stock Exchange .......... 5
Figure 1. 2: The Performance of KSE-100 Index during 1995-2016 ............................ 7
Figure 1. 3: Shares Offered and Funds Raised during 1995-2016 ................................. 8
Figure 3. 1: Year-wise Number of IPOs in the Sample ............................................... 70
Figure 3. 2: Sector-wise Number of IPOs in the Sample ............................................. 71
Figure 4. 1: IPO Valuation and Pricing process involve in New Offerings .............. 118
Figure 4. 2: Aftermarket IPO Valuation and Performance Analysis ......................... 139
Figure 4. 3: Yearly Listed IPOs and IER in Hot-Cold Issue Market ......................... 158
Figure 4. 4: Sector-wise Initial Excess Returns Analysis .......................................... 162
Figure 4. 5: Year-wise Long-run Returns Analysis using BHAR and CAR ............. 175
Figure 4. 6: Year-wise Long-run Returns using BHAR and CAR ............................ 176
Figure 4. 7: Sector-wise Long-run returns Analysis using BHAR and CAR ............ 178
Figure 4. 8: Privatization and Non-Privatization IPOs Long-run Returns Analysis.. 179
Figure 4. 9: Long-run Performance Analysis of Financial and Non-Financial IPOs 182
Figure A. 1: Equal- and Value-weighted Monthly returns using BHAR and CAR .. 255
Figure A. 2: Year-wise Long run returns for year 1 using BHAR and CAR ............ 256
Figure A. 3: Year-wise Long run returns for year 2 using BHAR and CAR ............ 256
Figure A. 4: Year-wise Long run returns for year 3 using BHAR and CAR ............ 256
Figure A. 5: Year-wise Long run returns for year 4 using BHAR and CAR ............ 257
Figure A. 6: Year-wise Long run returns for year 5 using BHAR and CAR ............ 257
Figure A. 7: Sector-wise Long run returns for year 1, 3 & 5 using CAR .................. 258
Figure A. 8: Sector-wise Long run returns for year 1, 3 & 5 using BHAR ............... 258
Figure A. 9: The cumulative effect of New-Listings & De-Listings in PSX ............ 265
XIX
List of Abbreviations
ADB Asian Development Bank
AMCs Asset Management Companies
APE Absolute Prediction Errors
BHAR Buy-and-Hold Abnormal Returns
CAPM Capital Asset Pricing Model
CAR Cumulative Abnormal Returns
DCF Discounted Cash Flow
DDM Dividend Discount Model
EW Equally Weighted Returns
FF3F Fama-French Three Factor
FF5F Fama-French Five Factor Model
FTSE The Financial Times Stock Exchange
GOP Government of Pakistan
IER Initial Excess Returns
IPOs Initial Public Offerings
ISE Islamabad Stock Exchange
KSE Karachi Stock Exchange
LRR Long run Returns
LSE Lahore Stock Exchange
MOF Ministry of Finance
MSCI Morgan Stanley Capital International
NCCPL National Clearing Company of Pakistan
PSX Pakistan Stock Exchange
SBP State Bank of Pakistan
SECP Securities and Exchange Commission of Pakistan
SEO Seasoned Equity Offerings
SPE Signed Prediction Errors
VW Value Weighted Returns
1
1. Chapter 1
1. Introduction
2
1.1 Background of the study
Thousands of public limited companies around the globe have offered their
shares to the general public through primary market in the last couple of decades. An
Initial Public Offering (IPO) is a procedure whereby a firm sells its shares first time to
the general public through the fixed price or book-building auction. Firms can finance
their operations and investments through internal sources as well as external sources
of funding. Internal sources mostly refer as earnings and retained earnings, and
external sources mostly refer as debt financing and equity financing. A decision to
raise equity capital through selling stock to the general public is a significant event in
the lifecycle of issuer firm. A number of IPO studies have raised the concerns whether
firms decide to go public mainly to finance their future investments or for other
reasons such as market timing (Colak & Gunay, 2011; Agathee et al., 2012a). Chang
et al., (2012) argue that most firms decide to go public in hot-issue1 market due to the
several IPO stylized facts such as leave less money2 on the table, favorable investor
sentiments 3 , overvalued stock prices and to avoid time-varying-adverse-selection
problem. There are several theoretical explanations found by researchers that why
firms raising equity capital through selling stocks to the general public and/or
institutions such as plans for future growth, dilution of ownership, exit strategy, to
increase liquidity, cheap cost of capital, to increase external monitoring, reduction in
financial leverage, mergers and acquisitions, to increase equity investment, and
corporate control and to reduce agency costs4 etc (Burkart and Lee, 2008; Lyandres et
al, 2007; Kim and Weisbach, 2008; Huyghebaert and Hulle, 2006; Ritter and Welch,
2002; Pagano & Panetta, 1998).
The initial public offering is a complex procedure because market participants
are uncertain about the worth of IPO firms’ equity and unable to predict the market
demand for their shares. Therefore the issuing firm passes on the IPO offer price
decision to underwriters who act as a valuation expert and certify as financial advisor
1 A situation, when the number of new offerings gets listed with great pace than the usual number of
new offerings in a certain time period. 2 Leave less money on the table means underwriters deliberately set offer price discount at the time of
going public to compensate the market liquidity risk for more details see J. Ritter (1984, 1988) and
Ritter, J. R., & Welch, I. (2002). 3 When investors are least concerned about market risk, valuation and willing to pay high prices based
on good current and forecasted macroeconomic factors 4 When company actual performance is less than the potential performance due to conflicts of interests
between managers and owners is called agency cost.
3
to set preliminary offer prices. In practice, the lead underwriters estimate ex-ante fair
value estimates of the issuing firms using several valuation methods. For this purpose,
the lead underwriters employ different valuation methods based on the firm-specific
characteristics, macroeconomic factors, aggregate stock market returns and market
returns volatility before the IPOs to estimate fair values for new offerings.
Roosenboom (2012) argues that it is difficult to access actual IPO valuation process
employed by investment banks to value IPOs because in most of the countries it is not
mandatory to disclose in the prospectus. Bradley et al., (2003) mentioned that the US
underwriters are not allowed to publish predictions about IPO firms’ valuations.
Aggarwal et al., (2009) examine that how valuations of new offerings has changed
over different state of economies. They find that firms having more negative earnings
before IPOs produce higher valuation and firms with more positive earnings have
higher valuation than firms with less positive earnings. Abdulai (2015), Roosenboom
(2012) and Deloof et al., (2009) document that the lead underwriters deliberately
discount the offer price to attract more participation in the bidding process. Their
findings reveal that the prestigious underwriters offered less offer price discount and
vice versa.
It has been observed from the existing literature of corporate finance
particularly in IPO studies that there are two main irregularities identified namely as;
(1) abnormal returns in very few days, and (2) long-run underperformance of IPO
firms. Ibbotson and Jaffe (1975) and Ritter (1988) first-time document the IPO market
fluctuations over time, both the aftermarket IPO performance and the number of firms
went public vary substantially. A number of IPO studies reported the first day
abnormal returns vary over the last couple of decades across the countries (see Table
2.2). Loughran and Ritter (2004) unveil that the level of IPO underpricing in the US
markets is 11.70%, while Jelic and Briston (1999) document that the level of IPO
underpricing in the emerging market of Hungry is 52%. Ritter (2011) points out that
the initial abnormal returns in the emerging markets of India and Malaysia are higher
than the developed markets of the US and UK. Ritter (1991) first-time documents the
long-run underperformance using data of US IPOs. Kooli and Suret (2004) also
confirmed the long-run underperformance in the Canadian IPO product market.
Govindasamy (2010) documented the long-run underperformance using the sample of
229 South Africa IPOs floated during the 1995-2006 period. Bossin and Sentis (2014)
4
report the long-run underperformance in France and Brown (1999) reports negative
returns for the UK. Bossin and Sentis (2014) used both the event-study and calendar-
time approaches to estimate the long-run performance of both clusters (i.e. orphan
IPOs5 and non-orphan IPOs6). They observed poor performance of both types of IPOs
in the long run relative to the market portfolio during the sample period.
1.2 Overview of IPOs in the Pakistan Stock Exchange
This section provides the introduction, types and history of all stock exchanges in
Pakistan. This section also provides the information of IPOs activity in Pakistan since
last couple of decades.
1.2.1 The Pakistan Stock Exchange
The Pakistan Stock Exchange (PSX) was established on January 11, 2016 when the
Government of Pakistan (GOP) decided to merge the Karachi Stock Exchange (KSE)
established in 1947, the Lahore Stock Exchange (LSE) established in 1970 and the
Islamabad Stock Exchange (ISE) established in 1992 under the Corporatization,
Demutualization and Integration Act, 2012 process which initially defined in August
2012 in order to reduce market fragmentation and build a sturdy stance for attracting
strategic partner required for technological expertise. The PSX was converted into a
public limited company with the ownership rights were segregated from the trading
rights. The PSX has three trading floors in Karachi, Lahore and Islamabad while the
establishment is laid down in Karachi to regulate and surveillance the market
activities. In recent years, the Pakistan stock exchange was reclassified as an MSCI
Emerging Market in the 2016-2017 review and FTSE as a Secondary Emerging
Market because PSX delivered above 40% returns in 2012, 2013 and 2014 (see Table
1.1) which reclassified it as the best emerging market awards in each year (source;
The News, July 22, 2016). Figure 1.1 presents the regulatory institutions and capital
market structure in Pakistan.
As on June 30, 2017, there were 560 companies listed on the exchange with 35
sectors, the total listed capital was PKR 1,312 billion and the total market
capitalization was PKR 9,522 billion. The PSX has now seven indices as the KSE-
5 The IPOs without carrying recommendations from the financial analyst during IPOs placement process 6 The IPOs carrying recommendations from the financial analyst during IPOs placement process.
5
ALL Shares Index, KSE-100 Index, KSE-30 Index, PSX-KMI All Shares Index,
KMI-30 Index, Oil & Gas Sector Index and Banking Sector Index (Source: PSX Daily
quotation of June 30, 2017). There are about 400 members which are brokerage
houses of the exchange, from out of 400 members, 21 members are operating as asset
management companies (AMCs). At current, the investors registered on the PSX
capital market includes as 1,886 foreign institutions, 883 local institutional and above
220,000 individual investors (PSX Annual Report 2017).
Figure 1. 1: Regulatory and Operational Structure of Pakistan Stock Exchange
1.2.2 The History of Karachi Stock Exchange
One of the oldest stock exchanges in the South Asia, the Karachi Stock Exchange was
established on September 18, 1947, Karachi, Pakistan. It starts working by five firms
with a paid-up capital of PKR 37 million rupees and launched the manual trading
system generally known as the open-out-cry system. Before 1991, the KSE50 index
was used as a benchmark, while in November, 1991 the KSE100 index was launched
with a base index of 2,000. The KSE was one of the largest stock exchanges in
6
Pakistan before the implementation of demutualization and integration process. The
KSE was declared one of the top 10 stock markets in the world in 2015. According to
Bloomberg, the benchmark of PSX (KSE100 index) is the 3rd best benchmark in the
world since 2009. Since its establishment, over 70 years of age, the KSE has enabled
by helping a wide range of participants from retail investors to foreign institutional
investors, the investor’s community and the firms listed for trading. Specifically, the
KSE engaged helping the large firms and/or new entrepreneurs to generate long-term
funds from the general public by selling their shares.
1.2.3 The Performance of Karachi Stock Exchange
According to Ahmad (2000), the KSE is unpredictable and the most volatile exchange
in the world. However, the KSE has shown hard to believe performance during the
2000-2010 periods and also in subsequent years except in 2001 and 2008 years due to
the US internet bubble crisis and the US Subprime mortgage crisis respectively. When
the market benchmark dropped more than 50% during 2008 due to the US Subprime
mortgage crisis effect, the management of KSE decided to lock the further drop in
benchmark index, which creates another issue of liquidity crunch because investors
were unable to sell their shares anymore for a couple of months. Due to this
irregularity, MSCI reclassify the status of KSE from emerging markets to the frontier
market, which raised another panic for the investors. The KSE-100 Index has
produced 48.9%, 49.4% and 27.2% gains in 2012, 2013 and 2014 years respectively.
Due to this tremendous performance, KSE remains at top 10 markets of the world. In
2016, the KSE was reclassified as an MSCI Emerging Markets and FTSE as a
Secondary Emerging Market due to the consistent performance of PSX and the recent
capital market developments. At present, the PSX is well thought-out a possible
investment opportunity for the investor community. Table 1.1 and Figure 1.2 present
the historical performance of KSE during the 1995-2016 periods.
7
Figure 1. 2: The Performance of KSE-100 Index during 1995-2016
Table 1. 1: The Performance of KSE-100 Index during 1995-2016
Year %age Gain Opening Closing Min Max Std Dev Days
2016 45.681 33,229 47,807 30,565 47,807 4,433.49 248
2015 2.132 32,480 32,816 28,927 36,229 1,193.30 249
2014 27.196 25,609 32,131 25,479 32,149 1,725.02 246
2013 49.427 16,795 25,261 16,108 25,579 2,656.20 247
2012 48.976 11,282 16,905 10,909 16,943 1,522.99 249
2011 -5.613 11,849 11,348 10,842 12,682 429.24 248
2010 28.077 9,438 12,022 9,230 12,031 620.94 248
2009 60.050 5,753 9,387 4,815 9,846 1,354.85 246
2008 -58.337 13,353 5,865 5,865 15,676 2,566.31 218
2007 40.204 10,067 14,077 10,067 14,815 1,204.40 245
2006 5.063 9,652 10,041 8,769 12,274 666.01 241
2005 53.681 6,222 9,557 6,222 10,126 892.42 250
2004 39.066 4,474 6,218 4,474 6,218 321.81 248
2003 65.528 2,701 4,472 2,356 4,604 688.47 248
2002 112.198 1,273 2,701 1,273 2,701 274.31 261
2001 -15.557 1,508 1,273 1,075 1,550 97.39 261
2000 7.004 1,457 1,508 1,276 2,054 202.15 260
1999 49.053 945 1,409 852 1,430 134.87 261
1998 -46.104 1,754 945 766 1,754 318.18 261
1997 30.796 1,341 1,754 1,339 2,068 166.99 261
1996 -10.656 1,501 1,341 1,332 1,854 133.12 262
1995 -28.010 2,085 1,501 1,323 2,085 161.28 258
Source: PSX DataStream
0
10,000
20,000
30,000
40,000
50,000
60,000
Jan
-04
-19
95
Jan
-04
-19
96
Jan
-04
-19
97
Jan
-04
-19
98
Jan
-04
-19
99
Jan
-04
-20
00
Jan
-04
-20
01
Jan
-04
-20
02
Jan
-04
-20
03
Jan
-04
-20
04
Jan
-04
-20
05
Jan
-04
-20
06
Jan
-04
-20
07
Jan
-04
-20
08
Jan
-04
-20
09
Jan
-04
-20
10
Jan
-04
-20
11
Jan
-04
-20
12
Jan
-04
-20
13
Jan
-04
-20
14
Jan
-04
-20
15
Jan
-04
-20
16
KSE 100 Index
KSE 100 Index
8
1.2.4 The History of Initial Public Offerings in Pakistan
From 1990 onwards, the capital market of Pakistan was accessible for foreign
investors under the capital market liberalization reform process. Due to heavy
investments plunge, the hundreds of Pakistani private and state-owned enterprises
decided to go public to raise equity capital for their expansions and debt repayments.
From 1992 to 1997, 272 IPOs were launched in the KSE, however, the highest
number of IPOs (86 firms) were launched in 1992 followed by 1994 (73 IPOs). The
evidence can be observed from Table 1.2, from 2000 to 2016 (the sample period of
this study) 90 IPOs were issued in the Pakistan where the year 2005 and 2007 were
the best years in terms of number of IPOs issued in each year. The performance of
IPOs issued in 1998 and 1999 was disappointing due to the Pakistan Atomic Missile
Tests and the Military Takeover. In addition, the IPOs activity in recent years is also
unsatisfactory as compared with the firms registered at SECP where thousands of
private and public limited firms registered in SECP during the study period,
particularly 3385, 3925, 3960, 4587, 5000 and 6200 in the year of 2011, 2012, 2013,
2014, 2015 and 2016 respectively (Source: the annual reports released by SECP).
Figure 1. 3: Shares Offered and Funds Raised during 1995-2016
The study did not include IPOs listed before 1999 primarily for two reasons: (1) many
scholars have stated the cases of “fly-by-night” capitalist who wear down the
investors' wealth from 1991 to 1997; (2) after 1998, KSE introduced computerized
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
0
5
10
15
20
25
30
35
40
45
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
IPO Activity During 1995-2016
New Listings Funds Raised (million)
9
trading platform with full automation of back-office operations such as electronic
cash settlement and electronic transfer of shares facilities. From 2000 onward,
Securities and Exchange Commission of Pakistan (SECP) introduced more stringent
regulations.
Table 1. 2: Shares Offered and Funds Raised during 1995-2016
Year New Listings Shares Offered Funds Raised
1995 41 569,405,300 837,578,250
1996 30 222,652,500 313,650,000
1997 4 66,250,000 66,250,000
1998 1 9,960,000 9,960,000
1999 0 0 0
2000 3 46,500,000 542,000,000
2001 2 21,250,000 300,000,000
2002 4 87,754,000 877,540,000
2003 3 42,250,000 470,937,500
2004 8 298,606,710 9,287,835,220
2005 14 424,922,000 11,766,795,000
2006 3 91,100,000 1,126,500,000
2007 11 253,535,500 15,988,747,000
2008 9 171,000,000 4,552,500,000
2009 3 151,160,000 1,511,600,000
2010 6 514,333,334 6,758,000,020
2011 4 189,216,000 3,078,664,480
2012 3 50,000,000 500,000,000
2013 1 56,976,000 1,253,472,000
2014 5 213,949,500 6,547,888,500
2015 7 728,377,857 14,442,470,707
2016 4 159,751,000 4,459,437,500
Total 4,368,949,701 84,691,826,177
Source: SECP official website
1.2.5 The Listing Procedure of IPOs in Pakistan Stock Exchange
As discussed earlier, the decision of going public is an important phase of the life
cycle of the issuing firm. IPOs not only allow to raise funds from the primary market
to meet their financial needs but also offer an opportunity to investors to take part in
the growth of issuing firms. In each country, different regulatory institutions are
engaged to control, regulate and monitor the process of new listings. In Pakistan, the
SECP and the PSX are the main regulatory institutions and the step-by-step listing
procedure with respect to expected time involved is reported in Table 1.3.
10
• According to section 3 of The Issue of Capital Rule 1996, the issuing firm can
raise capital through the public offer for the first time which owns a loan-
based project or an equity-based project.
• According to section 4 of The Issue of Capital Rule 1996, In case of shares
offered at a premium, the issuing firm shall have the profitable operational
record of at least one year and justification of premium shall be disclosed in
the prospectus.
• According to Companies Ordinance 1984, the ordinance allows new issuance
to the public only through the approval of prospectus from the stock exchange
and commission. According to section 9 of SEC Ordinance 1996, an issuer
firm who plans to get listed on PSX shall submit an application.
• According to section 5 of PSX Rule Book,
o An issuing firm shall have a post-issue paid-up capital of at least PKR
200 million.
o In case of post-issue paid-up capital of an issuing firm is up to PKR
500 million then the allocation of capital to the general public shall not
be less than 25% of the post-issue paid-up capital.
o In case of post-issue paid-up capital of an issuing firm is above PKR
500 million then the allocation of capital to the general public shall be
at least PKR 125 million or 12.50% of the post-issue paid-up capital,
whichever is higher.
o An issuer may allocate shares up to 5% and 25% of the public offer to
their employees and overseas Pakistani respectively.
Most of the information has been extracted from AKD Securities Limited7, provided
the details of new offerings in Pakistan with respect to expected time required for
each step. The step-by-step listing procedure of new offerings is reported in the Table
1.3.
7 AKD Securities Limited is one of the leading investment bankers in Pakistan and Mr. Jasim Sarwar facilitate the researcher in order to provide detail information and required data for IPO placement related activities.
11
Table 1. 3: Step by Step IPO Listing Procedure in Pakistan Stock Exchange
Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 Week 10 Week 11 Week 12 Week 13 Week 14 Week 15 – Week 17
Step 1
Agreement with Lead
Underwriter/Advisors Step 2
Due diligence of company Drafting of Prospectus Step 3
Financial Feasibility & Model
Drafting of the information
Memorandum Filing of Listing
Application to PSX
(Payment of Listing Fees)
Step 4
Appointment of BR/ LA/
Balloter/ CDS
Eligibility/Shares Registrar
Application to SECP for
Approval of Preliminary
Prospectus
Step 5
Floor Price Setting Circulation and
Publication of Pre.
Prospectus in
newspaper
Step 6
Finalization of Prospectus Clearance of Preliminary
Prospectus by PSX Publication and
Circulation of Final
prospectus
Step 7
Floor Price Approval Marketing/Road
Shows
Balloting in case of
Oversubscription Note:
Approval from Cabinet
Committee on
Privatization incase of a
State-owned Enterprise
Approval for Listing by
SECP BR Book Runner
Book Building
Registration / Dutch
Auction (not required
in case of fixed-price)
Refund of Bids from
BookBuilding Portion
Refund of unsuccessful Bidders
LA Legal Advisor
Transfer of Shares to
investor CDS Accounts
CDS Central Depository System
Strike Price General Public
Subscription
Formal Listing
PSX Pakistan Stock Exchange
SECP Securities & Exchange Commission of Pakistan
Receipt of Funds
Sources: Chapter 5: Listing of Companies and Securities Regulations of PSX
12
1.3 The Issues Related to IPO Valuations and Pricing
Apparently, it has been observed that the discussion regarding selecting
comparable firms for valuation is very little in the emerging economies. As a result,
the practitioners usually employ explanation of comparable firms based on developing
economies when valuing IPOs using multiples in the circumstance of emerging
economies (Goh et al., 2015; Ivashkovskaya & Kuznetsov, 2007). By comparing the
predicted valuation accuracy adjusted by the country risk of the US and Russia,
Ivashkovskaya & Kuznetsov (2007) argued that the price-book (P/B) and enterprise
value-sales (EV/S) multiples produce more valuation accuracy than the price-earnings
(P/E) valuation accuracy. In the beginning, this study hypothesized that the
performance of valuation accuracy of numerous valuation methods in emerging
economies may be varying than the developed economies. Park and Lee (2003)
observed that the Japanese investment banks usually prefer multiples for the
simplicity of IPOs valuation compared to the direct valuation methods. Schreiner and
Spremann (2007) conjectured that the criterion to select comparable firms is
complicated to be recognized because each firm confronts different business issues.
Damodaran (2007) argued that the practitioners unable to implement multiples
correctly because they mostly depend on subjective decisions. Goh et al., (2015)
examined the valuation accuracy of Malaysian IPOs listed in the agribusiness sector.
They examined that the median is a more accurate measure to estimate the
performance of valuation methods and price-earnings multiple outperform than the
other multiple measures. Mills (2005) findings reveal that the discounted cash flow
valuation method is widely used by practitioners. Kaplan and Ruback (1995)
compared the performance of DCF valuations with the valuations achieved from the
comparable firms from similar industries and find that the DCF valuations produced
high valuation accuracy than obtained from the comparables. This study disagree with
the conjecture of Houston et al., (2006), post-IPO valuation estimates are similar as
underwriter’s valuation before the IPOs. Furthermore, an extant literature of IPO
valuations (Kim and Ritter, 1999; Liu et al., 2002; Pumanandam & Swaminathan,
2004; How, Lam & Yeo, 2007; Colaco, Cesari & Hedge, 2013, 2017; Yoon, 2015;
Herawati, Achsani, Hartoyo & Sembel 2017) has mainly focused on the ex-post data
to estimate the fair value of new offerings. Roosenboom (2007) was emphatic about
this lack of attention when noticing that the IPO literature, particularly on valuation is
13
not very thick. In emerging market of Africa, Hearn (2010) mentioned the lack of
attention about the valuation of IPO literature and predicted the accuracy of several
valuation methods. Pereiro (2006) compounded the problems of less attention on IPO
valuation and pricing studies in emerging markets, investment banks usually applying
traditional methods used for valuation in the developed economies. Nwude (2010)
argued that the appropriate methods for Nigerian IPOs valuation and pricing are
problematic because of an extant literature unable to provide clear-cut valuation
methods.
It has been observed from a number of IPO studies, the underwriters did not disclose
their ex-ante valuation procedures in the offering documents (such as Prospectus). In
many advanced economies, researchers used analysts’ reports to estimate their fair
value estimates rather than the actual valuation procedures followed by the lead
underwriters. In the literature, only Abdulai (2015) in Ghana, Goh et al., (2015) in
Malaysia, Roosenboom (2012, 2007) in France and Deloof et al., (2009, 2002) in
Belgium used information of valuation methods disclosed in the prospectus to
examine the choice and valuation accuracy of different valuation methods. The
empirical findings of an extant literature about valuation accuracy based on the ex-
ante fair value estimates disclosed in the prospectuses rather than the ex-post
valuation estimates are not similar.
1.4 The Issues Related with IPO aftermarket Price Performance
The implementation of IPO process has two well-known puzzling stylized
facts across the world, namely as, 1) IPO underpricing (short-run abnormal returns)
phenomenon and 2) long-run underperformance (market adjusted negative returns for
subsequent years after listing) phenomenon. IPO underpricing is the difference
between preliminary offer price of IPO and first day closing price. These initial
returns are significantly positive across the countries but vary in percentages
(Ljungqvist et, al., 2006). Various theories have been proposed to explain the IPO
underpricing such as short run abnormal returns occurs due to agency theory (Jensen
and Meckling, 1976), information asymmetry hypothesis (Rock, 1986; Carter &
Manaster, 1990, Dolvin & Pyles, 2006), signaling theory (Allen & Faulhaber, 1989;
Welch, 1989; Jegaesh et al., 1993), the lawsuit hypothesis (Tinic, 1988; Hughes &
Thakor, 1992; Hao, 2007), prestigious underwriter hypothesis (Gordon & Jin, 1993;
14
Kumar and Tsesekos, 1993), and the ownership dispersion hypothesis (Booth & Chua,
1996). The aforementioned underpricing theories are thoroughly discussed in the
literature review chapter Section 2.2. When there is uncertainty about the 8intrinsic
value of IPO firm then some momentous mispricing is to be expected. Loughran and
Ritter (2002) explain the phenomenon of underpricing as “Leave so much money on
the table” when the market price is higher than the IPO offer price, due to bulky
underpricing, issuer loses wealth than they expected to be. It’s difficult to assess the
exact causes to IPO underpricing.
The IPO long-run underperformance is another anomaly observed in the
domain of public equity aftermarket performance analysis. The long-run poor
performance is explored by various scholars, over many years, across the countries.
Starting from (Aggarwal and Rivoli, 1990; Ritter, 1991; Loughran and Ritter, 1995;
Lowery, 2003), first time uncover the confirmation of negative abnormal returns over
the three and five years after the IPO. They argued that if IPOs are overvalued
systematically then in the long run market adjust this mispricing as long-run
underperformance. They called this underperformance phenomenon is called as ‘Fad’.
They conclude their findings that the long-run underperformance of IPOs reflect in
investor behavioral biases that cause them to be overoptimistic when IPOs are issued
in the hot-issue markets because when more information becomes available, the
market adjusts the initial mispricing in the long run price performance. This argument
is also confirmed by various scholars (Peterle and Berk, 2016; Agathee et al., 2012a;
Colak and Gunay, 2011; Moorman, 2010; Helwege and Liang, 2004; Nanda and Yun,
1997). A number of IPO studies have been trying to discover the reasons of long-run
underperformance under as the heterogeneous expectations hypothesis proposed by
(Miller, 1977; Houge et al., 2001), the agency hypothesis proposed by (Carter et al.,
1998; Johnson and Miller, 1988; Megginson and Weiss, 1991; Carter and Manaster,
1990; Brav and Gompers, 1997), the signaling hypothesis proposed by (Welch, 1989;
Jegadeesh et al., 1993; Ljunqvist, 1996; Koh et al., 1996). The aforementioned long-
run underperformance theories are thoroughly discussed in the literature review
chapter Section 2.3.
8 Intrinsic value is same as market value and it is derived based on current financial position, discount
all business risk factors and future cash-flows.
15
Another implication of long-run underperformance caused by inaccurate
estimation of abnormal returns is discussed by various scholars. This inconsistency
has been investigated in three different ways as: (1) the issue of accuracy and biases
of measurement methods such as event-time approach (BHARs and CARs) and
calendar-time approach (Asset pricing models such as CAPM, Fama-French multi-
factor model, Corhart four-factor model etc) are discussed by (Loughran and Ritter,
2000; Barber and Lyon, 1996; Fama, 1998), (2) the second issue with
underperformance is the selection of market benchmarks to risk-adjusted returns are
discussed by (Kothari and Warner, 1997; Barber and Lyon, 1997; Ritter and Welch,
2002; Gompers and Lerner, 2003; Choi, Lee and Megginson, 2010), and (3) the third
issue with underperformance is the power of statistical tests used for the long-term
performance analysis are discussed by (Loughran and Ritter, 1995; Brav, 2000;
Jenkinson and Ljunqvist, 2001). The issues related to long-run performance
measurement are thoroughly discussed in the literature review chapter Section 2.3.6.
Pakistan economy is effectively in a state of metamorphoses – simultaneously
developing, growing and transforming from loose regulated financial system to
synchronize. The primary markets of developing countries like Pakistan also confront
similar problems of IPO puzzling stylized facts in the process of implementing IPOs
in the capital market. The proposed study aims to examine the IPOs valuation on the
basis of prospectus document which includes all relevant information about the IPO
issuer firms.
1.5 Problem Statement
It is important to take an academic review of the IPOs valuation practice on the PSX,
because in recent years, a seeming lack of confidence is observed by Pakistani
unlisted firms who are eligible to fulfill the prerequisites to issue IPOs but they
intentionally preferred to stay as private firms due to the issues about pre-issue
valuation biases, post-issue ‘mispricing’ on early trading days and inconsistent
policies by the regulators as discussed in the literature. In 1997-98, Asian
Development Bank (ADB) aid $50 million for capital market liberalization and
financial reforms. But disappointingly, since the financial reforms have been
implemented, the capital market is facing the issues of i) the stumpy market depth and
16
size, and ii) the high pace of delisting firms is more than the pace of new IPO
offerings in the market.
Table A.16 & Figure A.9 (see Appendix) present the details of year-wise delisting of
firms and listing IPO firms from 1997 to 2016, which motivated the scholar to
investigate the insights of ex-ante valuation process and ex-post price performance of
Pakistani IPO firms.
1.6 Research Objectives and Questions
The research objectives and research questions are discussed in this section.
1.6.1 Research Objectives
i. To investigate the firm-specific characteristics and market-related factors that had
influenced the choice of valuation methods when valuing IPOs.
ii. To investigate the firm-specific characteristics and market-related factors that had
influenced the valuation biase and accuracy associated with each valuation
method.
iii. To examine the usefulness of prospectus information on the initial valuations to
set preliminary prices in pre-issue pricing process.
iv. To inspect the impact of fundamental, ex ante risk and signaling factors on the
IPOs short-run and long-run adjusted returns.
v. To validate the robustness of IPOs long-run underperformance using asset pricing
models (Calendar-time approach).
1.6.2 Research Questions
i. What firm-specific and market-related factors had influenced the choice of several
valuation methods that are employed when valuing IPOs?
ii. What firm-specific and market-related factors had influenced the valuation biases
and accuracy associated with each valuation method?
iii. Does the prospectus information help to price the IPO firms in the ex-ante pricing
decision process?
iv. Does the prospectus information have an explanatory power on the short-run and
long-run adjusted returns?
17
v. Does the calendar-time approach validate the IPOs long-run underperformance
anomaly?
1.7 Significance of the Study
This study examines the valuation dynamics of Pakistani Initial Public Offerings:
their valuation practices, motivations and implications. Most developed market
challenges are generally augmented in the emerging markets. This study has five main
contributions. First, this study contributes and explains the IPO valuation practice in
Pakistan using the underwriter’s valuation analysis information published in the
prospectus documents. The findings suggest that the Pakistani underwriters frequently
used dividend discount model, discounted cash flow and comparable multiples
valuation approaches to estimate the fair value of issuing firm’s equity. The findings
also suggest that the investment banks choose each valuation method on the basis of
firm-specific characteristics, aggregate stock market returns and aggregate stock
market volatility before the IPO.
Second, based on the underwriter’s valuation analysis information, this study
evaluates the valuation biases and accuracy of each valuation method employed by
lead underwriters when valuing IPOs. The finding suggests that the DCF produce less
valuation bias and high valuation accuracy than the other valuation methods. This
disagree with Purnanandam and Swaminathan (2004), Kim and Ritter (1999), How,
Lam and Yeo (2007), Chang and Tang (2007), Sahoo and Rajib (2013), Yoon (2015),
and Colaco, Cesari and Hegde (2017) who completely rely on comparable multiples
valuations on the basis of post-IPO financial data rather than the real valuation
estimates of underwriters. So, the practitioners can’t completely rely on explanations
from the developed market findings.
Third, this study, first time in Pakistan, used the unique data in terms of, (1) the cash
and stock dividends adjusted prices to accurately measure the short-run and long-run
price performance of IPO firms, (2) for pre-IPO valuation analysis, 88 out of 94 IPOs
were taken as sample, (3) the 70% (65 out of 94) IPOs were used for post-IPO price
performance analysis that only survive at least five years since the date of formal
listings in the market, and (4) 225 non-IPO firms data (that remain listed from 2000 to
2017 in PSX) were used in capital asset pricing model (CAPM), Fama-French 3 factor
18
(FF3F) & 5 factor (FF5F) models to counter estimate the long-run IPO portfolio
performance.
Fourth, most studies analyzed the relationship of IPO valuation and underpricing
phenomenon (Klein, 1996; Beatty et al, 2002, Cynthia J, et al, 2008; Roosenboom,
2007, 2012; Glordano et al, 2011; Huge, 2013) and another form of studies examined
the relationship between initial excess returns and long-run abnormal performance
analysis (Levis, 1993; Ritter, 1991; Lian and Wang, 2009; Xia et al, 2013). This study
examined the three different elevations involved in the implementation of IPO process
as (1) the pre-issue valuation analysis commenced by the lead underwriters, (2) the
initial excess returns analysis due to initial ‘mispricing’ phenomenon, and (3) the
long-run abnormal returns analysis due to long-run underperformance phenomenon.
Fifth, this study provide a comparative analysis of long-run performance using the
Event- and Calendar-time approaches to address the issues related to mis-
measurement of long-run returns. This study employed the cumulative abnormal
returns (CARs) and buy-and-hold abnormal returns (BHARs) as Event-time approach,
while capital asset pricing model (CAPM), Fama-French three factor (FF3F) and
Fama-French five factor (FF5F) models as Calendar-time approach to study the
sensitivity of price performance under equally- and value-weighted returns methods.
This study, first time, provides the application and comparison of most recent FF5F
model with other asset pricing models in the literature of the IPO, especially in the
context of emerging markets.
Finally, the existing literature inspected these issues for developed and developing
markets but the country like Pakistan is not taken into account and this study is
devoted to analyzing the numerous aspects related primary market of Pakistan. The
primary markets of developing countries like Pakistan also confront similar problems
in the process of implementing IPOs in the PSX. The findings of this study contribute
to the knowledge that to what extent the Pakistani capital market has converged on the
way to or diverged from a mature market. The portfolio managers can devise their
trading strategies under the findings of this study in order to get superior returns due
to the short run abnormal returns. The regulatory institutions can modify their
procedures for prospectus approval & listing permissions in order to control
19
irregularities, observed in pre-issue fair value estimates and post-issue underpricing &
underperformance, and to attract more listings get listed and to increase investor base.
1.8 Structure of Dissertation
This dissertation consists of five chapters. In the current Chapter one- Introduction,
the brief introduction and background of Initial Public Offerings (IPO), the brief
overview and listing procedure of IPOs in Pakistan, issues related IPO valuations,
issues related aftermarket performance of IPOs, research objectives and questions,
and significance of the study is discussed. The Chapter two is about Literature Review
and divided into three parts. First, literature about IPO valuation is discussed in two
ways as (1) IPO studies used ex-post data for the accuracy of valuation methods and
(2) IPO studies used ex-ante data for accuracy of valuation methods. Second, I discuss
the theoretical foundations and empirical evidence of short run abnormal returns and
long run underperformance of IPOs across the countries. Third, I discuss the issues
related methodologies to estimate long run IPO returns such as comparison of
Calendar Time and Event Study approaches. Chapter three is about Methodology and
Framework of this dissertation. In the first part, population and sample selection
criteria are discussed. The second part presents the methodology and hypotheses to
investigate the choice, bias and accuracy of the valuation methods used by the lead
underwriters. In the third part, discuss the methodology and hypotheses which
investigates the impact of fundamental factors, ex-ante risk factors and signaling
factors on the ex-post IPO performance. In the last, asset pricing models such
asCAPM, Fama-French three factor and Fama-French five-factor models are
discussed to estimate the IPO long-run performance. Chapter four presents the Results
and Discussion of this study. In the first part, the results of binary logistic and cross-
sectional regression models to estimate the impact of choice, bias and accuracy of ex-
ante valuation methods are discussed. In the second part, the results of cross-sectional
models are discussed to examine the impact of prospectus information on the ex-post
IPO returns. In the last part, the CAPM, Fama-French three- and five-factor models
are used to check the robustness of long-run underperformance. Chapter five is about
Summary and Conclusion of this study. In this chapter, the findings summary,
limitation and recommendations, direction for future research and policies
implications are discussed. The last contents of this dissertation are References and
Appendix.
20
2. Chapter 2
2. Literature Review
21
The initial public offerings (IPOs) have been scrutinized mainly with attention to ex-
ante valuation estimates, the theoretical rationalizations of short-run underpricing, the
methodological issues related to long-run underperformance and the most recent
corporate governance issues. A comprehensive literature reflects these issues in detail.
In this chapter, a number of studies are reviewed with respect to (1) the explanations
on the pre IPO valuation estimates with reference to choice, accuracy and bias of
alternative valuation methods, (2) the theoretical and empirical aspects of initial
excess returns also known as underpricing phenomenon and, (3) finally, the
explanatory power of prospectus and market information related to IPOs long-run
underperformance and relative priority of alternate long-run performance models such
as the event- and the calendar-time approaches.
2.1 IPO Valuation
The valuation about initial public offerings has devoted limited attention in the
existing literature. Valuation models have arrived a considerable empirical research in
the field of corporate finance in recent years. Based on the valuation theory, each
model should turn out the same valuation if they are properly constructed. On the
other hand, the relative dominance of each valuation model in academic research and
in practice is an unsettled issue. The literature regarding IPO valuations is categorized
on the basis of accounting information used by the lead underwriters and the financial
analysts’ to predict the valuation estimates.
2.1.1 Post-IPO Valuation Methods based studies
Boatsman and Baskin (1981) used two different types of price-earnings
models to compare the valuation accuracy of comparable firms and the accuracy
defined by the absolute values of prediction errors as a percentage of actual values.
Firstly, they select a random non-IPO firms from the similar sector, and then they
select firms from the similar sector with the most comparable last 10year average of
growth rate of earnings. They report that the valuation accuracy of the second model
is greater than the firm selected randomly in the first model.
Alford (1992) employs price-earnings multiples valuation method to
investigate the valuation accuracy of the IPO firms selected on the basis of industry,
firm size (a proxy for risk) and the growth rate of earnings using the data of
22
comparable firms. Alford also examines the impact of earnings with adjustment for
cross-sectional differences in the leverage. He documents the results of most similar
firms selected based on the industry, the median absolute prediction error measured as
absolute values of prediction errors as a %age of actual values is 24.50%. His findings
using the size in addition to industry membership did not improve the accuracy of
price-earnings multiples method. Lastly, his findings document that the price-earnings
valuation method with adjustment for differences in leverage for all comparable firms
drops the valuation accuracy. He concluded that the industry membership is an
effective principle for selecting comparable firms.
Kaplan and Ruback (1995) examine the valuation accuracy of discounted cash
flow (DCF) valuation method in the setting of the management buyout and the
recapitalization on a sample of 51 highly leverage transactions of large and mature
firms using the adjusted present value model. They pointed out that the event
transaction prices are nearby to the present value of forecasted cash-flows and not
rejected the hypothesis of forecasts are made to justify the prices. They also document
that a CAPM based DCF valuation method has almost the same intent of valuation
accuracy as a comparable firms valuation method as EBITDA used as an accounting
measure. Gilson et al., (2000) also investigate the valuation accuracy of DCF and the
comparable firms’ multiples using the data of firms emerging from the bankruptcy.
They find that the degree of valuation accuracy almost similar in the both approaches.
They also point out that the financial interest of numerous stakeholders in the
bankruptcy proceedings distresses the projections of cash-flows that are used. Cheng
and McNamara (2000) examine the valuation accuracy of P/B, P/E and a combine
(P/E-P/B) valuation methods using the most similar non-IPO firms selected based on
the industry association, size and return on equity. They point out that, when firm’s
value is not available then the combine price-earnings and price-book valuation
method in the comparable firms’ industry association is better than the other valuation
approaches. They also find that the price-earnings valuation approach performs well
than the price-book and a combine (P/E-P/B) valuation methods. These findings
indicate that the both earnings and book values are value relevant but earnings are
more important than the book values and they are not the perfect substitute for each
other. They find that the P/B outperform better than the P/E valuation method when
23
peers are selected based on size-effect and the extent of valuation accuracy surges
with firm size and the size of the target firm’s sector.
Berkman et al., (2000) investigate the valuation accuracy of conformist DCF
and the price-to-earnings valuation methods using the sample of 45 New Zealand
IPOs floated 1989 to 1995. They find that the P/E comparable and DCF have almost
similar valuation accuracy because the median value of absolute pricing error is
approximately 20.0% and cross-sectional model explains 70.0% variation in the
market price when deflated by book value.
Kim and Ritter (1999) examine that how IPOs offer prices are set using the
comparable firms’ multiples valuation of 190 US firms from the same industry listed
from 1992 to 1993. They employ price-sales, price-earnings (P/E), enterprise value to
operating cash-flow and enterprise value to sales ratios to evaluate the offer price
valuation accuracy of the comparable firms. They document that the forecasted price-
earnings (P/E) lead to all other multiples in the valuation accuracy and this valuation
accuracy is more predictive for older firms than for younger firms. They also
document that the forward earnings per share (EPS) for succeeding years lead the use
of present year EPS. Moreover, underwriters able to achieve more valuation accuracy
by campaigning market demand before setting a final price.
Bhojraj and Lee (2002) try to develop a systematic approach to select the
comparable firm from peer groups on the basis of market-based multiples used as
equity valuation techniques using the large sample of US newly listed firms during
1982-1998. They conjecture that the selection of peer firms based on specific
characteristics that compel the cross-sectional fluctuation in the specific multiples
valuation. This study suggests that the proposed systematic approach select
comparable firms on the basis of profitability, growth and the risk characteristics of
each valuation method as explained by the relevant theories. They test the efficacy of
their proposed methodology of comparable firms’ selection criteria using up to three
years forecasted enterprise value to sales and price to book value ratios. Results reveal
that comparable firms selected in this respect get sharp improvements than the
comparable firms selected based on different methods.
Purnanandam and Swaminathan (2004) examine that how IPOs are valued at
the offer prices relative to fair prices using the sample of 2000 US non-financial
24
newly listed firms during 1980-1997. They employ price to EBITDA, price to sales
and price to earnings multiples of comparable firms to estimate the fair values. An
industry grouping was chosen on the basis of Fama and French (1997) SIC industry
codes and comparable firms were chosen based on their operating characteristics.
They observed that the median IPO is overvalued about 14% to 50% based on the
comparable firms from industry peers. When they add forward earnings in the
comparable firms’ selection criteria, the median IPO overvaluation is noticed about
33%. When they add earnings forecast along the industry membership and sales then
the median IPO overvaluation is observed 14%. These findings point out that the IPO
investors mainly focusing on too much analyst forecasted growth and less on the prior
profitability in valuing IPOs.
Firth (2008) compares the IPO share prices offered at the time of listing on the
capital market with the other valuation methods using the data from China listed IPOs
from 1992 to 2002. Their results point out that the price-earnings multiples have more
valuation strength in order to offer the quality of the issuer firm.
How, Lam and Yeo (2007) examine the efficacy of P/E and MTB multiples
estimated on the basis of firms’ management forecast of earnings and book-value of
IPO firm’s equities provided in the prospectuses are strappingly related with the
average prices to earnings and price to book value of two similar firms chosen based
on the industry membership SIC code, earnings growth and size without any prior
adjustment in the multiples. They find out that median prediction errors using the IPO
offer prices of price-to-earnings ratio were (-6.44%) to (-21.54%) and (1.95%) to (-
13.63%) when the price-to-book value was used. They also calculate median
prediction errors using the first trading day closing prices were (-1.58%) to (-13.60%)
for price-to-earnings and (14.25%) to (-1.78%) for price-to-book value multiples. The
findings suggest that the P/E and P/BV used on the basis of management prediction
reported in the offering documents such as prospectuses are closely linked to the
average of P/E and P/BV of matched firms (selected based on the industry
association, earnings growth and size).
Chang and Tang (2007) investigate the explanatory power of commonly used
multiples valuation techniques such as price-earnings, price-book value and price-
sales ratios during the IPO pricing process using the sample of 84 IPOs listed on
25
Taiwan Stock Exchange during 1991-1992 and 1995-1996. They choose comparable
firms based on the comparable revenue characteristics and found the poor accuracy of
aforementioned valuation approaches in evaluating IPO pricing. They find the
positive beta coefficient for peer group medians and suggest that the higher values for
IPOs are associated with the higher values of the peer group.
Table 2. 1: Post IPO valuation studies under different comparable benchmarks
Author(s) Country Sample
Size
Sample
Period
Comparable firms selection
methodology
Boatsman and Baskin
(1981)
US 80 1976 Randomly selected firms from
same industry with 10years
earnings growth rate
LeClair (1990) US 1,165 1984 Similar industry, current earnings
and trailing earnings
Alford (1992) US 4,698 1978,82,86 Industry SIC code, the degree of
risk and earnings growth rate
Kim and Ritter (1999) US 190 1992-1993 Newly listed IPOs on same
industry and firms shortlisted by
research boutique
Lie and Lie (2002) US 8,621 1998-1999 Industry SIC code
Bhojraj and Lee (2002) US 741 1982-1998 Comparable firms on the basis of
risk, growth and profitability
How et al., (2002) Australia 275 1993-2000 Industry membership, size and
growth
Purnanandam and
Swaminathan (2004)
US 2,000 1980-1997 Industry membership, size and
profitability
Cotter et al., (2005) Australia 71 1995-1998 Similar industry
Chang and Tang (2007) Taiwan 84 1991,92,95,96 Newly listed IPOs from same
industry
How, Lam and Yeo
(2007)
Australia 275 1993-2000 Similar industry, growth and size
Firth et al., (2008) China 745 1992-2002 Similar industry
Sahoo and Rajib (2013) India 120 2002-2007 Industry membership, revenue,
book value and return on new-
worth
Source: compiled from various research studies
Campbell, Du, Rhee and Tang (2008) investigate the initial underpricing and
the valuation accuracy with respect to information asymmetry, investor sentiment and
the underwriter reputation using the sample of 2140 IPOs listed during 1970-2004.
They estimate that initial underpricing is larger for overvalued IPOs as compare to
undervalued IPOs and positive association with the information asymmetry and the
investor sentiment. This study employs methodology consistent with the
Purananadam and Swaminathan (2004) as price-earnings, price-sales and price-
26
EBITDA multiples approaches to estimate the intrinsic value of IPO firms. They
don’t find any supporting evidence of systematic overvaluation or undervaluation of
IPOs on the basis of comparable firm’s accounting ratios.
Colaco, Cesari & Hedge (2013) explore the effect of retail investor sentiments
on the initial IPO valuations. They employ search volume index (SVI) as retail
investor sentiment using the sample of 147 US IPO firms listed during 2004-2011.
They employ methodology as comparable with Punanandam and Swaminathan (2004)
and the Liu et al., (2002) as price-earnings, price-to-sales, price-to-EBITDA and
price-to-assets multiples to measure the fair value estimates as compared with the peer
group multiples. They find that a typical SVI trend before the early estimation is
directly linked to the price-to-sales, price-to-EBITDA and price-to-assets multiples.
They argue that the reward to high net-worth institutions and investment bankers for
their particular responsibilities in the book-building process may be unwarranted since
they free-ride on the retail investor sentiments behavior fashions toward initial
valuations and they have not rewarded anyway and forced to buy shares at the high
prices.
Sahoo and Rajib (2013) empirically investigate the impact of multiples
valuation with comparable firms on the basis of: (i) industry membership, (ii) industry
membership and revenue, (iii) industry membership, revenue and book value and, (iv)
industry membership, identical revenue characteristics and the return on net-worth in
estimating the IPO offer prices using the sample of 120 Indian IPO firms listed during
2002-2007. They find that the comparable firms were selected on the basis of the
same industry having similar revenue characteristics, which have the significant
impact on the IPO pricing. They suggest that the peer group PER multiples selected
on the basis of book value, growth and leverage improve the accuracy of multiples
valuations. They point out that the peer group multiples selected based on similar
industry, revenue growth, book value per share and return on equity have 78.20%
accuracy power than the other estimates.
Yoon (2015) conducts an empirical analysis of the three most commonly used
valuation methods for equity valuation such as P/E, EV/EBITDA and the EV/Sales
multiples to investigate the mispriced securities using the calendar time portfolio
regression as advocated by Fama and French (1992, 1998). This study used alpha in
conventional CAPM from value-weighted and equal-weighted regressions for all
27
portfolios are significant which indicates that the multiples valuation approach can be
known as a good estimator to mispriced securities. This paper also used alpha in
three-factor model from value-weighted and equal weighted regressions are
statistically significant after controlling the size and value effect. These results
suggest that mispricing is concentrated in smaller size firms and noticeable in an
equal-weighted designed when size is controlled for in a multifactor model. These
findings are inline with Loughran and Ritter (2000), and Fama (1998) who observed
that the abnormal returns in equal-weighted portfolios are driven by smaller firms.
Furthermore, EV/Sales multiples valuation produces higher alphas than in the P/E
ratio and EV/EBITDA ratios. In a value-weighted one factor CAPM regression, the
price-earnings multiples perform better than the Sales and EBITDA multiples.
Colaco, Cesari and Hegde (2017) examine the impact of retail investor’s
attention on the early IPO valuation. They scan the association between the SVI as a
measure of investor’s attention with IPO valuations through the price to sales, price to
total assets, and price to EBITDA ratios using the sample of IPOs floated during 2004
to 2011. According to the Jumpstart Our Business Startups Act 2012 in the US, the
lead underwriters and the issuers are not allowed to communicate and produce the
earnings estimates before the filing of the initial registration, produce soliciting offers
and raise signals to heighten the demand from investors during the filing of the
prospectus and the initial price is filed. They conclude that, in the absence of
restricted settings for lead underwriters and the institutional control demand, the retail
investor’s concentration take the major part in the initial IPO valuations regardless
that retail investors are painstaking absurd frequently.
Herawati, Achsani, Hartoyo and Sembel (2017) compare the valuation of IPO
offer prices defined by the underwriters at the time of listing and with price-earnings
ratios (PER) using the sample of 240 firms listed on the Indonesia Stock Exchange
during 2000-2014. The results of this study show that the share prices estimated
through price-earnings ratio is greater than 65% of values that are offered at the time
of formal listing. Furthermore, they compare the valuation differences to the extent of
underwriter reputation and find that all underwriters have offered a significant offer
price discount to the fair price estimates. They argue that the underwriters offered a
guarantee to issuer firm through creating sufficient demand to augment investor’s
participation and also to investors through investment opportunity in riskier securities.
28
They suggest that the existing studies considered the reputation of underwriters in the
process of IPOs and auditor’s reputation is also equally importation for the issuer firm
in order to prepare quality financial statements which play a vital role in the valuation
of IPO pricing process.
In the literature of post-issue financial data valuation studies conclude that the
most of the authors completely rely on the comparable multiples, on the basis of
similar industry, size, revenue growth and earnings, valuation methods to estimate the
fair value of issuing firm’s equity while very few studies, only compare the valuation
estimates using DCF and comparable multiples method. Boatsman and Baskin (1981),
Alford (1992), Chang and Tang (2007) argu that P/E selected on the basis of industry
membership produce more valuation accuracy than selected on the basis of other
benchmarks, while (Purnanandam and Swaminathan, 2004; Chang and Tang, 2007;
Campbell, Du, Rhee and Tang, 2008; Colaco, Cesari & Hedge, 2013) P/E selected on
the basis of forward earnings produce higher valuation accuracy than the
price/EBITDA and price/sales multiples. Kaplan and Ruback (1995), Gilson et al.,
(2000) and Berkman, Bradbury and Ferguson (2000) found similar valuation accuracy
of DCF and comparable multiples.
2.1.2 Pre IPO Valuation Methods based studies
Deloof, Maeseneire and Inghelbrech (2002) first time used pre IPO valuation
estimations disclosed in the prospectuses to investigate the valuation accuracy of IPO
offer prices and aftermarket returns used by the underwriters. They took 33 Belgium
non-financial IPO firms based on the valuation methods disclosed in the offering
documents listed during 1993-2000. They use the percentage of valuation errors
within 15%, of mean absolute valuation errors and the mean squared valuation errors
measures to estimate the accuracy of valuation methods. They find that the investment
bankers used Discounted Free Cash Flow (DFCF), Dividend Discount Model (DDM)
and Multiples (price/earnings, price/cash flow, enterprise value/EBITDA, enterprise
value/sales, price/book value, dividend yield and price/earnings growth peer ratios)
valuation methods in the Belgium primary market. They highlight that DFCF is the
most popular method in Belgium. They point out that the DDM tends to
underestimate the intrinsic value whereas the DFCF generates unbiased intrinsic value
estimates which reveal that the underwriters deliberately offer price discount by
29
relying on the valuation approach that tends to underestimate the value. Their results
document that multiples valuation with forward earnings and cash flows leads to more
accurate valuations than the multiples valuation on the basis of latest financial year
earnings and the cash flows because the offer prices are closer to stock prices
estimated on the basis of forecasted earnings and the cash flows by adjusting more
valuable price-sensitive information.
Cassia, Paleari and Vismara (2004) investigate that how underwriters use
valuation methods to set preliminary IPO prices and their valuation accuracy using the
data of 83 Italian newly listed firms during 1999-2002. They extract “real-world”
valuation details from the prospectus document and observed that the Italian lead
underwriters employ multiples valuation and DCF methods to price IPOs. They
observed that EV/EBITDA, P/E, P/BV and P/CF are most commonly used multiples
in Italy. These studies uncover that the investment banks mostly depend on P/BV and
P/E ratios during IPO pricing process. On the other side, the results of EV to sales and
EV to EBITDA are overvalued from comparable firms than the estimates from the
IPO firms. The results come from P/E, P/BV and P/EBITDA is closer to IPO offer
prices. They argue that the valuing IPO is a great responsibility to lead underwriters to
avoid mispricing and also to build the reputation. Underwriters deliberately use
conservative measures to lease good taste for investors and successfully subscribe
their deals.
Roosenboom (2007) investigates that how French underwriters choose the
valuation methods to set preliminary IPO offer prices using the sample of 228 French
non-financial newly listed firms from 1990 to 1999. He used pre IPO characteristics,
valuation methods disclosed in offering documents and the unique data of detailed
valuation reports from the leading financial advisors. He employs binary logit model
analysis to enlighten the choice and accuracy of valuation methods. This study
observed that the French lead underwriters used the DCF, DDM, the peer group
multiples valuation estimates and EVA valuation methods to price IPOs and these
valuation methods are selected on the basis of firm characteristics, the aggregate stock
market returns and volatility before IPOs. This study points out that underwriters have
a preference to choose peer group multiples valuation when valuing rapidly growing,
technology and profitable firms. They prefer to use DDM when valuing older firms
from the mature industries because the older firms pay a large portion of their
30
earnings as dividends and when the aggregate stock market returns are low before
going public. The discounted free cash flow model and the economic-value added
valuation measures used when aggregate stock market returns are high prior to IPO.
This study document that the choice of valuation method is purely based on firm’s
industry membership, the aggregate stock market returns and the aggregate stock
market returns volatility but also partially based on the capital market dynamics,
industry circumstances and the firm-specific factors. This study also documents the
significant determinants of offer price discount as the underwriters with high
reputation are linked with the lesser discounts, growth in trailing earnings are
associated with lesser discounts but large offer price discounts are associated with the
riskier firms.
Deloof, Maeseneire and Inghelbrech (2009) investigate that how do leading
underwriters value the IPOs using the numerous valuation methods to set the
preliminary offer prices and the valuation accuracy of these methods to produce
unbiased estimations of 49 non-financial IPO firms listed on the Euronext Brussels
(previously known as Brussels Stock Exchange) from 1993 to 2001. This study used:
(i) the percentage of offer price to value estimates differences within 15% and the
mean absolute error to investigate the pricing error, (ii) mean and median estimated
value to market value estimation errors to highlight the degree to which value
estimates are biased, and (iii) the percentage of valuation errors within 15% and the
mean absolute errors measures used to estimate the valuation accuracy. They
observed that the discounted free cash flow, the dividend discount and the peer group
multiples valuation approaches used by Belgium investment banks, of which DFCF is
noticed to value in all the IPOs, the offer prices are value relevant to DDM
estimations when DDM is used, and the peer group multiples valuation based on the
forecasted earnings and cash flows produce more accurate estimates than the based on
the latest financial earnings and the cash flows disclosed in the offering documents.
They point out that the underwriters deliberately offer price discounts to DFCF value
estimates, DFCF is the most popular and reliable valuation technique whereas DFCF
produces unbiased results. The DDM estimations are more consistent with IPO offer
prices than the other valuation methods as DDM tends to underestimate the values.
Roosenboom (2012) investigates that how investment banks decide the IPO
offer prices, an ex-ante estimate of the market value and how these fair valuations are
31
subsequently used as a basis for IPO pricing using the sample of 228 non-financial
firms listed in France from 1990 to 1999. This study used a unique access of detailed
valuation reports by lead underwriters during the IPO pricing process. This study
employs the methodology as comparable with Francis et al., (2000) to evaluate the
bias, accuracy and explain-ability of the valuation method. This study observed that
the French lead underwriters used peer multiples, DDM, DCF, economic value added
and analysts-specific valuation methods to value private firm owner’s equity. This
study observed that the multiples valuation of the comparable firms is the most
popular valuation approach by French lead underwriters. The findings reveal that
DDM, DCF and the multiples valuation have similar bias, accuracy and
explainability. The lead underwriters intentionally “leave money on the table” to fair
value estimate in the process of IPO pricing. Underwriters market this offer price
discount during the IPO marketing campaign to amplify the investor’s participation in
the IPO auction or book-building process. This results in partial recovery in offer
price discount and turns out the high offer price updates. The findings are inline with
partial adjustment in offer price discount phenomenon and this phenomenon cause the
reason of large underpricings after controlling the investor demand.
Abdulai (2015) investigates the impact of choice, pricing and performance of
different valuation measures on post-issue performance of 30 IPOs listed on the
Ghana Stock Exchange (GSE) during 1992-2012. He employed binary logit and
cross-sectional regressions to examine the performance of valuation measures, for that
purpose, data obtained from the prospectuses and the GSE DataStream. He argues that
the DCF most commonly used method by underwriters. The findings of comparable
multiples document that the P/E and P/B are the most common measures used as
multiples while the DDM is a least popular measure and only used by one IPO firm.
The findings reveal that the operating profitability before the IPO is a significant
determinant of choice of multiples while firm age, size and dividend payout are the
key determinants of direct valuation methods.
Goh, Rasli, Dziekonski and Khan (2015) investigate the choice of different
multiples of newly firms related agricultural business sector listed in the Malaysia
during 2003-2009. They argue that the accuracy of each valuation multiples is based
on the; (1) a reliable benchmark multiples, and (2) to recognize the comparative
performance of comparable multiples and core elements of agricultural business
32
sector. They employ Price-earnings (P/E), Price-book (P/B), Price-cash flow (P/CF),
Price-sales (P/S), price-total assets (P/TA) and Return-on-equity (ROE) ratios of
comparable firms based on the industry membership and the firm’s rate of growth.
The findings reveal that; (1) the median is an accurate measure to estimate the
comparable multiples and the accuracy of valuation measures, (2) the ROE used as a
control factor adjustment in the comparable firms leads to unbiased valuation
estimates, (3) the price-sales (P/S) ratio perform worst in the valuation performance
while price-earnings ratio produce more accurate valuations when valuing the
agribusiness IPOs.
The most recent study, Rasheed, Sohail, Din and Ijaz (2018) investigate that
how lead underwriters select alternative valuation methods to value IPOs listed in
Pakistan. This study also unfold the value revelancy of each valuation method used by
underwriters. This study used sample of 88 IPOs floated during 2000-2016. They find
that underwriters prefer to choose DDM method when firms have corporate payout
record before the IPO and offered through the prestigious underwriters. The
investment banks are more lsikely to choose DCF when valuing young firms, rapidly
growing firms and that have more assets-in-tangibility. The investment banks prefer
to select comparable multiples when valuing mature firms and IPOs offered in bullish
sentiments. The findings of value relevance analysis unveil that the DCF has lowest
but P/B has highest predictive power to market estimates. Amir et al., (2018) also find
similar results with Rasheed et al., (2018).
The findings of IPO literature that used underwriter’s real valuation analysis
information to estimate the fair value and accuracy of several valuation methods
employed when valuing IPOs demonstrate that investment banks employed DDM,
DCF, P/E, P/B and economic value added method to estimate fair value estimates.
Deloof et al., (2009) found the similar valuation accuracy of DDM and DCF while
Deloof, Maeseneire & Inghelbrecht (2002) and Berkman et al., (2000) found that the
DCF valuation method produce more valuation accuracy than the P/E, P/B and other
multiples to estimate fair value estimates. On the other hand, Cassia, paleari &
Vismara (2004), Schreiner and Spremann (2007), Demirakos, Strong & Walker
(2010), Roosenboom (2012) and, Goh, Rasli, Dziekonski & Khan (2015) argue that
the comparable multiples such as P/E and P/B valuation multiples outperform the
DDM and DCF in order to produce more accurate fair value estimates. The literature
33
on valuation biases also depict that the P/E ratio produce less valuation bias than the
other methods (Deloof, Maeseneire & Inghelbrecht, 2002; Cassia, paleari & Vismara,
2004; How, Lam & Yeo, 2007), while Deloof, Maeseneire & Inghelbrecht (2009)
document that the DCF produce less valuation bias than DDM and on the similar
pattern, Francis, Olsson & Oswald (2000) report DDM produce less valuation bias
than the DCF.
2.2 The IPO Underpricing Phenomenon
The IPO underpricing is the difference between the preliminary offer price of
newly offered equity and the closing price over the first few days. These initial returns
are statistically positively significant across the world but vary in percentage
(Ljungqvist, 2006). When there is uncertainty about the intrinsic value of IPO firms
then some momentous mispricing is to be expected. IPO underpricing phenomenon
was firstly documented by Reilly and Hatfield (1969) and they took the sample of 53
US IPO firms listed during 1963-1965 and found initial returns range from 18.30% to
20.20%. After Reilly and Hatfield (1969) study, many researchers found similar
results by taking different time intervals and the sample sizes across the world (e.g.,
Neuberger & Hammond, 1974; Logue, 1973; McDonld & Fischer, 1972). Ibbotson
(1975) firstly, measured the initial excess returns and provides the comprehensive
explanation of potential determinants. He provides six possible explanations of
underpricing in the IPOs are as (1) the US Securities Exchange Commission (SEC)
provide instructions to the underwriters, determine the IPO offer price below than the
expected intrinsic value; (2) IPO issuer firms and underwriters deliberately offered a
discount to the investors in the IPO offer price “to leave a good taste in IPO” to make
seasoned equity offerings at attractive price; (3) Underwriters may favor to their
investors through the large underpricing; (4) underwriters deliberately present offer
price discount in the IPOs to control the uncertainity of undersubscription and
credibility; (5) investors may compensate issuers for large offer price discount by
some undisclosed mechanism, and (6) underwriters may present offer price discount
to protect issuers by the investor’s lawsuit. Loughran and Ritter (2002) explain the
phenomenon of underpricing as “Leave so much money on the table” when market
prices are higher than the IPO offer prices, due to bulky underpricing, issuer loses
wealth than they expected to be. It’s difficult to assess the exact causes to IPO
34
underpricing; various researchers try to explain this phenomenon with numerous
models and propositions.
2.2.1 The winner’s Curse Hypothesis
The most widely used underpricing model was built by Rock (1986) to discuss
the theoretical foundations of the IPO underpricing. Rock (1986) theorizes that with
fixed-price offers, underpricing arises due to information asymmetry between the
market participants. Rock assumes that investors who have more information called
the informed investors and less informed investors are called uninformed investors.
The informed investors struggle only for ‘good’ issues and there is a high probability
that the uninformed investors obtained more ‘bad’ issues due to less information
about fair value. Therefore, it is expected that the ‘good’ issues have excess demand
by market participants and the ‘bad’ issues face excess supply. This adverse selection
behavior is known as a winner’s curse problem. Thus, underwriters deliberately offers
a discount in the IPO preliminary offer prices in order to encourage uninformed
investors. Beatty and Ritter (1986) also used the Rock (1986) model to propose
another justification of IPO underpricing based on the Rock’s assumptions of
asymmetric information. They argue that the ex-ante uncertainty about the fair value
of issuing firm may affect the IPO underpricing. They argue that the degree of ex-ante
uncertainty is positively related to the degree of IPO underpricing. They used the
number of uses of IPO proceeds and the reciprocal of IPO gross proceeds as a proxy
for ex-ante uncertainty. They argue that the issuer firm unwilling to share the
information of proceeds utilization plan in order to avoid increased exposure to the
lawsuits and to meet stringent disclosure requirements by market regulators. This
implies that the ex-ante uncertainty is a key determinant to explain the degree of
underpricing. They conclude that the degree of ex-ante certainty is positively related
to the extent of underpricing. Keloharju (1993) also confirm the presence of winner’s
curse problem using the Finnish Capital Market data. Michaely and Shaw (1994)
examined the Rock winner’s curse hypothesis using the two IPO samples of relatively
homogenous IPOs and the general IPOs respectively. They find that the results are
consistent with the Rock hypothesis but the level of underpricing vary in both
samples.
35
The level of ownership structure is an important factor during IPO valuation
process. The recent literature, Bruton et al., (2010) examined the impact of ownership
concentration on the IPO performance in the France and UK, and found that
concentration ownership improves IPO performance. As agency theory stated, the
numerous researchers argue that ownership concentration decreases the asymmetric
information problems associated with disperse ownership (Barry et al., 1990; Shleifer
and Vishny, 1997).
2.2.2 The Prestigious Underwriter Hypothesis
This hypothesis keeps the same assumptions on information asymmetry
hypothesis proposed by the Rock (1986) among market participants, the investment
bankers and issuers hold more inside information about firm’s prospects than the
outside investors called uninformed investors. Baron (1982) firstly introduced the
model of underwriter reputation based on the agency theory to focus on the optimal
behavior of issuers as the principals and the underwriters as the agents to explain the
underpricing. The Baron’s model suggests that a investment banker has more
information about the firm’s intrinsic value and in turn results in an excess demand of
an IPO. He reveals the positive relationship between the demand of an IPO and the
underwriter reputation. The prestigious underwriters leave less money on the table.
There is an inverse significant relationship between the underwriter reputation and the
level of underpricing. Muscarella and Vetsyupens (1989) argue that the reputed
offering agents also face underpricing in their own IPOs. Beatty and Ritter (1986)
demonstrate the role of underwriters by taking care of their credibility in the
underpricing equilibrium. Therefore, the underwriters deliberately present limited
offer price discount to increase excess demand due to three necessary conditions.
First, the underwriters are uncertain about the demand and aftermarket price of the
IPO firms. Second, the underwriters have non-salvage reputation investment at stake.
Third, the underwriters may suffer from committing fees and the commission if they
offer too much money in the form of large underpricing or little. They also argue that
if the underwriters cheat during the allocation process, could lose the confidence of
potential issuers and the credibility among market participants. Following these
studies, various researchers investigate the impact of underwriter reputation on the
degree of underpricing and found a negative association between the ranking of
investment bankers and the extent of initial excess returns.
36
Furthermore, Johnson and Miller (1988) examined the relationship between
the underwriter reputation and the degree of underpricing by proposing another
hypothesis of adding ex-ante uncertainty in the model. They illustrate that the
underwriter prestige cannot change the underpricing equilibrium when the pre-IPO
uncertainty has been taken into account. When more information is available to the
investors then excess demand occurs by the informed investors. Based on the mean-
variance efficiency theory, a risky firm could not control the underwriting costs by
choosing a prestigious underwriter; on the other hand, firms can control the
underpricing through a higher investment banker commission. Therefore, the total
IPO transaction proceeds costs are positively related to the extent of ex-ante
uncertainty about the firm’s intrinsic value and isolated the impact of underwriter
reputation on underpricing. They used a sample of 962 US IPOs from 1981 to1983
and find a negative association between the level of underwriter reputation and the
level of underpricing.
Carter and Manaster (1990) used similarly to (Johnson and Miller, 1988)
procedure to assign the value of underwriter reputation by examining the tombstone
advertisements. By analyzing the relative placement of underwriters on the
advertisements as compared to other peers, each underwriter is allotted a rank from
nine to zero with respect to their appearance from top to bottom on the advertisement
respectively. The results of this procedure assigned zero to least prestige and the nine
to most prestige underwriters. They used the sample of 501 US IPO firms listed
during 1979-1983 and support the argument that the underwriters launch only less
risky IPOs in order to safeguard their credibility in the financial market. Therefore,
from the issuer’s point of view, the less risky firms try to choose prestigious
underwriters to signal the quality of the firm’s prospects.
Kumar and Tsesekos (1993) investigate the relationship between the
underwriter reputation and the type of commitment agreement. They considered the
use of over-allotment option makes benefit to the investment bank to build their
relationship with the potential investors and reputation through the overallotment
option as a provision in the underwriting contract. Hence, the overallotment option
allows the investment banker to satisfy the excess demand of investors. The
underwriters having low ranking have more interest in building a relationship with the
investors and also making a more investor base. They hypothesize that there is a
37
negative association between the relative size of overallotment option and the degree
of underwriter reputation. From the issuer’s point of view, the underwriters are
certified as financial advisors and set the IPO offer prices consistent with the inside
information about the firm’s forecasted cash flows. They find that the type of the
underwriting agreement determined the level of support offered by the investment
bankers.
Logue et al., (2002) investigate the impact of underwriter reputation during the
IPO process. Their results depict that the underwriter reputation is a significant
determinant during the due diligence process and the allocation of shares to the
investors, but less related to the post-IPO price stabilization motions and not related to
the post-IPO returns in the long run. They also explored that the underwriter
reputation is a key determinant of the IPO underpricing but not related to the IPO
long-run performance.
Amihud, Hauser and Kirsh (2003) used the IPOs data of Tel Aviv Stock
Exchange to perform a powerful test of the Rock’s model. They have the subscription
of each investor and their allocation were undertaken by equal pro rate basis during
their study period. As similar to Rock’s model, the small investors allotted greater
allocation in overpriced IPOs. Furthermore, the large number of orders was submitted
by informed and uninformed investors in the more underpriced IPOs, while
uninformed investors submitted more orders in the overpriced IPOs and ultimately
earned adverse excess returns across the sample offerings in which they subscribed.
2.2.3 The Signaling Hypothesis
Allen & Faulhaber (1989), Welch (1989) and Grinblatt & Hwang (1989)
introduced the signaling model based on the information asymmetry assumption to
explain the underpricing anomaly. Though, this model supposed that the issuer firms
know better regarding the future prospects instead of the external investors or the
underwriters. They also assumed that there are two types of firms, (i) good firms and,
(ii) bad firms. Though, the outside investors unaware about the quality of firms until it
is publicly available. Consequently, the quality firms reveal inside information to
potential investors and deliberately offer a discount in the preliminary offer prices to
give a signal of quality firms. In the extant literature of the IPO, firms employ
numerous variables as a signal of their quality such as selection of prestige
38
underwriters or auditors, prior IPO quality of management and executives, quality of
share capital structure and pre-IPO debt borrowings, and others. If signals work
successfully, the good quality firms split themselves from fewer quality firms with the
help of aftermarket performance. According to Ibbotson’s words, issuers deliberately
underprice their IPOs to ‘leave a good taste in the market’; therefore, this will assist
the firms to successfully accomplish the seasoned equity offerings in the market.
Thus, deliberate underpricing used as a signal to fetch a high price in the subsequent
equity offerings and could recover the loss of initial offerings (Jegadessh et al., 1993).
Various signaling models show different empirical relationships between the
underpricing and the intrinsic value of the IPO firms, the degree of a firm’s quality,
the project uncertainty, the seasoned equity offerings and the market sentiment at the
time of issue. The Grinblatt & Hwang model associate the project uncertainty to the
level of underpricing and the proportion of shares held by initial sponsors at the time
of offerings. They argue that the degree of underpricing is an increasing function of
the percentage of shares offered in the IPO event.The Allen & Faulhaber model
discuss another implication about the hot issue of market timings and the degree of
underpricing. This model provides that the market timing issue happens to the specific
sector when a regulatory or economic shock significantly impacts the profitability of
that particular sector. The signaling models proposed that the degree of underpricing
is an increasing function of ex-ante uncertainty before the IPO. The proposed
association is entailed by exiting signaling models until the strident market is required
to attain equilibrium that guarantees the degree of underpricing. The Welch’s model
assumes that the initial public offerings always followed by seasoned equity offerings
and the market value of IPO firms drop less upon news of SEO when a firm tries to
adjust underpricing equilibrium in order to change investors previous’s conviction
about firm value.
Welch (1989) shows there is a significant positive association between the
probabilities of firms’ undertaking seasoned equity offerings and the degree of
underpricing. Michael & Shaw (1994) used different sample period and their
empirical findings inconsistent with the signaling models. They find that the firms
with less underpricing experience high earnings and payouts in the aftermarket are not
consistent with the signaling model assumptions. They also find that the degree of
underpricing is not positively linked with the firm value and the ex-ante uncertainty.
39
They do not find supporting evidence for the positive association between the degree
of underpricing and the post-IPO payout policy which is proposed by Allen &
Faulhaber in the signaling model. They also investigate the long-run performance and
find that the firms experience to undergo seasoned equity offerings in the aftermarket
outperforms the non-issuing firms. On the other hand, they also do not find any
association between the degree of underpricing and the proportion of shareholding
with insiders in the long run.
Jegadesh et al., (1993) used the sample of 1,985 US IPO firms listed during
1980 to 1986 to investigate the relationship between the underpricing and the
probability of undertaking SEOs. Their results support the evidence of signaling
model hypothesis and find the positive association between the probability of
undertaking SEO and the underpricing. Espenlaub andTonks (1998) used the UK
IPOs data to investigate the association between the post-IPO directors’ sales and the
probability of undertaking SEO. They argue that the initial shareholders deliberately
underprice the IPOs to get benefits in the seasoned equity offerings. They proposed
that the there is a positive association between the degree of underpricing and the
probability of directors’ sales in the SEOs. Their results indicate the positive
association between the size of post-IPO executives’ sales and the degree of
underpricing but they do not find evidence that the post-IPO directors’ sales are a
significant determinant of conducting seasoned equity offerings. Keasay and
McGuiness (1992) investigate the implication of positive association between the
underpricing and the firm value using the UK USM digital data and used fifth-day
market capitalization is as a proxy for firm value. They find a positive association
between the underpricing and the firm value as predicted by the signaling hypothesis.
Su and Fleisher (1999) test the signaling hypotheses using the 308 IPOs listed during
1987 to 1995 in the China. They find stunning initial returns of 949% and argue that
their findings are consistent with the signaling hypotheses and the degree of initial
excess returns appears to be large to indicate an effect of rational economic
bargaining.
Yu and Tse (2006) test the reasons of underpricing in the Chinese primary
market using the sample period from 1995 to 1998. Their results do not support the
evidence that the underpricing is not mainly because of the signaling hypothesis but
consistent with the winner’s curse hypothesis. Chorruk and Worthington (2010)
40
investigate the initial returns using the 136 Thai IPOs listed between 1997 and 2008
and they find a underpricing of 17.60% and recognize that the initial excess returns
has not declined over time. Low and Yong (2011) estimate underpricing using the 368
Malaysian IPOs listed between 2000 and 2007 where the fixed-price mechanism is
most common for IPO process and control the underwriter’s knowledge of cumulative
demand for offered equity. They document an underpricing of 30.80% and argue that
the issuers offer their equities at a lower price to attract high demand and the signal of
high-quality firms.
2.2.4 The Lawsuit Avoidance Hypothesis
The insurance hypothesis is an additional underpricing explanation which
initially proposed by Ibbotson (1975) argues that the underpricing is used to avoid the
lawsuit from the external investors. Later on, Tinic (1988) and Hughes & Thakor
(1992) test this hypothesis and argue that the issuer firms and the underwriters
deliberately underprice their IPOs to avoid lawsuits from the outside investors.
Tinic (1988) tests the lawsuit-avoidance hypothesis and argue that the
expected cost of lawsuit file by outside investors would be sky-scraping9 for issuing
firms for the reason that to conduct the due diligence examinations which entail a
number of uncertainties and complications. As a result, underwriters and issuer firms
deliberately underprice their IPOs in order to avoid lawsuits situations. Furthermore,
in the IPO placement process, the issuer usually unaware of the disclosure
requirements and may possibly seem to be an unimportant part of the information to
be shared probably be observed a material oversight in a common deed. Secondly, the
underwriter assess the quality of management and their capacity to perform
organizational operations in a good manner based on the judgment and subjective
evaluation. Even, the speculative qualities and risks about IPO security are usually
mentioned in the due-diligence and prospectus documents. While the underwriters
and issuers both are at risk to the legal obligations and want to protect them would be
to buy jointly insurance policy against possible losses. The issuers and the
underwriters along with the legal protection from potential damages, they produce
information about the firm in order to successfully complete the transaction process.
9 The issuance of IPO being under the very high influence of SEC compliance and face a heavy plenty incase of failure to comply.
41
This may amplify the chances of post-offering sue filings by the outside investors and
the possible losses for the issuer firms. To keep safe from this situation, the issuers
pay a large amount to make sure the quality of underwriter valuations and charge a
mutually decided fee cuts when it found to be dodging. If underwriters could
formulate the provable standards for a due diligence examination then the cost of
insurance policy could contain a premium for the moral hazard. Tinic argues that in
the absence of the legal protection against lawsuits, deliberate underpricing is an
efficient measure and the incomplete information disclosure in the offering documents
for both the underwriters and the issuers. Finally, deliberate underpricing reduces the
chance of the legal action and the potential damages in the situation of an adverse
evaluation.
Hughes & Thakor (1992) argue that the issuer firm hire underwriter to set the
preliminary offer price which acts as certify financial advisor and fully responsible
aftermarket lawsuits. They also argue that the underwriters set offer prices based on
their professional expertise and the subjective judgments, and would be litigated in
the future if it is observed the evidence of mispricing. The investment banks
deliberately underprice the IPOs to reduce the probability of litigation because of
legal action apparently expensive to underwriters and damage their reputation too.
They also suggest that there is a trade-off between minimizing the chance of a lawsuit
and maximize the IPO proceeds. In their proposed model, they suppose that the
reduction in litigation likelihood increases the IPO preliminary offer price, means the
positive association between the overpriced an IPO and the probability of a litigation.
Furthermore, deliberate underpricing reduce the probability of a litigation, adverse
ruling conditional being filed by investors and the underwriter reputation in case of an
unfavorable ruling.
In order to investigate the lawsuit-avoidance hypothesis, Tinic used 204 US
IPOs and use the year 1933 as a cut-off point in time about the Securities Act of 1933.
She divides the sample period into two categories. The first pool consists of 70 IPO
firms listed during 1923-1930, and the second pool consists of 134 IPO firms listed
during 1966-1971. Before the legislation of SEC Act of 1933, the principle of caveat
emptor filed for listing in an almost open way, the underwriter and issuer do not face
lawsuit uncertainty. Since 1933, the pace of underpricing has gone up with
simultaneously to amplify the probability of the future lawsuit. The evidence sustains
42
the proposition that the underwriters deliberately underprice their IPOs to avoid
litigation for misstatements in the offering documents. Drake and Vetsuypens (1993)
investigate 93 IPO firms who were subsequently sued under the legislation of SEC
Act of 1933 during the period 1969 to 1990. They explored that the buyers of
overpriced IPOs are just as probably sued as of underpriced ones. Therefore, they find
that the underpricing does not control the probability of a lawsuit-avoidance and also
argue that the deliberate underpricing is a costly form to avoid future litigations.
As compared to the lawsuit-avoidance hypothesis, Ruud (1993) argue that the
underwriters underprice their IPOs to support aftermarket trading to fall below the
certain level instead to avoid future litigations. The investment banker supports the
aftermarket share prices through the repurchasing of a sizeable number of shares
when it reaches below the offer price and resulting the price goes up higher than the
initial offer price and leave a good taste to investors. Jenkinson (1996) supports the
Ruud hypothesis that the underwriters deliberately offered a discount in the offering
price to leave a good taste to investors in the post-IPO period. Prabhala and Puri
(1999) present the evidence against the arguments proposed by Tinic. They compare
the firms listed before 1933 and firms listed during 1985 to 1994 and assume that the
firms listed prior 1933 should be riskier than the firms went public during 1985 to
1994 indicating that the underwriters faced more lawsuit uncertainty than the firms
went public during 1985 to 1994which used the IPO offer prices, the proportion of
shares offered and SD of initial returns as proxy for IPO risks and explored that the
firms which went public during 1985 to 1994 were riskier than the firms went public
before 1933. Finally, they argue that the differences in underpricing before 1933 and
during 1985 to 1994 are possibly on the basis of differences in the IPOs risk instead of
the risk obligatory by SEC Act of 1933. Lowry and Shu (2002) employ methodology
comparable with (Drake and Vetsuypens, 1933), but identify that they ignore the
endogeneity bias of the relationship between the underpricing and the risk of a
lawsuit. They empirically examine the lawsuit-avoidance hypothesis by investigating
the association between litigation uncertainty and underpricing. They supposed that
firms having higher chances of lawsuit produce higher underpricing to avoid legal
obligations and firms having large underpricing to control the chance of being sued.
They used the sample of 1,841 newly listed firms during 1988-1995 and used
43
simultaneous equation system to test proposed hypotheses. Their results provide the
evidence of a lawsuit-avoidance hypothesis.
Hao (2007) test the lawsuit-avoidance hypothesis using the IPOs listed
between 1996 and 2005 and finds no evidence for litigation risk hypothesis. She
suggests two possible rationales about results. First, the probability of being sued
about directors & officers, and errors & omissions has been replaced with IPO
underpricing as an alternative measure to control the potential damages of future legal
obligations. Second, she identifies that the Private Securities Litigation Reform Act of
1995 and the Securities Litigation Uniform Standards Act of 1998 have substantially
minimized the risk of the lawsuit by the US investment bankers, to control the
requirement to purchase litigation insurance in the course of underpricing. Lin,
Pukthuanthong and Walker (2013) investigate that the pre-IPO characteristics
including the initial excess returns have a minor effect on the litigation uncertainty
and the risk of being sued is commonly determined based on the ex-post activities
including comparable firms’ downturns. Hanley and Hoberg (2010) used recent IPOs
data and find that the issuer firms deliberately share more information about the
utilization plan of the IPO proceeds and the potential risk factors involved in the
ongoing business operations. This deliberate disclosures possibly reduce the
probability of litigation risk.
2.2.5 The Prospect theory
The prospect theory of Kahneman and Tversky (1979) argue that the
individuals devise options under uncertainty maximizing a value function other than
an expected utility function that set priority according to the extent of expected utility.
Loughran and Ritter (2002) proposed a prospect theory, based on the investor
behavior to explain why issuers are willing to leave money on the table. Theory
suggests that issuer firms only concentrate on changes in their wealth instead of the
lost of wealth in the underpricing. The assumption of this theory explains that the
issuers leave a lot of money on the table because they are expecting a higher price in
the initial days, resulting to offset the loss of wealth in underpricing and the net gains
in their corporate wealth. They explore that the initial excess returns are mainly
dependent on the degree of underpricing when preliminary prices increased as
compare to price bands. They unfold the explanatory power of initial returns which
44
primarily based on the information disclosed in the offering documents. Moreover,
they explore that the filing range of offer price does not completely represent the
public information. Loughran and Ritter build the arguments from the prospect theory
that the issuer could consider the opportunity costs of benefits and the losses varying
relative to expected IPO net proceeds. They proposed that the issuers are reluctant to
make guarantees in case of ex-post surprises due to unswerving asymmetry with
prospect theory. Loughran and Ritter further interpreted the prospect theory that the
degree of initial excess returns depends on competitive restraint. The competitive
forbearance particularly plausible in Japan with having a track record of
governmentally engineered and insist on cartels. On the other hand, in the US, they
find that the style of partial adjustment for low market share investment banks is same
as the other industry leaders.
Loughran and Ritter (2002) investigate that the public information
incorporated partially based on the willingness of both the issuers and underwriters
“higher than necessary” underpricing. They proposed a behavioral explanation that
IPO issuing firm care regarding the change in wealth than the extent of underpricing.
The initial sponsors’ wealth affect represents the sum of two elements: money left in
shape of underpricing and the dilution effect of shares offered in an IPO, plus money
gained as shares retained by initial shareholders. If the difference between two
elements more than the loss from first element then according to the prospect theory,
initial shareholders will be content.
2.2.6 An International Empirical Evidence
The extant literature on initial public offerings aftermarket performance has
provided support to the underpricing phenomenon approximately exists in across the
world markets. The various theoretical explanations of underpricing is already
discussed in the previous sections and this study document some empirical evidence
of this phenomenon from around the globe. In the US market, during 1980-1990
periods, underpricing was approximately 7%, but during 1990-2000 periods, it
becomes double around 15%. Liu and Ritter (2010) investigate the underpricing using
the sample period of from 2001 to 2008 and found the underpricing of 12%. These
studies point out that the degree of initial excess returns has changed over time.
45
Chambers and Dimson (2009) used UK IPOs data listed during 1989-2007 periods
and found the underpricing of 19%.
Perera and Kulendran (2012) study the Australian primary market and found
the underpricing of 25.47% on the first day and 23% on the 10th day using the
cumulative abnormal returns method. In a study of the New Zealand market,
Alqahtani and More (2012) investigate initial returns of IPOs and found the
underpricing of 9.16% due to lower risk of issuing firms. Falck (2013) tests various
theories of underpricing by the Norwegian IPOs data and document an underpricing
of 3.14%, their empirical findings support the information revelation theory was a key
explanation of underpricing phenomenon. Kucukkocaoglu (2008) investigate the
initial returns on early trading days using the IPOs data. He argues that the initial
returns were higher when investment banks used fixed price and book-building
mechanisms to launch the IPOs. He identified that the underpricing on the 1st trading
day and for the 1st month were 11.73% and 14.12% respectively. Bernnan and Franks
(1997) have investigated the dilution of ownership and the control of management
evolves in the IPO process in order to analyze and retain the post-IPO ownership due
to the IPO underpricing. They used 69 IPOs data from the London Stock Exchange
newly listed firms during the 1986-1989 period. Their empirical findings highlight
that the initial underpricing is used to reach over subscriptions which allow initial
promoters to differentiate against big bids to avert block holdings. They argue that the
initial promoters release their 2/3rd of their ownership in succeeding offerings of 7
years while other executive directors hold only docile shares holding. Field and
Sheehan (2004) test the hypothesis that the managers deliberately underpriced their
IPOs to dissolve the ownership structure to attain the private benefits from less
scrutinized by the other stakeholders. They used the sample of 953 IPOs data and the
binary logit model to empirically test the proposed hypothesis and conclude that there
is no significant association between the initial returns and the outside block holdings.
Scultz and Zaman (1994) used the sample of 72 IPOs data listed on NASDAQ and
examined the underwriter’s aftermarket stabilization activity for the first three trading
days since the issue of IPOs. They support the evidence of underwriters aftermarket
stabilization activities and the answer of deliberate underpricing. They find that the
underwriters support aftermarket share prices through the buying when the price fell
below a certain level for both issues of hot and cold, eventually the IPO price
46
amplifies over the offer price. Yuhong (2010) examine the initial underpricing of
IPOs data listed during the internet boom period from 1999 to 2000 and dividing the
sample firms into two categories; (1) internet IPO firms and, (2) non-internet IPO
firms. His results highlight the underpricing of internet IPO firms was 88.60 % as
compared with the underpricing of non-internet IPO firms to be 44.70 %.
Table 2.2 presents the overall summary of initial underpricing on the first
trading day across the world capital markets but the level of underpricing varies from
country to country. This table provides underpricing from both developed and
emerging markets literature. Existing literature from emerging markets on IPOs
aftermarket performance has documented the higher underpricing relative to the
developed markets (Loughran et al., 1994). During the 2000 to 2010 decade, Sohail
and Nasr (2007) used the sample of 50 IPOs listed from 2000 to 2005 and employ
market adjusted abnormal returns method to measure the initial excess returns in the
Pakistani primary market and, found on average underpricing of 35.66% on the first
trading day. They also investigated that the ex-post uncertainty, offer size, over-
subscription and the market capitalization at 5th trading day were the significant
determinants of the underpricing. Hassan and Quayes (2008) investigate the initial
excess returns on the few early days using the IPOs data of Bangladesh and document
that the underpricing on the first day and 21st trading day are 108% and 119%
respectively. Sahoo and Rajib (2010) used the sample of 92 IPOs listed on the Indian
primary market during the 2002-2006 periods and found the underpricing on the first
trading day was 46.55% which mainly attributed to the investors over expectations.
Samarkoon (2010) documented the initial excess returns on the first listing day was
33.50% in the Sri Lanka. In the Nigerian market, Adjasi et al., (2011) investigate the
underpricing anomaly using the sample of 77 IPOs listed during the 1990-2006 period
and found that the underpricing of 43.10% at the first trading day. In the Malaysian
primary market, Abubakar and Uzaki (2012) reported the initial underpricing of
35.87% using the sample of 476 IPOs listed from 2000 to 2011. Deb (2009)
investigates the relationship between, (1) the initial underpricing and the ex-ante
uncertainty as compared with the Beatty and Ritter (1986), and (2) the initial
underpricing and the ex-post uncertainty as compared with Ritter (1984). He found a
strong positive association of both uncertainties with the initial underpricing.
47
Table 2. 2: IPOs Initial Excess Returns Studies From International Literature
Sources: Table in Loughran, Ritter, and Rydqvist (2010)
48
Lamberto and Rath (2010) examined the effect of prospectus information on
the survival of Australian IPOs listed during the 1995-1997 period. Their results show
the positive association between the survival of IPO firms with predicted dividend
yield and the IPO offer size whereas, negative association with the ex-post
uncertainty. Their results reveal that the IPO osffer size and the predicted dividend
yield are significant determinants for the survival of IPO firms. Islam et al., (2010)
studied the relationship of the underpricing and its driving factors by undertaking the
sample of 191 IPOs of Bangladesh listed during the 1995-2005 period. Their findings
reveal the positive association of the initial returns with the firm age and firm size,
whereas, negative association with the offer size and industry type while premarket
sentiments is an insignificant factor.
Datar and Mao (2006) reported the highest degree of underpricing in the
China. They investigate the initial returns of the Chinese market by taking the sample
of 226 IPO firms listed during the 1990-1996 period and found that firms were on
average underpriced at 388%. Hoque and Mousa (2001) also presented the highest
level of initial excess returns in the Bangladesh. They examined the initial
underpricing using the sample of 113 IPOs data listed from 1984 to 2001 and found
that the firms were on average underpriced at 285%. Borges (2007) analyzed the
underpricing of IPO firms before and after the 1988 financial crisis using the sample
of 98 newly listed firms. He divides the sample into two sub-samples, by taking 51
IPO firms before 1988 and post-crisis 41 IPO firms. He observed the higher
underpricing of 87.50 % in the pre-crisis period while the lower underpricing of 11.10
% was estimated in the post-crisis period. He also compares the underpricing of IPOs
offered by fixed-price and the bookbuilding mechanisms, the results reveal that the
IPOs launched by bookbuilding mechanism underpriced more than the launched by
the fixed-price mechanism.
By having a look on international literature of initial underpricing, this study
supports the evidence of above 100% of underpricing is observed in the China,
Jordan, Bangladesh, Malaysia and Saudi Arabia, the initial underpricing from 50% to
100% is observed in the India, Greece, Thailand, Brazil, Portugal and Korea. The
table also documented the lowest underpricing such as less than 20% observed in the
Hong Kong, Russia, Austria, Egypt, Israel, Norway, Canada, Chile, Argentina, US &
Denmark.
49
2.2.7 Underpricing in Pakistan
An extant literature on Pakistani IPOs, only a few researchers try to explore
the initial underpricing and its predicting forces. The underpricing phenomenon has
been unfolded by various researchers after the initiative of Sohail & Nasr (2007) and
Rizwan & Khan (2007).
Table 2.3 presents the overall look of IPO literature regarding the initial
excess returns in the Pakistan by undertaking various methodologies, sample sizes and
the sample periods. The major issue of Pakistani capital market is the slow pace of
new listings on the PSX market as compared to the high pace of new offerings in the
same regional exchanges such as the India, Bangladesh, Malaysia, Saudi Arabia and
Bangladesh.
Sohail and Nasr (2007) first time in Pakistan examined the aftermarket
performance of IPOs and dynamics of their initial returns by taking the sample of 50
IPOs floated during 2000-2006. They used the methodology as comparable with
Aggarwal, Leal & Hernandez (1993) for post-issue price performance and found that
the Pakistani IPOs are on average underpriced by 35.66%. They also undertook the
regression models in order to understand the magnitude and driving factors of
underpricing. They found that the oversubscription at the time of launching IPOs, the
market capitalization of IPO firm at a 5th trading day and ex-ante uncertainty are
statistically significant determinants and positive relationship with the underpricing
while the offer size has a negative association with the level of underpricing and
confirm the asymmetry information hypotheses. On the other side, market volatility,
the proportion of shares offered and the P/E ratio have a little explanatory power of
underpricing. Sohail and Raheman (2009) broaden their analysis in the context of
aftermarket performance comparison between the financial sector and non-financial
sector IPO firms consistent with the Sohail and Nasr (2007). The sample contains 25
IPO firms from financial and non-financial sectors each. The findings reveal that the
non-financial firms were slightly more underpriced than the financial firms and their
explanatory variables were also different from each other. They reported that the
financial and non-financial sector firms were on average underpriced by 34.52% and
36.8% respectively. They also documented the reasons of initial return and found that
50
the oversubscription, offer size, ex-ante uncertainty and the ex-post market
capitalization on the early trading days has more explanatory power.
Table 2. 3: IPOs Initial Excess Returns Studies From Pakistani Literature
Sr. No Source Sample Size Sample Period Underpricing (%)
1 Sohail & Nasr (2007) 50 2000 - 2006 35.66
2 Rizwan & Khan (2007) 35 2000 - 2006 36.48
3 Javid (2009) 50 2005 - 2006 32.45
4 Sohail & Raheman (2009) 50 2000 - 2006 35.66
5 Sohail & Raheman (2010) 73 2000 - 2009 42.17
6 Kayani & Amjad (2011) 59 2000 - 2010 39.87
7 Afza, Yousaf & Alam (2013) 55 2000 - 2011 28.03
8 Mumtaz & Ahmed (2014) 75 2000 - 2011 30.30
9 Usman. (2014) 55 2001 - 2012 29.14
10 Yar & Javid (2014) 59 2000 - 2012 51.57
11 Kafayat & Rafey (2014) 30 2006 - 2013 68.22
12 Waseemullah &Sohail (2015) 26 2002 - 2011 35.49
13 Mumtaz, Smith & Ahmed (2016) 80 2000 - 2013 22.08
14 Mumtaz & Ahmed (2016) 90 1995 - 2010 15.27
15 Javid & Malik (2016) 72 2000 - 2015 23.32
16 Mumtaz, Smith & Ahmed (2016) 57 2000 - 2010 31.96
17 Sohail, Bilal, Rukh & Fatima (2018) 26 2010 - 2015 3.52
Sources: Compiled by author from various Pakistani studies
Rizwan and Khan (2007) examined and compared the aftermarket
performance of public vs. private sector IPOs and their explanatory components. They
used the sample of 35 IPOs listed during the 2000-2006 period and the sample was
consisted of 7 public sector IPO firms and 28 private sector IPO firms. They used the
methodology as comparable with the Asussenegg (2000) to estimate the short-run
aftermarket performance and found that the public sector IPOs were more underpriced
as compared to the private sector IPOs. The results of their finding of the
underpricing of the overall sample, the privatized firms and the private firms were
36.48%, 74.33% and 26.66% respectively. They also find that the proportion of shares
offered in an IPO and offer size are positive and significant determinants of the level
of initial underpricing. They argue that the Government of Pakistan provides support
to privatization commitments, offer benefits to small investors and to promote capital
market activities through the selling of large and well-renowned state-owned
enterprises at lower prices.
Sohail and Raheman (2010) investigate the initial returns on the 1st trading
day, 5th trading day, 10th trading day, 15th trading day & 20th trading day under the
51
normal, boom & the recession state of economies. They took the sample of 73 IPOs
listed on KSE from 2000 to 2009 period. They used the market adjusted return model
to estimate the short-run performance and found that the initial returns on the first
trading day in the overall sample period, normal, boom & the recession state of the
economy were 42.17%, 36.75%, 55.19% & 32% respectively. They used wealth
relative model as a sensitivity analysis & a robustness measure to compare the results
with the market adjusted returns method. The results of initial underpricing by wealth
relative model are consistent with the market adjusted returns. They also perform the
sectoral analysis and found that the Chemical, Engineering & the Oil and Gas
marketing IPO firms underpriced more than 100% on the first trading day. They argue
that the investors can earn abnormal profits if they buy stocks in the IPO process and
sell them on the first trading day closing prices.
Kayani and Amjad (2011) investigate the association between the level of
underpricing, before IPO investor’s interest and post-IPO trading volume using the
sample of 59 IPOs listed during the 2000-2010 period on KSE and provides the
evidence of initial underpricing of 39%. The number of shares subscribed by the
investors over the number of shares offered in the IPO ratio and the daily trading
volume over the number of shares offered in the IPO ratio was used as a proxy for
pre-IPO investor’s interest and post-IPO investor’s interest respectively. Their results
reveal that the initial returns are positively linked with the higher subscription ratio
and higher trading volumes over the few early trading days. They also used the cross-
sectional regression analysis to determine the reasons of the initial underpricing and
found that the firm size, offer size, Proportion of shares offered in the IPO ratio and
the ex-ante uncertainty were major predictors of the initial underpricing, whereas their
findings were consistent with the preceding studies of underpricing.
Afza, Yousaf and Alam (2013) examined the impact of information
asymmetry and the corporate governance (i-e the ownership structure and the board
composition) on the level of underpricing using the sample of 55 IPO firms who went
public from 2000 to 2011 period. This study also investigated the impact of corporate
governance as a moderating factor in the relationship between the initial underpricing
and the information asymmetry. They found that the Pakistani IPOs were underpriced
of 28.03% and corporate governance determinants add value to control the magnitude
of underpricing. They also documented that the information asymmetry is positively
52
linked with the degree of initial underpricing. They argue that CEO duality and the
institutional investment have an effect on the degree of initial excess returns.
Mumtaz and Ahmed (2014) employs the methodology as comparable with the
Sohail & Raheman (2010) and Ljungqvist et al., (2006) to estimate the initial
underpricing of 75 IPO firms listed during 2000-2011 period and found that the IPOs
underpriced on the first trading day and the thirtieth trading day was 30.30% and
24.17% respectively. They also used Extreme Bound Analysis (EBA) as comparable
to Levine & Renelt (1992) to examine the sensitivity and the robustness analysis of
the independent factors of initial underpricing. They found that the oversubscription,
ex-post uncertainty, financial leverage and the IPO offer prices were the true
predictors of the IPO underpricing.
Yar and Javid (2014) enlighten the underpricing phenomenon through the
relationship of initial returns, aftermarket liquidity and the ownership structure using
the cross-sectional data of 59 newly listed firms on KSE during the 2000-2012
periods. They found that the IPOs underpricing on the first trading day was 57.57%
and employ the methodology of logit model initially formed by Amemiya (1981) to
estimate the determinants of underpricing. They also employ the methodology as
comparable with the Booth and Chua (1996) to diffuse the ownership effect on the
IPO underpricing. They investigated that the M/B ratio, oversubscription, and the total
assets were the significant determinants of initial adjusted returns. They also expand
their results and found that the underpricing and the total assets were the significant
factors of the ownership structure. They also investigated that the underpricing, firm
size, ex-post uncertainty and the oversubscription were the major significant
determinants of aftermarket liquidity.
2.3 The IPO Long-run performance
The long-run poor performance is another anomaly observed in the domain of
public equity aftermarket performance analysis. Aggarwal and Rivoli (1990) first time
uncover the confirmation of underperformance over the longer periods. They called
this underperformance stance is a fad. The findings advocate that if private equities
methodically overvalued in the early trading days, market participants who acquired
shares on first trading day closing prices and hold them for longer time horizons
underperform the market. Ritter (1991) proposed various implications under the
53
findings of this study that why long-run returns analysis is noteworthy in corporate
finance. First, from the market participants’ point of view, the presence of returns
behavior fashions may offer investment alternatives for active trading strategies to get
abnormal returns. Second, the aftermarket return patterns are not constant over the
numerous time horizons and raised the question on the informational efficiency of
IPO product market. Third, the large variation in the volume of IPOs has been
observed over time. Finally, the cost of external equity capital raised by undertaking
the initial public offerings not only depends on the transaction cost take place in IPO
privatization and left money on the table in terms of underpricing. Many other studies
are motivated by Ritter’s study and has been attempting to assess the long-run
performance in many developed and emerging markets. The proposed hypotheses and
an extant literature have conversed below.
2.3.1 Fad Hypothesis
Aggarwal and Rivoli (1990) initially proposed this hypothesis based on the
evidence of long-run underperformance. They unable to defined the implications of
this phenomenon called as a fad in the primary market. Ritter (1991) investigates the
IPO valuation and the aftermarket long-run performance using the sample of 1,526
US newly listed firms during 1975-1984. They evaluate their three years aftermarket
performance and then compare the performance of firms from the similar industry and
market capitalization. This study finds that mostly firms went public when share
prices of comparable firms were at peak. He also observes negative returns as
compared to investing in comparable firms from the similar group based on the
industry and the capitalization listed in the US market. Therefore, it can be concluded
that the IPO is a profitable opportunity only if invest on the floatation prices and
release on the early days of listing. He suggests three different implications for long-
run underperformance, namely; the risk mis-measurement, the bad-luck and fads.
Though, the pragmatic finding does not hold the risk-measurement and the bad-luck
explanations. He presents the evidence of hot issues as firms decide to go public when
market sentiments are bullish and investors willing to pay higher prices. The IPO
share prices adjust their equilibrium as more information turns out to be available
publicly.
54
Loughran and Ritter (1995) motivated and broaden the Ritter’s study and
argue that firms went public when they observe firms from similar industry trading at
a high P/E and P/BV multiples. This results in the positive bias on investor initial IPO
valuations about upcoming offerings. The investors took valuation on the basis of
rapid growth in the trailing earnings which often disappear after the offering. They
advocate that the rational investors cannot beat the other investors’ over-valuations
about new offerings. They suggest that when underwriters leave the control of IPO
share prices to market forces then the share prices adjust their origin and this results in
underperformance of IPOs.
Lowery (2003) concludes that long-run underperformance of hot market IPOs
is sensitive to test specification by using the IPOs data of US floated firms during
1973-1996. She estimates the abnormal returns of IPOs as comparable with matched
size and BTM portfolio benchmarks. She observed that the negative association
between the IPO activity and long-run performance is strongest when using raw
returns, and weakest when using matched-size and BTM portfolio benchmarks. She
finds that the issuer firms more likely to go public when comparable valuation
multiples are high.
In sum, the long-run underperformance entails that the total cost of external
equity capital raised through public offerings is not excessive for private equity
issuers.10 The more established firms face a larger cost of external equity capital than
the smaller firms who want to expand their operations considered to be a risky
investment.
2.3.2 Heterogeneous expectations hypothesis
This hypothesis was initially proposed by (Miller, 1977). He theoretically
enlightens the unwind supposition of homogeneous expectations of investors;
therefore, a divergence of opinion arises among the market participants in the IPO
market. This study also argues that the short selling is restricted in the IPO shares;
hence the prices are defined by the optimistic investors. In the long run period, IPO
shares shifted in the normal regime of market regulations and more information
10 The major part of the transaction cost of raising equity capital is offset by the realized long run underperformance, for those firms who decide to go public when share prices of the comparable firms are at peak and the investors are optimistic about the future prospects.
55
become available then IPO share prices adjust their equilibrium. Therefore, this study
proposed the conjecture that the initial overvaluation is translated due to divergence of
opinion among initial investors and as a result long-run underperformance.
Houge et al., (2001) investigate the relationship between the degree of
variation of opinion among investors and the long-run performance of 2,025 newly
listed firms in the US from 1993 to 1996. They employ the time of the first trade, the
%opening bid-ask spread, and the flipping ratio as a proxy for the divergence of
opinion among investors perceived as the uncertainty bear by the number of IPO
market makers and tend to lead to opinion divergence. They find that the larger bid-
ask spread, long delay of first trade and the larger flipping activity impact the long-
run poor performance over three years. They conclude that the extent of uncertainty
faced by the IPO market makers, significantly influence the long-run poor
performance.
2.3.3 Agency Hypothesis
Carter et al., (1998) used several proxies to investigate the underwriter
reputation using the sample of US IPO firms listed during 1979-1991. They employ
the regression model to explain the explanatory power of underwriter reputation by
taking initial returns as a dependent variable and the long-run performance as a
dependent variable in the second model. They extract underwriter reputation measures
from (Johnson and Miller, 1988; Megginson and Weiss, 1991; Carter and Manaster,
1990) in both models individually as well as all together. The results of the initial
return model, each underwriter reputation measure are significantly linked with the
initial returns. On the other side, only Carter et al., reputation measure stay significant
when estimated at the same time. The results of long-run returns analysis, produce
relatively less poor that bring by prestigious underwriters. They conclude that the
long-run underperformance is directly linked to the extent of underwriter reputation.
Logue et al., (2002) find that despite the impact of underwriter reputation, the
underwriter activities in the process of IPO are significantly bonded with the IPO
long-run returns.
Brav and Gompers (1997) also investigate the role of agent regarding the long-
run underperformance of 934 venture-backed IPOs and 3,407 non-venture-backed
IPOs listed from 1972 to 1992. They document that the venture-back IPO firms
56
outperform the market while non-venture-backed IPOs returns remain constant over
time in the long run. They extend their analysis by undertaking numerous comparable
firms’ benchmarks, and the Fama and French 3 factor model. They point out that the
venture-backed IPOs do not negatively perform in the long run while non-venture-
backed smaller IPOs do underperform. On the other side, the negative performance in
long-run is not due to as observed that the comparable non-IPO firms from the same
size and the book value-to-market price ratio also underperform as the IPO firms.
In sum, the venture-capitalists and lead underwriters play an important role in
the IPO valuation oversight by initial investors and also affect the IPO valuation in the
long run.
2.3.4 Signaling Hypothesis
As discussed in the earlier sections, the signaling hypothesis exhibits that the
underwriters and the issuers produce some signals to the investors before the formal
listing in the market to reveal the fair value of IPO shares. Even it is inevitable to
define the underpricing anomaly during the initial trading days, there are some
implications for long-run performance as well.
As the signaling hypothesis suppose that the initial public offerings are
followed by seasoned equity offerings. Jegadeesh et al., (1993) demonstrates several
implications of the long-run performance of newly listed firms. First, they argue that
the firms raising additional equity capital after the fund raised in the IPOs are high
value; therefore these firms outperform the market over long period. Second, firms
having abnormal returns in initial trading days outperform the market in the long run
as well. Third, the quality IPOs hold a higher proportion of shares in early days and
leave a more money for the initial investors in the market to perform better in the long
run. Welch (1989) investigates the relationship between the initial returns and the
long-run performance. He finds that the firms offer abnormal returns in the early
trading days outperform the market of non-issuing firms. Ritter (1991), Ljunqvist
(1996) and, Jain and Kini (1994) find that the firms who produce abnormal returns in
the initial days do not demonstrate superior post-listing returns, in the long run,
relative to those that do not. Koh et al., (1996) investigate the relationship between the
shares retained by old shareholders and the aftermarket long-term returns using the
data of newly listed firms in Singapore. They find that firms having a major portion of
57
shares retained by old shareholders outperform the market in the long run. These
results are inconsistent with the findings of Ljungqvist (1996).
2.3.5 An International Empirical Evidence
Ritter (1991) used a sample of 1,526 US IPO firms listed during 1975-1984
and found a significant underperformance in the long run. This study used CAR by
adjusting the size portfolio as the benchmark and found the underperformance of (-
10.20%) and (-29.10%) for one- and three-years respectively. In addition, he swapped
the benchmark with NYSE index and found the presence of muscular
underperformance using the same sample. He argued that the small and medium
enterprises experience more underperformance than the larger enterprises. Loughran
(1993) used the data of US IPO firms and the Nasdaq index used as a benchmark to
estimate the long-run abnormal returns. He found a significant underperformance of (-
60.00%). Levis (1993) first time investigate the long-run performance in the UK using
a sample of 712 IPOs listed from 1980 to 1988. He points out the value of size-effect
and presents the long-run adjusted returns on the basis of Hoare Govett Smaller
Companies (HGSC) index, Financial Times Actuaries (FTA) index and All Shares
Equally-weighted Index as benchmarks. He finds the post-IPO underperformance
over three years of between (-8.0%) and (-23.0%) depending on the benchmark used.
Loughran and Ritter (1995) examined the long-run underperformance using
the sample of 4,753 US IPOs listed during the 1970-1990 period. They find that the
underperformance was observed (-26.90%) over three years while the
underperformance has increased to (-50.00%) when estimated for five years. They
point out that the long-run performance was worse during the high IPO activity
periods as compared with the low IPO activity periods. Hwang and Jayaraman (1995)
estimate the long-run IPOs performance of 182 Japanese newly listed firms and
conclude that the equal-weighted CAR is significantly negative (-14.98%), however,
at the same time value-weighted CAR is significantly positive (16.44%).
Barber and Lyon (1997) examine the long-run performance using the CAR
and BHAR methodologies of same IPOs data. They point out that the results of long-
run performance are not in line with both techniques. They preferred the BHAR
approach by justifying the cause that CAR method does not watch the investment
strategy of market participants if the securities held for longer time periods. On the
58
other hand, they also criticized the BHAR method due to skewness problem when
compounded abnormal returns are estimated on the monthly basis. To cope the issue
of skewness, Barber, Lyon and Tsai (1999) suggested the skewness adjusted model to
estimate the long-run abnormal returns.
Espenlaub et al., (2000) re-investigate the longer period returns of the UK
newly listed firms during 1985-1995 and they compare the long-run performance on
the basis of several methods: the CAPM, Fama-French three-factor, size control
portfolio, Ibbotson Returns Across Securities and Times (RTA) and the value-
weighted multi-index using the HGSC index. They found the long-run
underperformance over five years for CAPM, Fama and French three factors, size
control portfolio and the RATS while insignificant abnormal returns were found when
using the HGSC index as a benchmark.
Ritter and Welch (2002) estimate the long-run performance using the two
different alternatives: market and matched firms benchmarks (market to book ratio
and market capitalization). They estimate the long-run underperformance by -23.40%
when the market index was used as a benchmark while matching firms’ benchmark
produces underperformance -5.10%.
In the study of Gompers and Lerner (2003), measure the long-run performance
over five years of 3,661 US IPOs floating during 1935-1972 and compare the long-
run performance based on the different methods: BHAR, CAR, CAPM and Fama-
French three factor model. They found the long-run underperformance when BHAR
on the value-weighted basis was applied. They also observed that the IPOs
outperformed over five years when BHAR and CAR on the equally-weighted basis
were applied. Furthermore, no underperformance was examined by the CAPM and
the Fama-French methodologies.
Kooli and Suret (2004) also confirmed the long-run underperformance in the
Canadian IPO product market. They investigate the long-run abnormal returns over
three to five years by selecting the sample of 445 IPO firms listed from 1991 to 1998
and non-issuing matched firms used as a benchmark. They documented that the non-
issuing matched firms adjusted performance of -19.96% and -26.5% over the period
of three- and five-year respectively. Eckbo and Norli (2005) used Fama-French model
with modification by selecting a rolling portfolio strategy and found variations in the
59
long-run underperformance. The Jensen’s alpha explored to be insignificant to dictate
the long-run underperformance by adjusting the risk factors of value, size and market.
In the Malaysia, Ahmad-Zaluki (2007) select the sample of 454 newly listed
firms during 1990-2000 on the main board and the second board. They investigate the
long-run abnormal returns using the BHAR, CAR and the Fama-French three factor
model. They observed that the IPOs outperform the market under the approach of
BHAR and CAR. Though, the Fama-French three factor model produces negative
abnormal returns.
In the UK, Gregory et al., (2009) discussed the problems with various
benchmarks to examine the long run abnormal performance by choosing the large
sample of 2,499 UK IPOs floating during the 1975-2004 period. They found the
underperformance of -12.60% when a value-weighted portfolio is used as a
benchmark for the period of three years. They also observed that the
underperformance increased by -31.60% when an equally-weighted portfolio is used
as a benchmark for the period of five years.
Govindasamy (2010) documented the long-run underperformance using the
sample of 229 South Africa IPOs floating during 1995-2006 and used All Shares
Index of JSE as a benchmark. He found the long-run underperformance of -50.0%
after the period of 36 months. These results are in line with the other emerging market
literature. In India, Sahoo and Rajib (2010) document the long run abnormal returns
using the sample of 92 IPO firms listed during the 2004-2006 period on the Bombay
and National Stock Exchanges. Their results of long-run underperformance were not
in line with other Asian countries like the Sri Lanka, Bangladesh and the Pakistan.
Gopalaswamy et al., (2008) investigate the long-run performance using the data of
fixed price issues and the Bookbuilding issues. They conclude that the IPOs issued by
Bookbuilding process perform better after the one, two and three years than the IPOs
issued by Fixed price method.
Anton et al., (2011) examined the aftermarket pricing performance using the
data of Spanish IPOs listed during the period of 2000-2010 in the short and medium
terms. They point out that the Spanish IPOs outperform the market in the short-run
whereas the medium term performance was observed to be worse.
60
Table 2. 4: IPOs Long-run Performance Studies From International Literature
Authors Country Sample
Period
Sample
Size
Abnormal
Returns-%
Underperformance
for the period
Jewartowski and Lizinska (2012) Poland 1998-2008 142 -22.62 36 – Months
Komenkul et al., (2012) Thailand 2001-2012 136 -16.60 36 – Months
Belghitar and Dixon (2012) UK 1992-1996 335 -14.0 36 – Months
Bossin and Sentis (2014) France 1991-2005 207 -28.85 36 – Months
Islam et al., (2012) Bangladesh 1992-2006 163 -38.4 44 – Months
Thomadakis, Nounis and
Gounopoulos (2012)
Greece 1994-2002 254 -16.12 36 – Months
Brau (2012) US 1985-2003 3,547 -17.10 36 – Months
Su et al., (2011) China 1996-2005 936 8.6 36 – Months
Sahoo and Rajib (2010) India 2002-2006 92 41.91 36 – Months
Govindasamy (2010) South Africa 1995-2006 229 -50.00 36 – Months
Chi, Wang and Young (2010) China 1996-2002 897 9.69 36 – Months
Chi, McWha and Young (2010) New Zealand 1991-2005 114 -27.81 36 – Months
Chorruk and Worthington (2010) Thailand 1997-2008 141 -25.39 36 – Months
Gregory et al., (2009) UK 1975-2004 2499 -12.60 36 – Months
Goergen, Khurshed and
Mudambi (2007)
UK 1991-1995 240 -21.98 36 – Months
Zaluki et al., (2007) Malaysia 1990-2000 454 0.04 36 – Months
Campbell and Goodacre (2007) Malaysia 1990-2000 454 -2.01 36 – Months
Drobetz, Kammernmann and
Walchli (2005)
Switzerland 1983-2000 53 -173.46 120 – Months
Kooli and Suret (2004) Canada 1991-1998 445 -20.70 60 – Months
Gompers and Lerner (2003) US 1935-1972 3,661 -33.40 60 – Months
Ritter and Welch (2002) US 1980-2001 6,249 -23.40 36 – Months
Espenlaub, Gregory and Tonks
(2000)
UK 1985-1992 588 -21.30 60 – Months
Stehle et al., (2000) Germany 1960-1992 562 -9.01 36 – Months
Allen, Morkel-Kingsbury and
Piboonthanakiat (1999)
Thailand 1985-1992 143 10.02 36 – Months
Hwang and Jayaraman (1995) Japan 182 -14.98 36 – Months
Loughran and Ritter (1995) US 1970-1990 4,753 -50.0 60 – Months
Levis (1993) UK 1980-1988 712 -22.96 36 – Months
Ritter (1991) US 1975-1984 1,526 -29.10 36 – Months
Source: Compiled from various international articles
From the literature of China, Su et al., (2011) found the positive abnormal
returns by selecting the sample of 936 IPOs listed during the 1996-2005 period on the
Shenzhen and Shanghai Stock Exchanges using the matched firms as a benchmark.
The results reveal that the matched firms’ adjusted abnormal returns outperform the
market by 4.60% over the period of 24 months and these returns increases to 8.60%
over the period of 36 months. They concluded the misspecification of evaluation
61
methods with respect to the choice of benchmarks and confirmed the previous studies
of misspecification of the model.
Table 2. 5: IPOs Long-run performance Studies using the Event-Time Approach
Author(s) Country Period Sample
Size
Methodology Abnormal Returns over -%
1-Year 2-Year 3-Year 5-Year
Aggarwal, Leal and
Hernandez (1993)
Chile 1982-1990 36 MAR +1.1 -2 -23.7 --
Aggarwal, Leal and
Hernandez (1993)
Mexico 1987-1990 44 MAR -19.6 -- -- --
Aggarwal, Leal and
Hernandez (1993)
Brazil 1980-1990 62 MAR -9 -34.9** -47.0** --
Levis (1993) UK 1980-1988 712 CAR VW +1.6 -5.2 -11.4 --
CAR EW -7.2*** -17.3*** -23.0*** --
Lee, Taylor and Walter
(1996)
Australia 1976-1989 266 CAR -13.5 -31.0*** -51.3*** -30.9***
Allen, Morkel-Kingsbury
and Piboonthanakiat
(1999)
Thailand 1985-1992 150 CAR EW -5.4 -10.5 +10.0 --
CAR VW +1.4 +2.2 +27.5 --
Jakobsen and Sorensen
(2000)
Denmark 1984-1992 76 BHAR market -1.8 -23.5*** -30.4***
BHAR matching -6.6 -21.6*** -13.1*
Stehle, Ehrhardt and
Przyborowsky (2000)
Germany 1960-1992 187 BHAR VW +1.2 -7.0** -5.0
BHAR EW +2.4 -4.4 +1.5
Arosio, Giudici and
Paleari (2001)
Italy 1985-1999 150 BHAR -7.5 -12.5** -11.5
Lyn and Zychowicz
(2003)
Hungry 1991-1998 33 MAR -3.3 +1.2 -4.9
Lyn and Zychowicz
(2003)
Poland 1991-1998 108 MAR -4.1 +3.4 -24.4
Kooli and Suret (2004) Canada 1991-1998 445 CAR EW -10.8** -12.4 -16.9 -25.7
CAR VW -6.84** -8.7*** -9.4*** -19.2***
Alvarez and Gonzalez
(2005)
Spain 1987-1997 52 BHAR +6.1 -- -28.2*** -21.0**
Drobetz, Kammermann
and Walchli (2005)
Switzerland 1983-2000 109 BHAR -2.1 +0.9 -1.7 -26.2
CAR -8.5* -8.3 -7.5 -31.2***
Cheng and Shiu (2005) Taiwan 1988-2002 917 BHAR EW -3.4*
BHAR VW -10.0***
CAR EW -9.4*
CAR VW -22.7***
Bildik and Yilmaz (2007) Turkey 1990-2000 244 CAR EW +2.3 -13.0 -84.5
CAR VW +0.7 -4.0 -24.0
Thomadakis, Nounis and
Gounopoulos (2007)
Greece 1994-2002 254 MAR 15.7*** -8.1** -31.4**
Ahmad-Zaluki, Campbell
and Goodacre (2007)
Malaysia 1990-2000 454 CAR small EW +5.4 +2.5 +0.4
CAR small VW +0.04 -5.8 -8.2
BHAR mkt EW 11.6*** +21.1*** +17.9***
BHAR mkt VW +1.7 +3.9 -14.2***
***Significant at the 1% level, **Significant at the 5% level and *Significant at the 10% level
Source: Compiled from Choi, Lee and Megginson (2010)
In the developed market of France, Bossin and Sentis (2014) noted the long-
run underperformance by selecting the sample of 207 France IPO firms listed from
1991 to 2005 using the size and book to market as benchmarks. They documented that
62
the long-run adjusted performance of -28.85% and -68.10% under the benchmarks of
size and book to market respectively.
In Thailand, Komenkul et al., (2012) selected 136 Thailand IPOs listed from
2000 to 2012 and support the presence of long-run underperformance using the event-
study methodologies. They found the underperformance of -16.60% and -19.60%
using the BHAR and CAR respectively. In the study of Bangladesh, Islam et al.,
(2012) used a sample of 163 IPOs listed from 1992-2006 and found the long-run
underperformance of 38.40% at the end of the 44th month as compared with the
industry benchmark indices. In addition, they used size as a benchmark and found the
severe long-run underperformance in the smaller issues than the larger issues.
Bossin and Sentis (2014) examined the long-run performance of French IPOs
floated during 1991-2005. They categorized the sample into two groups as (1) orphan
IPOs (without carrying recommendations from the financial analysts) and, (2) non-
orphan IPOs (carrying recommendations from the financial analysts). They used both
the event-study and calendar-time approaches to estimate the long-run performance of
both clusters. They observed poor performance by the both types of IPOs in the long
run relative to the market portfolio during the sample period. They argue that the
financial analysts’ recommendations are significant for the initial year 1, however, for
the year 3 to the year 5, analysts’ recommendation unable to drive the long-run
performance of IPOs.
This draft finds only one study, which used Fama-French 5-factor model (the
most recent multifactor asset pricing model proposed by Fama & French (2015)) to
estimate the long-run performance of IPOs. In Sri Lanka, Ediriwickrama and Azeez
(2016) investigate the IPOs long-run underperformance anomaly using the numerous
asset pricing models (known as calendar time approaches). They employ Sharpe-
Lintner CAPM, Zero Beta CAPM, Fama-French 3-factor model, Carhart 4-factor
model and Fama-French 5-factor model to investigate the long-run performance of 51
IPOs floated in the Colombo Stock Exchange during 2000-2012. Based on the
constant coefficient (Jensen’s Alpha), they found that the IPO portfolios
underperform in the long run as compare to market benchmarks using the value- and
equally-weighted methods. They argue that the value-weighted models are more
63
appropriate to jointly explain the variation in the IPO long-run performance than the
equally-weighted models.
2.3.6 Difficulties with Long-term Returns Measurement
An accurate measure of long-term returns and the variation in the aftermarket
performance for longer time horizons may be attributable to several reasons. The
initial literature on long-run performance demonstrates that there are two methods to
estimate the long-run performance; the Cumulative Abnormal Returns (CAR) and the
Buy-and-Hold Abnormal Returns (BHAR).
The first issue with long-run performance measurement is the biasness. Barber
and Lyon (1996) investigate the accuracy of both CAR and BHAR methods to
measure long-term returns. They find that the CAR methodology undergoes from the
measurement bias because CAR is a biased predictor for BHAR. As a result, they
support the use of the BHAR to assess long-term abnormal returns. Fama (1998)
findings varied to Barber and Lyon (1996) and favored the cumulative abnormal
return model to estimate the long-run performance of IPOs. They point out that the
CAR is easy to scrutinize the linearity behavior of averages in the longer time
horizons. He preferred the CAR model which is a good estimator for multi-periods as
the CAR averages amplify linearly and the standard-error amplify with the square
root. At the same time, he criticized the buy-hold abnormal returns model is not in
shape to estimate multi-period returns because BHAR grows exponentially instead
linearly that finally raise the measurement problem.
The second issue with the long-run performance measurement is the selection
of market benchmarks. An Extant literature presents three different benchmarks to
estimate the long-run performance; the market indices, comparable firms from the
similar industry and the Fama-French three factor model. Barber and Lyon (1997)
identify few biases with using of market indices as the benchmark are; rebalancing
bias, skewness bias and newly listed firms’ bias. They suggest that the portfolio of
comparable firms selected on the basis of firm-specific characteristics is a good
benchmark. Whereas, Kothari and Warner (1997) disagree with the use of the
portfolio of comparable firms as a benchmark referred to as pre-event survivorship
bias. Ritter and Welch (2002) examine the accuracy of long-run performance using
two benchmarks; market and matched firms. They find that the market adjusted long-
64
run returns after three years was assessed to be -23.40%, while long-run performance
using the matched firms was observed to be -5.1%. Gompers and Lerner (2003) used
Fama-French three-factor model as a benchmark to estimate the long-run performance
and their results support the evidence of long-run underperformance. Eckbo and Norli
(2005) using the Fama-French three-factor model and Jenson’s alpha to estimate the
long-term performance. They point out that the results are not constant for longer time
horizons when they used Fama-French model with the modification of rolling
portfolio strategy. This study also finds that the Jenson’s alpha was insignificant to
demonstrate the long-run underperformance after adjusting the risk factors of value,
size and the market. Choi, Lee and Megginson (2010) also focused on the
methodology problem for long-run performance using the sample of 241 IPOs from
42 countries using the matched firms based on the size, book-to-market and the Fama-
French three-factor model benchmarks to assess the long-term returns. They find that
the IPOs under the Fama-French methodology is significant and outperform the
market in the several longer time horizons. However, the average returns of BHAR by
adjusting the size and the book-to-market multiples are not statistically significant and
outperform the market as well.
Lastly, few researchers also talk about the power of statistical tests used for
the long-term performance analysis. Loughran and Ritter (1995), and Brav (2000)
argue that the test statistics absence from the independence of observations as the
long-run returns of IPO firms may be correlated in the calendar time. Jenkinson and
Ljunqvist (2001) used the sample of internet IPOs listed during the bubble period and
verify this stance by showing the reduction in cross-sectional variance in the long-run
abnormal returns.
65
Table 2. 6: IPOs Long-run performance Studies using Calendar-Time Approach
Author(s) Country Period Methods
Espenlaub, Gregory, & Tonks
(2000)
UK 1985-1992 CAR & FF3F
Gompers and Lerner (2003 USA 1935-1972 CAR, BHAR, CAPM
& FF3F
De Silva Rosa, Velayuthen &
Walter (2003)
Australia 1991-1999 WR, BHAR & FF3F
Boabang (2005) Canada 1990-2000 CAR & FF3F
Ahmad-Zalukai, Campbell &
Goodacre (2007)
Malaysia 1990-2000 CAR, BHAR &
FF3F
Pukthuanthong-Le & Varaiya
(2007(
USA 1993-2002 BHAR &FF3F
Chi, Wang & Yong (2010) China 1996-2002 CAR, BHAR &
FF3F
Moshirian, Ng & Wu (2010) China, Malaysia, Japan,
Hongkong, Korea, Singapore
1991-2004 BHAR &FF3F
How, Ngo & Verhoeven (2011) Australia 1992-2004 CAR, BHAR &
FF3F
Brau, Couch & Sutton (2012) USA 1985-2003 BHAR &FF3F
Thomadakis, Nounis, &
Gounopoulos (2012)
Greece 1994-2002 CAPM, FF3F & C4F
Liu, Uchida & Gao (2012) China 2000-2007 WR, BHAR & FF3F
Boissin, & Sentis (2014) France 1991-2005 BHAR & FF3F
Mumtaz, Smith & Ahmed (2016) Pakistan 2000-2010 CAR, BHAR &FF3F
Ediriwickrama, & Azeez (2016) Sri Lanka 2000-2012 CAPM, Cohart4F,
FF3F & FF5F
Ediriwickrama, & Azeez (2017) Sri Lanka 2003-2015 FF3F, proposed 4-
Factor model
Source: compiled from various international articles
2.3.7 Long-run IPOs Performance in Pakistan
An extant literature on Pakistani IPOs, only a few researchers has try to unfold
the long-run underperformance anomaly for three to five years after the initiative of
Sohail & Nasr (2007) and Rizwan & Khan (2007).
Table 2.7 presents the summary of long-run underperformance of Pakistani
IPOs gauged by various researchers. Sohail and Nasr (2007) evaluate the one-year
pricing performance of 50 newly listed firms during 2000-2006 and document the
market adjusted returns using the CARs and BHARs by -19.67% and -38.10%
respectively. Rizwan and Khan (2007) estimate the long-run performance of 35
privatization and the private-owned IPOs from 2000 to 2006 period. They find that
the market adjusted returns using the buy-and-hold returns for one and two years are -
11.26% and -23.68% respectively. They also document that the aftermarket two-year
66
performance of privatization and private-owned IPOs by 12.69% and -33.11%
respectively. They argue that the privatization IPOs outperform the market than the
private-owned IPOs because privatization firms mature enough to overcome the
market uncertainties and perform well in the long run.
Table 2. 7: IPOs Long-run Performance Studies From Pakistani Literature
Authors Sample
Period
Sample
Size Methodology
Abnormal Returns over %
Year1 Year2 Year3 Year5
Sohail & Nasr (2007) 2000-2006 50 BHAR -38.10 -- -- --
CAR -19.67 -- -- --
Rizwan & Khan
(2007)
2000-2006 35 BHAR -11.26 -23.68 -- --
Mumtaz & Smith
(2015) 1995-2008 35 BHAR -- -- -29.50 -69.70
Mumtaz & Ahmed
(2016)
1995-2010 90 CAR VW -24.60 -18.90 -23.40 --
BHAR VW -15.20 -7.40 -19.80 --
CAR EW -19.90 -18.60 -23.20 --
BHAR EW -19.00 -15.30 -24.20 --
FF3F Y3EW -0.068 Y3VW -0.105
Carhart4F Y3EW -0.034 Y3VW -0.118
Javid & Malik (2016) 2000-2015 72 BHAR 12.76 -18.00 -42.49 -65.54
Mumtaz, Smith &
Ahmed (2016)
2000-2010 57 CAR VW -22.80 -19.30 -22.50 --
BHAR VW -11.00 +7.70 7.70 --
CAR EW -27.40 -16.60 -17.90 --
BHAR EW -26.30 -22.90 -32.70 --
Source: compiled from various Pakistani articles
Mumtaz and Smith (2015) examine the long-run performance over three and
five years using the sample of 35 IPOs from 1995 to 2008. They find that the IPO
firms underperform over three and five years using the buy-and-hold abnormal returns
by -29.50% and -69.70% respectively. They exclude the first month returns to avoid
the potential bias from the price adjustment of initial excess returns and once more
estimate the longer period’s performance over three and five years using the BHAR,
IPOs underperform by 22.80% and 61.70% respectively. They argue that the results
are consistent with the existing literature and investment in Pakistani IPOs is not
valuable over longer period horizons.
Mumtaz and Ahmed (2016) estimate the long-run pricing performance using
the BHAR and CARs of sized-based matched firms and the equal-weighted matched
firms. They used the sample of 90 IPOs listed from 1995 to 2010 in the KSE and
support the presence of underperformance phenomenon. The higher percentage of
underperformance is recognized when compared with sized-based matched firms
approach. Javid and Malik (2016) also validate the long-run underperformance of
67
privatization and private-owned firms using the sample of 72 IPOs during 2000-2015.
Mumtaz, Smith and Ahmed (2016) investigate the aftermarket performance of newly
listed firms using the event-study and calendar-time approaches by adjusting the
sized-based matched firms benchmark over three years. They point out that the long-
run underperformance is greater when using the CARs, exemplify significantly
negative abnormal returns. In the Calendar-time approach, the constant coefficient
(Jensen’s Alpha) related to Fama-French 3-factor and Carhart 4-factor models appear
to be negative in the equally- and value-weighted portfolios respectively.
By having a look on Pakistani literature of long-run returns analysis, this study
supports the evidence of long run underperformance is observed in the PSX, the long
run underperformance from -18% to -69% is observed in three to five years using
BHAR (Javid & Malik, 2016; Mumtaz & Smith, 2015). The most Pakistani studies
used BHAR and CAR methodology to estimate long run abnormal return while only
Mumtaz & Ahmed (2016) first time used Fama-French 3-factor and Carhart’s 4-factor
models to validate the long-run underperformance. Only Mumtaz, Smith & Ahmed
(2016) and Javid & Malik (2016) estimate the determinants of long run performance.
They find that financial leverage, offer size, firm beta and percentage of shares
offered are the key driving forces of long run abnormal returns.
68
3. Chapter 3
3. Research Methodology
69
This chapter explains the sample description and research methodology employed to
investigate the choice and accuracy of the IPO valuation methods and the
determinants of aftermarket performance of IPOs listed in Pakistan. A number of
valuation models have been conversing about IPO performance in the extant
literature. The discussion about ex-ante valuation and ex-post performance includes
the theoretical foundation of valuation models, initial excess returns models, and the
long-run performance models. This chapter as a whole describes the variables used in
empirical models and related hypotheses on valuation and aftermarket performance.
Section 3.1 provides the details of sample selection criteria and description of the
sample. Section 3.2 & 3.3 provides details of theoretical basis of research
methodology employed to examine the choice, bias, and accuracy of each valuation
method for fair-value estimates disclosed in prospectus documents. Section 3.4
provides the comprehensive explanation of the valuation methods used in IPO and
non-IPO studies, and formation of multivariate regression models to determine the
factors that impact on offer price valuation. Section 3.5 & 3.6 provide the
methodology and related hypotheses to explore the effect of prospectus information
on the short-run underpricing returns and long-run performance respectively.
3.1 Data and Sample Description
This study starts with all newly listed firms on the Pakistan Stock Exchange
previously known as Karachi Stock Exchange during 2000-2016. A total of 126 firms
went public on PSX during the sample period. As similar to an extant literature, this
study excludes 38 firms from the population as: excludes 16 firms that went public
without publishing prospectus documents because these firms get listed due to Specie
Dividend announced by their parent firms to their existing shareholders, merger and
demerger events, excludes 16 firms listed as closed-end mutual funds because their
reporting environments are not comparable with the other sector IPO firms and six
prospectus documents are missing in the data. These sample selection criteria result as
a sample of 88 (70%) firms that went public. Table 3.1 presents the summary of IPOs
and sampling selection criteria stretches across the years during sample period and
indicates that most IPOs went public during 2004-2008. Table 3.2 presents the
summary of IPOs across the sectors in which they are operating. The findings
highlight that most IPOs offered by investment securities & banks, commercial banks
and power generation & distribution sectors.
70
Table 3. 1: Sample Selection Criteria and Description
Sr. No Year Total Listed
Firms
Restrictions Sample
IPO
Firms Without IPO
Listings
Mutual
Funds
Missing
IPOs
1 2000 3 0 0 0 3
2 2001 3 0 0 1 2
3 2002 4 0 0 0 4
4 2003 6 2 0 1 3
5 2004 17 1 6 2 8
6 2005 19 1 4 0 14
7 2006 10 3 4 1 2
8 2007 15 2 2 1 10
9 2008 10 1 0 0 9
10 2009 4 1 0 0 3
11 2010 6 0 0 0 6
12 2011 4 0 0 0 4
13 2012 4 1 0 0 3
14 2013 3 2 0 0 1
15 2014 6 1 0 0 5
16 2015 8 1 0 0 7
17 2016 4 0 0 0 4
Total Firms 126 16 16 6 88
Source: Equity Listing History section on PSX data portal
Figure 3. 1: Year-wise Number of IPOs in the Sample
0
2
4
6
8
10
12
14
Year-wise Listed IPOs on PSX
71
Table 3. 2: Sector-wise IPO firms in the sample
Sr. No. Sector Name Listed firms
1 Automobile & Electrical Goods 4
2 Cement 5
3 Chemicals 6
4 Commercial Banks 10
5 Engineering 7
6 Fertilizers 2
7 Food & Allied Products 2
8 Insurance & Leasing 1
9 Investment Securities & Banks 12
10 Modaraba 3
11 Oil & Gas 6
12 Power Generation & Distribution 7
13 Property & Investment 3
14 Technology & Communication 12
15 Textile 6
16 Transportation & Communication 2
Total Listed Firms 88
Source: Equity Listing History section on PSX data portal
Figure 3. 2: Sector-wise Number of IPOs in the Sample
0
2
4
6
8
10
12
Sector-wise Listed IPOs on PSX
72
Figure 3.1 presents the year-wise distribution of the number of IPOs in the
sample period which shows that most private firms went public during 2004 to 2008
because of high market valuations, high GDP growth, low inflation and large pace of
FDI during 2004 to 2008 (Source: Pakistan Economic Survey (2004-05, 2005-06,
2006-07, 2007-08) and annual reports of SBP, and (Sohail and Raheman, 2010)). It
also shows that the rate of new floatation went down during the Internet Bubble crisis
and US Subprime mortgage crisis periods because most world markets dented the
relative valuations for the new placements and PSX also confront similar issues
during crisis periods. Figure 3.2 presents the sector-wise distribution of the number of
IPOs in the sample period which shows the sector-wise development trend in the
country. Nearly all unlisted commercial banks went public during the sample period
due to tight SBP regulations put into action that functional banks must register
themselves in the capital market within the six months from their date of
incorporation. Many of the investment companies and banks got registered and raised
equity capital because of attractive market valuations, the high pace of foreign
portfolio investment in the PSX, increased market transparency due to capital market
reforms, succeed world best performing market on different time of periods and
historically peaked the bench market index.
This study used secondary data for the pre- and post-IPO valuation analysis.
The data used in the choice and accuracy of underwriter’s valuation approaches is
hand collected from the prospectus documents that were published at the time of
listing on the PSX. The IPO share price data is collected from the official websites of
PSX and business recorder. The unique feature of this study is the cash and stock
dividends adjusted share prices data that has been used to estimate the long-run
performance of IPOs using the event-time and calendar-time approaches. The data
used in the different asset pricing models have been extracted from the annual reports
of 225 non-IPO firms that remain listed during the sample period on the PSX.
3.2 Theoretical Background
IPO process usually takes a long time and the uptight process for a private
firm that wants to raise long-term capital through selling ordinary shares to the
general public. To get permission for going public, firms need to fulfill a list of
requirements by market regulators. Firms that want to list on the capital market need
73
to submit a comprehensive document after the due diligence is called a prospectus.
This document must include all relevant information such as the structure of share
capital, the purpose of IPO proceeds, future prospects, valuation methods used to
estimate fair value estimates, history of IPO firms, preceding financial statements
before the IPO and the profile of the management team and detail of associated
companies.
Loughran and Ritter (1995), Lowery (2003), and Colak and Gunay (2011)
argue that as more information becomes available in the long run, the relationship
between IPO pricing and ex-ante risk factors disappears slowly as discussed earlier.
Therefore, ex-ante risk factors are positively correlated with expected returns. This
study also checks the relationship between signaling factors and the IPO pricing as
well as aftermarket performance. As the practice, stock prices are determined by the
current fundamentals, future prospects and discount expected risk factors. Peterle and
Berk (2016) and Agathee et al., (2012a) inside investors know more information
about future prospects than outside investors. Inside investors disclose firm’s
prospects as e.g., signals to forthcoming investors. If outside investors incorporate
signals in their decision to participate in an IPO then it could reduce the mispricing
and IPO firms could get utmost proceeds than they expected to be.
3.3 The Choice, Bias and Accuracy of Valuation Methods
This section constructs the methodology based on the theoretical basis and empirical
evidence to explain the selection of several valuation methods, bias and accuracy of
each valuation technique reported in the prospectus documents.
3.3.1 Enlightening the choice of valuation methods
In this part, this study constructs an econometric methodology to enlighten the
choice for a particular valuation model employed by the lead underwriters in the
domain of new offerings. This study employs the binary logit regression model
comparable with Roosenboom (2012, 2007) and Deloof, Maeseneire and Inghelbrecht
(2009) to examine the cross-sectional determinants of the selection of the valuation
methods used by the lead underwriters. The binary logit model is used, due to binary
outcomes of each valuation method variable (dummy variable), to predict the
probabilities of binary outcome of a given predictor variables. This study estimates
the following models.
74
Model 1: Based on the valuation method selected
𝑉𝑎𝑙𝑢𝑎𝑡𝑖𝑜𝑛_𝑀𝑒𝑡ℎ𝑜𝑑
= 𝛽0 + 𝛽1𝐿𝑛𝑆𝑖𝑧𝑒𝑖 + 𝛽2𝐿𝑛(1 + 𝑎𝑔𝑒)𝑖 + 𝛽3𝐴𝐼𝑃𝑖 + 𝛽4𝑃𝑅𝑂𝐹𝑖
+ 𝛽5𝐺𝑅𝑂𝑊𝑖 + 𝛽6𝐷𝐼𝑉𝑖 + 𝛽7𝑇𝑒𝑐ℎ𝑖 + 𝛽8𝑀𝑘𝑡𝑅𝑒𝑡𝑖 + 𝛽9𝑆𝐷𝑖 + 𝛽10𝑈𝑅𝑒𝑝𝑖
+ 𝛽11𝐷𝑖𝑙𝑢𝑡𝑖𝑜𝑛𝐹𝑎𝑐𝑡𝑟𝑖 + 𝜀𝑖
Equation (1)
In the preceding econometric equation, the dependent variable
(Valuation_Method) is a dummy variable that equals one if underwriter choose
comparable multiples valuation method for IPOs valuation and zero otherwise. On the
same pattern, if the underwriter choose discounted cash flow method then the
dependent variable equals one and zero otherwise and so on for other methods. The
peer group multiples valuation method is used for comparative valuation while DCF
and DDM valuation models are direct valuation techniques.
The firm size is estimated by the natural logarithm of total assets as Log(Total
Assets) from latest financial year statements disclosed in the prospectus prior to IPO.
This study employed natural log of total assets to normalize the distribution of data
and to control the ‘scale effect’ issue. Beatty and Ritter (1986), and Ritter (1984)
argue that the larger IPO firms can be easily estimated as they are more stable in
terms of market share, revenue growth, payout history and forecasted cash flows. This
helps to employ direct valuation models such as DCF and DDM more probably. The
firm age is estimated using natural logarithm of one plus firm age i.e. Log(1+Age) is
considered as an ex-ante uncertainty. Ritter (1984) argues that the degree of
uncertainty inversely associated with the age of the firm. Kim and Ritter (1999) point
out that it is complicated to predict forecasted cash flows and payouts for young firms
without creating financial statements track record because many of their estimations
based on the expectations regarding future growth rates, which significantly differ
from in each case. Therefore, mature firms are probably to be priced using direct
valuation models. This study estimates asset tangibility (AIP) as the ratio of property,
plant and equipment to total assets of the latest preceding year disclosed in the
prospectus. Lev (2001) argues that the accounting numbers supposed to be a good
estimator of the firm value of the tangible assets than the intangible assets. Therefore,
the firms with high asset tangibility increased the use of accounting-based valuation
75
methods such as economic value added method. This study theorizes that the lead
underwriters probably employ comparable multiples to value IPOs when they predict
that the issuer firm is comparatively profitable (PROF) in the year of issue. If the lead
underwriters are predicting that the issuing firm expected to be marginally profitable
or documenting loss on the basis of a due diligence report, IPO valuation estimates
and management prospects about future cash flows, they prefer to use other valuation
methods than the multiples valuation because applying the P/E ratio in these situations
results negative or low multiples valuation.
H1: The lead underwriters are more likely to use direct valuation methods
such as dividend discount model and discounted cash flow method when
valuing large firms.
H2: The lead underwriters are more likely to use direct valuation methods
when valuing older firms.
H3: The lead underwriters are more likely to use accounting based valuation
methods when valuing firms having higher asset tangibility.
H4: The lead underwriters are more likely to use multiples valuation method
for those that comparatively profitable in the year of issue.
In this study, predicted growth in sales in the IPO year is used as a proxy for growth
opportunities. The rapidly growing firms face challenges of cash imbalances in the
short to medium term because the capital investments are more than the cash inflows.
Penmen (2001) argue that the discounted cash flow model treat capital investment as
a loss of value and free cash flow model unable to recognize the value that does not
engage cash flows. In addition, the rapidly growing firms more likely to keep earnings
as capital reserves than to offer cash dividends. These firms are valued by comparable
multiples because of persistent investment in growth opportunities. Therefore, the
study hypothesize that the rapidly growing firms more likely to be valued using
comparable multiples than the direct valuation methods. One of the most important
features of our data is that Pakistani IPO firms typically report their historical payouts
(DIV) and/or payout pattern of associated companies in the offering document. Firms
that have a track record to announce high dividend are perceived as quality firms.
Bhattacharya (1979) argues that the high-quality firms only announce dividends to
76
their shareholders to signal their quality. In his theoretical explanations, dividends
tend to be a non-recoverable expense and visible signal to stakeholders while low-
quality firms prefer to use internal financing instead of expensive outside financing. A
trend of consistent payout (DIV) in the preceding years compels underwriters to value
IPO firms using the dividend discount model. Damodaran (1994) points out that the
dividend discount model is the best measure for valuing stable and high dividend
paying firms. Therefore, it is hypothesized that the underwriters are more likely to use
the dividend discount model during setting the fair value estimate for IPO firms that
pay a large portion of their earnings as the dividends in the past.
H5: The lead underwriters are more likely to use multiples valuation method
when valuing rapidly growing firms.
H6: The lead underwriters are more likely to use dividend discount model for
those that pay a large portion of their earnings as dividends in the past.
Bartov et al., (2002) argue that technology-based companies are more difficult to
assess their fair value estimate because a major division of their fair value comes from
their growth opportunities. Therefore, it is anticipated that the technology firms are
probably to be assessed using the multiples valuation than the direct estimation
methods such as DCF and DDM models because these models do not incorporate the
value of growth options in the fair value estimates. A dummy variable is used to
control the impact of technology companies (Tech) that equals to one if the IPO firm
is high-tech firm and zero otherwise. Discounted cash flow estimates are sensitive to
forecasted cash flows and discount rates as estimated by underwriters to value IPOs.
DeAnglo (1990) points out that it is not very convincing to external shareholders due
to the sensitivity of the value estimates made by lead underwriters, investors are more
likely to go for those IPOs in which window of opportunity prevails. A persistent rise
in aggregate stock returns may point out the window of opportunity. For that purpose,
this study adds market returns (MktRet) during a six-month interval from 185 trading
days prior to IPO and 5 days before the formal listing of IPO firm. This study assumes
that the likelihood of employing the discounted cash flow method increases when the
aggregate stock market returns before going public are high. The assumption of the
DDM method contrasts with the DCF model because DDM is used when aggregate
stock market returns are poor. Baker and Wurgler (2004) proposed a theory that the
77
investors pay more attention to buy those stocks that pay dividends on a regular basis.
Therefore, it is anticipated that during the bearish momentum, investors are more
likely to buy dividend-paying stocks when aggregate stock market returns are
declining.
H7: The lead underwriters are more likely to use multiples valuation method
when valuing technology firms.
H8a: The lead underwriters are more likely to use the discounted cash flow
method when aggregate stock market returns before the IPO are high.
H8b: The lead underwriters are more likely to use the dividend discount model
when aggregate stock market returns before the IPO are low.
Roosenboom (2007) argues that the investors are uncertain about the fundamental
values of securities when aggregate market returns are highly volatile. In this study,
standard deviation (SD) of the benchmark index is included during a six-month
interval from 185 trading days prior to IPO and 5 days before the formal listing of
IPO firm. The investment bankers choose a direct valuation method to cater the
investors demand by controlling the impact of market volatility. Therefore, It is
anticipated that the direct valuation methods are to be used more often when market
returns are more volatile beforee IPOs. In this model, underwriter reputation is
included as a control variable. This study employs underwriter market share as a
proxy for underwriter reputation (URep). Following Roosenboom (2012) and
Ljungqvist & Wilhelm (2002), underwriter reputation is used as a dummy variable
and take a value of 1 for prestigious underwriters and 0 for less reputed underwriters.
Carter and Manaster (1990) suggest that the prestigious underwriters are considered to
be more experts in valuing IPO firms. The magnitude of underwriter reputation may
impact the choice of valuation method although this study is unable to find any
particular prior prediction regarding the valuation model by prestigious underwriters.
The rapidly growing firms offer a large portion of their shares in general public
offerings to finance their future expansions and investments while mature firms small
part in the public offering. Therefore, if the value of dilution factor (DilutionFactr) is
large then underwriters more likely to use comparable multiples and direct valuation
methods for small dilution factors.
78
H9a: The lead underwriters are more likely to use direct valuation methods
when aggregate stock market volatility is large before the IPO.
H9b: The lead underwriters are more likely to use multiples valuation methods
when aggregate stock market volatility is low before the IPO.
H10a: The lead underwriters are more likely to use direct valuation method
when valuing IPOs that offered a large proportion of the outstanding shares in
the IPO.
H10b: The lead underwriters are more likely to use multiples valuation
method when valuing IPOs that offered a small proportion of the outstanding
shares in the IPO.
3.3.2 Bias and Accuracy of Valuation Methods
Numerous studies deal with the valuation bias and accuracy of each method
used by underwriters during setting the IPO value estimates. Boatsman and Baskin
(1981) examine the peer group multiples valuation method and found that the
valuation process produces improved results when matched firms are selected on the
basis of same industry and trailing earnings growth. Alford (1992) investigates the
valuation accuracy of earnings per share (EPS) when firms are chosen based on
earnings growth, same industry, size and leverage. He shows that valuation errors
decline when matched firms choosen from one-digit SIC code ot two and three.
Kaplan and Ruback (1995) compare the valuation accuracy of peer group multiples
with the discounted cash flow model and show that the estimates determined on the
basis of comparable multiples underestimate the transaction value. Bakar and Ruback
(1999) investigate the valuation accuracy using the harmonic mean of multiples
chosen on the basis of sales, EBIT and EBITDA and reported that the performance
adjusted by EBITDA carry out better estimation than sales and EBIT. Bhojraj and Lee
(2002) put extra focus on the collection of matched firms used for valuation accuracy
and employ a linear regression model that predict the warranted multiples for each
focused firm. Liu et al., (2002) found that forecasted and trailing earnings offer the
best estimates followed by the book value of equity and multiples based on cash flow
measures. Liu et al., (2004) provide evidence under the findings of international
79
markets that multiples on the basis of price-earnings perform best for valuation
accuracy.
Several studies of valuation accuracy defined the prediction errors on the basis
of market value (Roosenboom, 2012, Deloof et al., 2009, Liu et al., 2002; Francis et
al., 2000). This study estimates the prediction errors based on the market values
instead of IPO offer values because the lead underwriters sometimes involve, on the
same time, valuing fair-value estimates and a decision process of IPO offer prices. For
instance, if the underwriters want higher offer prices, then they may choose
comparable firms with high multiples and if the lead underwriters want lower offer
prices, then they may select comparable firms with low multiples (Kim and Ritter,
1999). The key insight of this approach, the prediction errors should contain the
underwriters’ deliberate offer price discounts that they use to set preliminary offer
prices and as a marketing tool to create excess demand from investors in primary
auction. These prediction errors assist underwriters to decide which valuation method
is better to value IPOs and which is not.
This study follows the methodology to estimate bias and accuracy of valuation
methods as used in Kim and Ritter (1999), Francis et al., (2000), Deloof et al., (2009)
and Roosenboom (2012). In this study, first, this section estimate the signed
prediction errors (SPE) of each valuation method as
𝑆𝑖𝑔𝑛𝑒𝑑 𝑃𝑟𝑒𝑑𝑖𝑐𝑡𝑖𝑜𝑛 𝑒𝑟𝑟𝑜𝑟𝑠 = (𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑉𝑎𝑙𝑢𝑒𝑖,𝑝𝑟𝑒𝐼𝑃𝑂 − 𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒𝑖)
𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑒𝑖
Equation (2)
Where Estimated Valuei,preIPO is the value estimate determined using the lead
underwriter’s valuation methods when valuing IPO firm ‘i’ and Market Value is the
closing price of the first trading day in the stock market. In this study, both first
trading day closing prices and IPO offer prices as market values are used to estimate
the signed prediction errors to better understand the biases linked to each valuation
method. The signed prediction errors based on offer prices are important because, in
most of the cases, underwriters at same times are involved to estimate fair value
estimates and to set IPO offer prices. For instance, if the underwriters desire higher
offer prices, then they choose comparable firms having high multiples and if the
underwriters want to control the overpricing then they may choose comparable firms
80
with low multiples (Kim and Ritter, 1999). In this part, the Wilcoxon sign rank test
would apply to the median values of signed prediction errors to examine the bias of
each valuation method used by the lead underwriters during the evaluation process.
To compare and analyze the performance of several valuation methods used
by the lead underwriters in terms of valuation accuracy, this study inspects the
measures of dispersion for the pooled distribution of absolute prediction errors.
According to Schreiner & Spremann (2007), Cassia et al., (2004), Purnanandam &
Swaminathan (2004) and Kim & Ritter (1999) the important measures of valuation
accuracy are the mean absolute prediction errors (APE) and the percentage of signed
prediction errors below 15% of actual market values. By doing this, the findings turn
into comparable with related studies follow these measures to draw inferences.
Absolute prediction errors capture the valuation accuracy of each valuation method
and can be estimated as
𝐴𝑏𝑠𝑜𝑙𝑢𝑡𝑒 𝑃𝑟𝑒𝑑𝑖𝑐𝑡𝑖𝑜𝑛 𝐸𝑟𝑟𝑜𝑟𝑠 = |(𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑉𝑎𝑙𝑢𝑒𝑖,𝑝𝑟𝑒𝐼𝑃𝑂 − 𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒𝑖)
𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒𝑖|
Equation (3)
The absolute prediction errors are the absolute values of the signed prediction
errors. The percentage of signed prediction valuations within 15% of the actual
valuation method is estimated as
[log(𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑉𝑎𝑙𝑢𝑒𝑖,𝑝𝑟𝑒𝐼𝑃𝑂) − log(𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒𝑖)] < 0.15
Equation (4)
In this part, the study examines OLS regression to test the valuation relevancy
and predicting the power of each valuation method by conducting a Wald-test to
probe whether the intercept term is statistically different from zero and the slope
coefficient statistically different from one. This methodology assists to examine the
ability of value relevancy to explain cross-sectional variation in the market values.
The valuation theories conjecture that the market value of IPO firms is directly
proportional to the pre-IPO value estimates examined by the investment banks. If the
valuation models give unbiased estimates of market values then the intercept would
equal zero and the slope coefficient equals one. According to Cassia, Paleari &
81
Vismara (2004) for the purpose of valuation relevancy, a more broad-spectrum
regression model may be used as:
𝐿𝑜𝑔(𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒𝑖) = 𝛽0 + 𝛽1𝐿𝑜𝑔( 𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑉𝑎𝑙𝑢𝑒𝑖,𝑝𝑟𝑒𝐼𝑃𝑂) + 𝜀𝑖
Equation (5)
Where Estimated Valuei,preIPO is the value estimates determined by lead
underwriters to use valuation methods when valuing IPO firm ‘i’ and Market Value is
the closing price of the first trading day in the stock market, β0 and β1 are the intercept
and slope of a regression model respectively and εi is the pricing error. In equation
(5), the natural logarithm of Market Value is used as a dependent variable and the
natural logarithm of Estimated Value calculated through several valuation models
used as the independent variable as discussed in the earlier.
This study employed cross-sectional ordinary least square regression models
as comparable with Roosenboom (2012), the first time in the literature, to estimate the
association between these bias and accuracy estimates with the IPO firms’
characteristics as discussed in the earlier section.
Model 2: Based on the bias estimates
𝑆𝑃𝐸 = 𝛽0 + 𝛽1𝐿𝑛𝑆𝑖𝑧𝑒𝑖 + 𝛽2𝐿𝑛(1 + 𝑎𝑔𝑒)𝑖 + 𝛽3𝐴𝐼𝑃𝑖 + 𝛽4𝑃𝑅𝑂𝐹𝑖 + 𝛽5𝐺𝑅𝑂𝑊𝑖
+ 𝛽6𝐷𝐼𝑉𝑖 + 𝛽7𝑇𝑒𝑐ℎ𝑖 + 𝛽8𝑀𝑘𝑡𝑅𝑒𝑡𝑖 + 𝛽9𝑆𝐷𝑖 + 𝛽10𝑈𝑅𝑒𝑝𝑖
+ 𝛽11𝐷𝑖𝑙𝑢𝑡𝑖𝑜𝑛𝐹𝑎𝑐𝑡𝑟𝑖 + 𝜀𝑖
Equation (6)
Model 3: Based on the accuracy estimates
𝐴𝑃𝐸 = 𝛽0 + 𝛽1𝐿𝑛𝑆𝑖𝑧𝑒𝑖 + 𝛽2𝐿𝑛(1 + 𝑎𝑔𝑒)𝑖 + 𝛽3𝐴𝐼𝑃𝑖 + 𝛽4𝑃𝑅𝑂𝐹𝑖 + 𝛽5𝐺𝑅𝑂𝑊𝑖
+ 𝛽6𝐷𝐼𝑉𝑖 + 𝛽7𝑇𝑒𝑐ℎ𝑖 + 𝛽8𝑀𝑘𝑡𝑅𝑒𝑡𝑖 + 𝛽9𝑆𝐷𝑖 + 𝛽10𝑈𝑅𝑒𝑝𝑖
+ 𝛽11𝐷𝑖𝑙𝑢𝑡𝑖𝑜𝑛𝐹𝑎𝑐𝑡𝑟𝑖 + 𝜀𝑖
Equation (7)
82
Table 3. 3: Operational definitions of variables used in Equation (1), (6) and (7)
Variable Definition
Dependent Variables
Valuation_Method Valuation_Method is a dummy variable that equals one if
underwriter uses peer group multiples valuation method and zero
other wise. Dummy variable equals one if underwriter uses DCF
model and zero other wise. Dummy variable equals one if
underwriter uses DDM model and zero other wise.
SPE The signed prediction errors is measured through the percentage
difference between the estimated value during pre-issue pricing
process and market value over first day market values for each
valuation method separately.
APE The absolute prediction errors are the absolute values of signed
prediction errors for each valuation method seperately.
Independent Variables
LnSize Natural logarithm of total assets for the latest financial year
disclosed in the prospectus.
Ln(1+Age) Natural logarithm of one plus firm age (difference between the IPO
listing year minus the date of incorporation)
AIP Ratio of property, plant and equipment to total assets for the latest
financial year disclosed in the prospectus
GROW Forecasted sales growth during the current year
Div Dummy variable equals one if the IPO firm has a track record of
payout history and/or disclosed dividend policy in the prospectus
prior to an IPO and zero otherwise
Tech Dummy variable equals one if the IPO firm belongs to a technology
industry and zero otherwise
MktRet The aggregate market returns during a 180 days interval from the
185th trading day before to 5th trading day before the formal listing
date of IPO firm. (KSE100 index has been used for market returns)
SD The standard deviation of daily market returns during a 180 days
interval from the 185th trading day before to 5th trading day before
the formal listing date of IPO firm.
URep A dummy variable for underwriter reputation, take the value of 1 for
prestigious underwriters and 0 for less reputed underwriters
Prof The ratio of current year forecasted EBIT to current year forecasted
sales
Dilution Factor The ratio of newly issued shares over Total post-issue outstanding
shares
83
3.4 The Basic Valuation Model
An extant literature reveals that different researchers used different IPO
valuation methods. McCarthy (1999) argues that the IPO pricing process is all about
science and art. The scientific division starts when the investment banker and issuer
firm set IPO offer price based on historical information. The art division starts when
more market information such as peer multiples are incorporated to set IPO offer
price. From the academic point of view, Kim and Ritter (1999) estimate the
effectiveness of peer multiples methods. They found that comparable firms valuation
methods are worthy and the inclusion of market information increases the pricing
accuracy. Berkman et al., (2000) explore that peer multiples methods and Discount
Cash Flow (DCF) method have the same accuracy in the process of IPO pricing
decision.
Many researchers used accounting-based valuation models to value the IPO
(e.g., Rees, 1997; Fama and French, 1998; Easton and Sommer, 2003; Richardson and
Tinaikar 2004). Few researchers expand their models by adding more firm
characteristics as per their research objectives. According to Easton and Harris (1991)
basic valuation model, the price is a function of firms’ book value (BV), earnings (E)
and dividends (Div). Rees (1999) includes the dummy variable of negative earnings to
discriminate the effect of loss building firms on the IPO pricing. Rees (1997) uses
non-IPO firms’ data and find that dividends payout is a proxy for permanent income
which gives a positive signal to the firm value. The sample contains both large and
small firms with respect to assets and revenue. Many large firms have the track record
of dividends before the IPO and release the projected dividend in the prospectus,
while many small firms neither offer any payout before the IPO nor promise for any
payout in near future. Therefore, the firms having a track history of offering dividends
payout is added as signaling information in the initial valuation model to estimate the
firm value. Aggarwal et al., (2009) find that firms having higher negative earnings
before IPOs produce higher valuation and vice versa. The initial valuation model also
includes the dummy variable (D) to control the impact of negative earnings on
valuation. The underwriter in a new stock offering serves as the intermediary between
the company seeking to issue shares in an initial public offering (IPO) and investors.
The leading underwriter form a syndicate of investment bankers and provide
guarantees through underwriting agreement to issuer firm that investment banks
84
purchase entire stock if left in the IPO process. Thus, the basic valuation model is
given below.
𝑃0 = 𝛽0 𝐵𝑉𝑖 + 𝛽1𝐸𝑖 + 𝛽2𝐷𝑖𝑁𝐸 + 𝛽3𝐷𝑖𝑣𝑖 + 𝜀𝑖
Equation (8)
Where P0 is the IPO offer price, BV is the book value of shareholder’s equity,
E is earnings, D1NE is a dummy variable of negative earnings and Div is the
dividends disclosed in the prospectus.
It is already reported that the sample of IPO firms consists of large and small
firms that went public on the PSX. The market capitalization of each firm is a proxy
for firm size. A review of sample IPO firms, there is a scale difference which may
mislead the results interpreted. Rees (1999) and Brown et al., (1999) highlights the
existence of scale effects when they perform firm level analysis. Kothari and
Zimmerman (1996) argue that per share data can dilute the impact of
heteroscedasticity problem but does not an adequate control for the scale effects as
new offerings bring with variable sizes. Barth and Kallapur (1996) advocate two
procedures to control the scale effects; first, deflate through a scale proxy or addition
of scale proxy as a further explanatory variable. They argue that deflation by a
regression variable can mitigate the heteroscedasticity and coefficient bias, while the
addition of scale proxy as a separate independent variable can only mitigate
coefficient bias. So, this study uses deflation by a scale proxy followed earlier
literature.
It has been explored from existing literature that various researchers used four
different deflators in cross-sectional valuation studies as proxies for scale effect: First,
Hirschey (1985) used ‘sales’ as a proxy for scale effect. Second, Rees (1997) and
Beaver, Hand and Landsman (1999) used ‘total number of shares’ as a proxy for scale
effect. Third, Lo and Lys (2000) used ‘opening market price’ as a proxy for scale
effect. Finally, Core et al., (2003), Danbolt and Rees (2002) and, Easton (1998) used
‘book value’ as a proxy for scale effect.
This study already used per share data which robotically is deflated by the
total outstanding shares. Brown et al., (1999) argue that per share data is not an
appropriate measure to control scale effect as shares come in different sizes.
85
Therefore, per share data is not a suitable deflator for the valuation model. This study
analyses the IPO valuation on offer price and opening market value as a deflator
might mislead the results discussed. If market value considered as a deflator (P0/P0)
then the dependent variable produces a constant value one for all observations and it
will become the complicated model as WLS regression, as an advocate of Easton and
Sommer (2003). Barth and Kallapur (1996) suggest that book value is a suitable scale
factor than sales or opening market value for two reasons. First, book value is already
an independent variable and helps to mitigate coefficient bias. Second, when book
value is considered as a scale factor, it transforms the dependent variable to price-to-
book value (P/BV) ratio, which is a commonly used ratio despite price-to-sales ratio
or price-to-price ratio. From IPO literature, Kim and Ritter (1999) explore that
market-to-book value ratio increase the predictive power of valuation model, which is
comparable with peer competitor multiples. Keasey and McGuiness (1992) and, How
and Yeo (2000) used price to book value ratios as a deflator in their price valuation
models – information as disclosed in the prospectus. This study follows Kim and
Ritter (1999) and uses the book value as a scale factor.
𝑷𝟎
𝑩𝑽𝑖= 𝛽0
𝐵𝑉𝑖
𝐵𝑉𝑖+ 𝛽1
𝐸𝑖
𝐵𝑉𝑖+ 𝛽2𝐷𝑖 + 𝛽3
𝐷𝑖𝑣𝑖
𝐵𝑉𝑖+ 𝜀𝑖
Equation (9)
The definition of variables is same as discussed in the previous equation scaled by
book value per share.
3.4.1 The IPO Valuation Model
The main research objective of this study is “to investigate the impact of
fundamental factors, ex-ante risk factors and signaling factors on IPO initial
valuations and aftermarket performance”. As addressed earlier, various researchers
(Roosemboom, 2007, 2012; Xia et al., 2012; Deloof et al., 2007; Kim and Ritter,
1999; Ritter, 1984) explored that IPOs ex-ante risk factors impact on IPO pricing
process and their aftermarket successive performance. An extant literature on IPO
signaling studies, a number of signaling factors have been estimated and few
significant factors have been uncovered. This section only presents the empirical
model related to IPO pricing valuation and empirical models related to estimating the
relationship between the risk factors and the signal variables with subsequent
performance to be discussed in the next two sections. The ex-ante risk factors are
86
classified into two categories such as financial and non-financial risk factors. The
capital availability risk and financial leverage are used as financial risk factors. The
non-financial risk variables are the capacity risk, efficiency risk, IPO gross proceeds
and firm’s beta. Based on extant literature on signaling variables, this study identifies
three signal variables as; the underwriter reputation, proportion of shares offered and
the age of IPO firms.
The financial leverage (FinLev) is the degree to which a company uses
external capital (e.g., debt and preferred stock) to attain additional assets. The
excessive use of debt financing increases its financial leverage in terms of high-
interest payments, which negatively impact on firms’ core earnings. A company
should keep its optimal capital structure in mind when making the financing decision
to ensure any increases in debt financing increase the intrinsic value of a firm.
According to studies about IPO valuation and financial leverage (Modigliani and
Miller, 1966); higher the proportion of debt in the capital structure that increases the
systematic risk of a firm’s value and the probability of going bankrupt. Therefore, it is
established that higher leverage, increases the level of insolvency. Loughran and
McDonald (2013), Thomas (2011), and Roosenboom (2012) estimate association
between the financial leverage and the firm’s equity. They argue that pre-IPO higher
financial leverage as a proxy for ex-ante risk increases the deflation of equity
valuation. The listing rules of new offerings required that offering document should
publish financial statements at least three years before IPO. The pre-issue debt ratios
from latest financial statements is used as financial leverage.
The second ex-ante financial risk factor used in this research is capital
availability risk (CaptlRsk). Generally, firms used internal and external capital to
finance their resources. Internal capital includes paid-up capital against common
equity and retained earnings while external capital includes debt and preferred equity.
The excessive use of external capital increases their financial risk because of high-
interest payments while internal capital raises less risk. Therefore, a large portion of
retained income represents the higher capital availability and lower financial risk. The
capital availability is estimated through an average of retained earnings ratios from
the latest financial statements publish in the prospectus documents. Mathematically, it
can be written as; Retained Earnings Ratio = 1 – Payout Ratio
87
Furthermore, the valuation methodology also incorporates the four non-
financial risk factors. The efficiency risk (EffRsk) is the first non-financial risk factor,
which depicts the operating efficiency. The firms’ with high operating efficiency
means the less cost of goods sold and less efficiency risk. An extant literature on
operating efficiency, various researchers explored the positive relationship between
the operating efficiency and the financial performance of the IPO firms. Jain and Kini
(1994) investigate the operating performance of US newly listed firms during 1976 to
1988 after going public. Their findings show that aftermarket operating performance
is positively related to equity retention by initial shareholders. They argue that the
firms having large offer price discount perform worst after the IPO. Mikkelson et al.,
(1997) investigate the post-IPO operating performance and its determinants during
1980 to 1983. They find out that the operating performance decline after going public
and the variation in operating performance explained mostly by firm’s size and firm’s
age rather than the ownership structure. Kim, kitsabunnarat and Nofsinger examined
the Thai IPOs. They explore a significant positive relationship between the post-IPO
operating performance and the financial performance. They find a significant drop in
operating performance followed by subsequent time periods. They investigate that the
operating performance significantly explained by sales growth, capital expenditures
and the asset turnover ratio. The reason for adding efficiency risk factor in our
methodology is due to unavailability of complete information about IPO firms in the
prospectus. The efficiency risk factor is measured by an average of the ratio of cost of
goods sold (CGS) over firm turnover disclosed in the offering documents. Therefore,
there is an inverse relationship between the operating performance and the ratio of
CGS over firms’ turnover.
The second non-financial risk factor is the capacity risk (CpctyRsk), which is
added to the valuation model. The capacity risk is defined as, the probability of
success of the new venture over firms’ capacity. In this research, larger the portion of
IPO proceeds used to be investment activities tends to large uncertainty of returns
from the new venture. Most, IPO firms decide to go public to raise equity capital to
finance their operations and a particular investment project for business expansion.
Though, in many cases IPO proceeds are used to provide an exit strategy, to repay
debt or to redeem preferred equity. An extant literature on utilization of IPO proceeds
describes the impact of IPO proceeds on the underpricing or aftermarket performance
88
(e.g., Leone et al., 2003; Espenlaub et al., 1999; Klein 1996). This study investigates
the impact of the proposed utilization plan of IPO proceeds on the IPO pricing and
aftermarket performance. This research uses the ratio of the proposed utilization plan
of IPO proceeds over net IPO proceeds as a proxy for capacity risk. Higher the
proportion of utilization plan over the IPO proceeds indicates higher the capacity risk.
Beaver et al,. (1970) argues that firm beta is an important factor to evaluate the
riskiness of the firm’s value. The firm beta is defined as to estimate the sensitivity of
change in the firm share price due to the variation in the market index. Though, due to
unavailability of share prices before formal listing in the market, the standard
deviation of IPO aftermarket prices for 180 trading days from the date of formal
listing on the market. A firm having higher beta is perceived to be riskier. Therefore,
firm’s beta is negatively correlated with aftermarket performance. KSE100 index is
considered to estimate market return.
The IPO gross proceeds (OffrSize) is used as a non-financial risk factor in the
valuation model. The offer size is broadly used as a proxy for the level of risk of the
IPO firms. Perhaps, firms having small offer size are more speculative than the large
size offerings (Ritter, 1984; Beatty and Ritter, 1986; Loughran and Ritter, 1995).
Bessler and Thies (2007) and, Agarwal, Liu and Rhee (2008) find that offer size is a
significant determinant of long-term performance. Sohail and Nasr (2006), Loughran
and Ritter (2002), Carter et al, (1998) and Pinkle (1998) investigates the association
between offer size and firm valuation. They find the inverse association between
short-run and long-run performances. The offer size measured through as a product of
the number of shares offered in IPO with the offer price.
The underwriter reputation (UndRep) is a first signaling variable in the
valuation model. The reason for adding underwriter reputation into the valuation
model is based on the theoretical basis. Baron (1982) enlightens the important role in
defining the allocation of shares and their subscriptions. In the IPO process, an
underwriter is bound to make an underwritten agreement to sell all shares with IPO
issuing firm and responsible if shares left unsold. To avoid the under-subscription
risk, the prestigious underwriters only engage in quality IPOs. Therefore, the firms
having lower quality could not bear prestigious underwriters underwriting fees then
these transactions are sponsored by less prestigious underwriters. It is conjectured that
IPOs sponsored by prestigious underwriters offered at a higher price than the firms
89
sponsored by not-prestigious underwriters. Dominique et al., (2013) investigate the
reputation of underwriter and the auditors on the IPO pricing and aftermarket
performance. Underwriters deliberately offer discount in the offer price than the
intrinsic value to minimize the risk of under subscription of offered capital. Their
empirical results show that the firms choose prestigious underwriters for IPO process
leave less money on the table. Mudrik and Imam (2002) emphasize that firms prefer
to hire reputed underwriters due to the reduction in undervaluation of equity. Kim et
al (1995) also explored that there is a negative and significant relationship between
the underwriter reputation and the level of underpricing. This study hold a dummy
variable underwriter prestige and its value equals 1 if IPO is offered through high
reputed underwriter and 0 for the less reputed underwriter. An extant literature on
underwriter reputation estimation, two most common approaches are adopted to
calculate the underwriter reputation classifications such as (1) underwriters included
in the specific year’s ranking of underwriter list and, (2) the underwriters relative
market share in preceding IPOs (Carter and Manaster, 1990; Meggison and Weiss,
1991). The methodology employs to estimate the classification of underwriter
reputation is comparable with Keasay and McGuiness (1992). The ranking
categorization is on the basis of how many newly listed firms underwritten by various
underwriters in the preceding time periods.
The firm age, a signaling factor, is used in the valuation model as a signaling
variable to enlighten the different IPO puzzling facts. The firm age imitates the
maturity level and size gained in the product market, tend to reflect the stability of
business operations and the market share. Ritter (1999) argues that it is complicated to
forecast predicted cash flows and corporate payouts of young firms without having
their track record. Older firms supposed to be less risky because of more experience
and stability in the business operations. The firm age is a key characteristic of the IPO
firm which uses to explain the variation in the aftermarket returns. The firm age is
estimated as the difference between the date of incorporation and the date of formal
listing on the stock exchange.
Firstly, Leland and Pyle (1977) suggest that the percentage of shares retained
by initial sponsors at the time of dilution of ownership contains very useful
information for outside investors. Most stock exchanges, at the time of the IPO,
require a certain limit of minimum shares to be retained by the initial shareholders and
90
the associates. Leland and Pyle (1977) observe that the large shareholder’s decision
about the percentage of ownership retention show the extent of confidence about the
firm’s future prospects. On the other hand, larger the part of shares offered in the IPO,
loses the confidence of large shareholders on the firm’s prospects. Beatty and Ritter
(1986) conjectur that firms offered less offer size are more speculative than larger
offer IPOs. Thus, there is a negetive association between the proportion of shares
offered and underpricing level. Prior literature used a number of proxies to estimate
the association between the ownership retention and the aftermarket valuation. In this
study, the percentage of shares offered to the general public is used as the inverse
proxy for the retained ownership.
The empirical valuation model used based on the theoretical foundations to
examine the association between the fundamental factors, the ex-ante risk factors and
the signaling factors are put together in the following equation.
𝑷𝟎
𝑩𝑽𝒊= 𝛽0 + 𝛽1
𝐸𝑖
𝐵𝑉𝑖+ 𝛽2𝐷𝑖 + 𝛽3
𝐷𝑖𝑣𝑖
𝐵𝑉𝑖+ 𝛽
4𝐹𝑖𝑛𝐿𝑒𝑣𝑖 + 𝛽
5𝐶𝑎𝑝𝑡𝑙𝑅𝑠𝑘
𝑖
+ 𝛽6𝐸𝑓𝑓𝑅𝑠𝑘
𝑖+ 𝛽
7𝐶𝑝𝑐𝑡𝑦𝑅𝑠𝑘
𝑖+ 𝛽
8𝐹𝑟𝑚𝐵𝑒𝑡𝑎𝑖 + 𝛽
9𝑂𝑓𝑓𝑟𝑆𝑖𝑧𝑒
𝑖
+ 𝛽10
𝑈𝑛𝑑𝑅𝑒𝑝𝑖
+ 𝛽11
𝐹𝑟𝑚𝐴𝑔𝑒𝑖
+ 𝛽12
𝑃𝑂𝑆𝑖 + 𝜀𝑖
Equation (10)
Where FinLev is the ex-ante financial leverage before IPO and is measured
through the average of logarithm of the pre-IPO debt ratios, CaptlRsk is the capital
(e.g., internal capital) availability risk prior to IPO and measured through the average
of pre-IPO retained earnings to net income, EffRsk is the pre-IPO efficiency risk and
measured through the average of the ratio of pre-IPO cost of goods sold over net
sales, CpctyRsk is the capacity risk prior to IPO and measured through the ratio of
IPO proceeds utilization plan disclosed in the prospectus over total IPO proceeds,
FrmBeta is a firm beta as proxy of price volatility and measured as a standard
deviation of aftermarket IPO prices for 180 trading days, OffrSize is the IPO gross
proceeds and measured through the product of number of shares offered to general
public and offer prices, UndRep is the underwriter reputation prior to IPO and
measured through a dummy variable take the value of 1 for prestigious underwriter
and 0 for less reputed underwriter. In this study, the cut-off point as the underwriters
participate in the preceding IPOs is 6 to bifurcate the underwriter prestige, FrmAge is
the age of issuing firm at the time of IPO and measured through the difference
91
between the date of incorporation and the formal date of listing on the stock
exchange, POS is the proportion of shares unsold by initial shareholders at the time of
IPO and measured as an inverse proxy for proportion of shares offered in the IPO.
Table 3. 4: Operational definitions of variables used in IPO valuation model
Variables Definition
P/BV The preliminary offer price scaled by book value (BV) of
shareholder’s equity per share
E Latest Earnings Per Share disclosed in the prospectus
D A dummy variable for negative earnings reported in the
prospectus document take the value of 1 for negative earnings
and 0 otherwise
UA A dummy variable for underwriting agreement in fixed price and
book building auctions take the value of 1 for IPOs offered
through book building and 0 for fixed price auction
Div Proposed dividend in the latest financial year disclosed in the
prospectus
Financial Leverage
(FinLev)
The ratio of Total Liabilities over Total Assets using the latest
year financial statements data disclosed in the prospectus is also
known as debt ratio.
Capital Availability Risk
(CaptlRsk)
The ratio of retained earnings over net income using the latest
financial statements data before IPO disclosed in the prospectus.
Efficiency Risk (EffRsk) The ratio of cost of goods sold (CGS) over net sales disclosed in
the latest financial statements before the IPO
Capacity Risk
(CpctyRsk)
The ratio of proposed investments plan disclosed in the
prospectus over the IPO gross proceeds
Firm Beta (FrmBeta) Firm beta is a proxy for price volatility, the standard deviation of
IPO aftermarket prices for 180 trading days since the date of
formal listing on the market.
Offer Size (OffrSize) Product of shares offered in the IPO and offer price
Underwriter Reputation
(UndRep)
A dummy variable for underwriter reputation, take the value of 1
for prestigious underwriters and 0 for less reputed underwriters.
Firm Age (FrmAge) Calculate by a natural log of one plus firm’s age: Log (1+Age)
Firm’s age measured through the difference between the date of
incorporation and the date of formal listing on the exchange.
Shares offered in IPO
(POS)
The proportion of shares offered in the IPO over post-IPO
outstanding shares
92
3.4.2 Hypotheses about the Valuation Model
Based on the theoretical foundations and the valuation model, it has been
concluded that the price is an increasing function of the book value of shareholder’s
equity and its regular earnings. An existing literature expresses the negative impact of
pre-IPO negative earnings in the IPO valuation process. Hayn (1995) highlights that
loss in preceding years than profits history is less enlightening about firm’s future
prospects. McCarthy (1999) argues that the lead underwriters use accounting
information such as book value, dividends and earnings to set the preliminary IPO
offer prices. As already discussed in the literature review chapter, Klein (1996) and
Beatty et al., (2002) used trailing earnings disclosed in the prospectuses while Kim &
Ritter (1999) and How & Yeo (2001) used forecasted earnings to estimate the fair
value of new offerings and their findings conclude that the earnings have large
predicting power of initial prices. In addition, previous literature also finds the
significant role of the dividends on the stock valuation. The dividend policy also
illustrates the strength of firm’s future cash flows which significantly impact on the
share price variation. Hypotheses regarding the fundamental variables are mentioned
below.
H11a: The book value of shareholder’s equity is positively related to the
preliminary offer prices.
H11b: The latest’s financial year earnings per share is positively related to the
preliminary offer prices.
H11c: There is a positive relationship between negative earnings per share
and the preliminary offer prices.
H11d: The dividend policy disclosed in the prospectus is positively related to
the preliminary offer prices.
As discussed earlier, based on risk-aversion hypothesis, it is the supposition that firms
having high risk, lower the share prices. According to Loughran & McDonald (2013)
and Thomas (2011) pre-IPO higher financial leverage as a proxy for ex-ante risk
increases the deflation of equity valuation. Modiliani & Miller (1966) argue that
higher leverage increases the risk of insolvency. Fama and French (1998) argue that
93
the IPO firms follow the pecking order theory as they prefer to finance their
investments first with the internal resources such as retained earnings, then with the
external resources such as debt and issuing equity. Keasey and Short (1997)
investigate a positive relationship between the IPO market prices and the IPO
proceeds. The offer size is generally used as a proxy for the level of risk of the IPO
firms (Aggarwal, Liu and Rhee, 2008: Bessler and Thies, 2007). This implies that the
market price of any security fluctuates based on the available information and the
risky assets face lower demand. The hypotheses regarding ex-ante risk factors and the
offer prices are mentioned below.
H12a: The firm’s financial leverage before the IPO is negatively related to the
preliminary offer prices.
H12b: There is a positive relationship between the pre-IPO capital availability
risk and the preliminary offer prices.
H12c: There is a negative relationship between the pre-IPO firm’s efficiency
risk and the preliminary offer prices.
H12d: There is a positive relationship between the pre-IPO firm’s capacity
risk and the preliminary offer prices.
H12e: There is a negative relationship between the pre-IPO firm’s beta and
the preliminary offer prices.
H12f: There is a negative relationship between the IPO offered size and the
preliminary offer prices.
Based on the existing studies regarding signaling factors, it is conjectured that the
signaling factors have a positive impact on the IPO valuations. Yung (2011) argues
that prestigious underwriters should have an advantage in information production
because of a large network with high net-worth institutions and/or individuals resulted
in greater price revision in the auction.The hypotheses about signaling factors and the
IPO valuations are discussed below.
H13a: There is a positive relationship between the underwriter reputation and
the preliminary offer prices.
94
H13b: There is a positive relationship between the firm’s age and the
preliminary offer prices.
H13c: There is a negative relationship between the percentage of shares
offered in the IPO and the preliminary offer prices.
3.5 The Initial Excess Return Model
The rationale to construct valuation model is to address the issue of whether
the underwriters and market participants employ prospectus data to value IPO firm’s
equity. Furthermore, it also draws attention to the difference of opinions between the
market participants about the utility of prospectus information on the IPO pricing.
Therefore, this study investigates whether the difference of opinions has any
significant impact on the aftermarket performance. Various researchers argue that
initial excess returns on the first trading day is still an unsettled puzzling stylized fact.
The initial excess returns model attempts to scan whether the prospectus information
(fundamental factors, signaling factors and ex-ante risk factors) has any explanatory
power to resolve this puzzle.
Beatty and Ritter (1986) find that the level of underpricing is directly
proportional to the extent of pre-IPO uncertainty, at the same time Feltham et al.,
(1991) highlight that ex-ante risk attributes published in prospectus documents have a
significant explanatory power of IPO underpricing. Various researchers unfold the
role of pre-issue earnings on the initial valuations. Firth (1998) investigates the impact
of pre-IPO earnings on the aftermarket performance. He finds that the prior IPO
earnings have the significant explanatory power of post-IPO one-year cumulative
returns but on the other hand, it loses its significance in the post-IPO three years
returns. Easton and Harris (1991) used non-IPO data to investigate the association
between the earnings and the post-issue excess returns, and find a positive and
statistically significant association. This study offers a robust evidence of significance
of earnings in the IPO valuations. The initial excess returns model only adds the book
value and earnings as fundamental factors and exclude dummy of negative earnings
and the dividends reported in prospectus to keep the IER model simple.
Su (1999) finds the positive association between the pre-IPO financial
leverage and initial excess returns. He uses three proxies (e.g., debt to total assets,
95
debt to equity, and debt to equity ratios) for financial leverage, but found only debt to
total assets ratio is significant with initial returns. As already discussed, the capital
availability is considered as an ex-ante risk determinent. A large amount of retained
earnings available to the IPO firms reveals lower risk and lower initial returns are
anticipated. Jain and Kini (1994) investigate the impact of aftermarket operating
performance on the aftermarket price performance. They argue that the operating
efficiency affects the aftermarket performance and find that less efficient firms
perform better in the post-IPO periods. Therefore, it is anticipated that the efficiency
risk has a positive association with aftermarket initial returns. Various researchers talk
about the relationship between the IPO gross proceeds and the extent of underpricing.
Some of them used IPO transaction proceeds as a size control variable. On the other
hand, this study only considered the IPO proceeds utilization plan as investments over
the net proceeds to indicate a level of risk in the future as published in prospectus
documents. The ratio of utilization plan for investment over net IPO proceeds shared
in the prospectus document is considered as the capacity risk. It is anticipated that
higher the portion of proposed utilization reveals lower the risk. It is anticipated that
there is a positive association between the capacity risk and the initial excess returns.
An extant literature examine the impact of industry beta as a proxy for firm beta on
the initial returns. Though in this study industry beta is used only for the IPO
valuation model as a proxy for firm beta due to unavailability of the market price of
newly listed firms and in the post-IPO period, firm beta is used as an explanatory
variable for underpricing. It is anticipated that the firm’s beta is positively related to
the initial excess returns.
Johnson and Miller (1988) examine the association between the level of
underpricing and underwriter prestige. They argue that IPO offered by reputed
underwriters be inclined less underpriced than the IPOs offered through less reputed
underwriters. Faltham et al., (1991) examine the impact of firm’s age and offer size
on the level of underpricingand find that firm’s age and offer size are negatively
associated with the degree of underpricing. Koh and Walter (1992) show the negative
association between the proportion of shares retained by initial shareholders and the
degree of underpricing. According to (Beatty et al., 2002), this study includes a
residual error term in the initial excess return model to examine the impact of other
factors. However, the underpricing take place as a result of various valuation methods
96
on IPO pricing and unobservable variables, the residual error term from the IPO
valuation model is incorporated added into the initial excess returns model. Therefore,
there is an inverse relationship between the valuation residuals and the initial excess
returns.
The IPO initial excess returns model is devised by considering the impact of
potential fundamentals, ex-ante risk factors, signaling factors and the residuals from
the IPO valuation model on initial returns as follows:
𝑰𝑬𝑹𝒊 = 𝛽0 + 𝛽1
𝐸𝑖
𝑃𝑜+ 𝛽2
𝐵𝑉𝑖
𝑃𝑜+ 𝛽
3𝐹𝑖𝑛𝐿𝑒𝑣𝑖 + 𝛽
4𝐶𝑎𝑝𝑡𝑙𝑅𝑠𝑘
𝑖+ 𝛽
5𝐸𝑓𝑓𝑅𝑠𝑘
𝑖
+ 𝛽6𝐶𝑝𝑐𝑡𝑦𝑅𝑠𝑘
𝑖+ 𝛽
7𝐹𝑟𝑚𝐵𝑒𝑡𝑎𝑖 + 𝛽
8𝑂𝑓𝑓𝑟𝑆𝑖𝑧𝑒
𝑖+ 𝛽
9𝑈𝑛𝑑𝑅𝑒𝑝
𝑖
+ 𝛽10
𝐹𝑟𝑚𝐴𝑔𝑒𝑖
+ 𝛽11
𝑃𝑂𝑆𝑖 + 𝛽12
𝑅𝑒𝑠𝑖 + 𝜀𝑖
Equation (11)
Where IER is the initial excess returns on the first trading day closing price. This
study estimates initial excess returns by the market adjusted returns method and the
wealth relative method.
The mathematical equation of the market adjusted returns model can be expressed as,
𝐼𝐸𝑅 = [(1 + 𝑅𝑖,𝑡)
(1 + 𝑅𝑚,𝑡)− 1] ∗ 100
Equation (12)
Where Rit is the natural log of the first trading day closing price divided by the
IPO offer price for stock i at tth trading day and Rmt is the natural log of the first
trading day closing market index divided by the before IPO last day closing market
index for stock i at tth trading day.
The mathematical equation of wealth relative model can be expressed as,
𝑊𝑅 = [1 + (1𝑛⁄ ) ∑ 𝑅𝑖,𝑡
𝑛
𝑖=1
] / [1 + (1𝑛⁄ ) ∑ 𝑅𝑚,𝑡
𝑛
𝑚=1
]
Equation (13)
Where Rit is the return of stock i at the tth trading day and Rmt is the market
returns for stock i at the tth trading day, n is the number of IPO firms in the sample.
97
The value of wealth relative model is above one reflects that the IPO firms
outperformed the market and less than one implies that the IPO firms underperformed
on the tth trading day.
The fundamental variables used in the IER model are book value and the
earnings, both are deflated by IPO preliminary offer prices. Now, these variables have
become the book value to market value ratio and the earnings to market price ratio.
The signaling and ex-ante risk factors are already defined in table 3.4 while Resit is
used as a proxy for unobservable variables of stock ‘i’ and estimated by the
standardized residual values from IPO valuation model.
3.5.1 Hypotheses about the Initial Excess Returns Model
Based on the theoretical foundations discussed regarding the initial returns, a
number of working hypotheses are proposed about variables used in the initial excess
returns model. The book value of shareholder’s equity of IPO offer price ratio is used
to identify, either stock is estimated as neither undervalued nor overvalued. Therefore,
it is anticipated that higher the book value to offer price ratio possibly increase the
underpricing. Beneda & Zhang (2009) conclude that the higher estimates of book
value disclosed boost investor’s confidence to invest more in initial trading periods.
Easton & Harris (1991) argue that earnings have a significant explanatory power in
aftermarket performance. They suggest that there is a positive association between
earnings forecast and short run returns.
H14a: There is a positive relationship between the book value of shareholder’s
equity over offer prices and the initial excess returns.
H14b: There is a positive relationship between the earnings before IPO over
offer prices and short-term returns.
An extant literature on IPO studies (Reber and Vencappa, 2016; mumtaz,smith &
ahmed, 2016; Banerjee, Dai & Shrestha, 2011; Lowry, officer & Schwert, 2010; afza,
Yousaf & alam, 2013; Miller and Reilly, 1987; Beatty and Ritter, 1986; Ritter, 1984),
advocate that the firms with greater ex-ante risk experience larger initial excess
returns. Mumtaz, Smith & Ahmed (2016) and Hedge & Miller (1996) find the inverse
correlation between the degree of financial leverage and the degree of underpricing.
Reber & Vencappa (2016), which investigate the positive and significant relationship
98
of the uses of IPO proceeds as investment related to initial excess returns. The offer
size is generally used as a proxy for the level of risk of the IPO firms (Aggarwal, Liu
and Rhee, 2008: Bessler and Thies, 2007). The firms offered large offer size in the
IPOs is riskier and in turn higher initial excess returns.
H15a: The average pre-IPO financial leverage is positively related to the IPO
initial excess returns.
H15b: The firm’s capital availability risk before the IPO is negatively
associated with the IPO initial excess returns.
H15c: The firm’s efficiency risk before the IPO is positively associated with
the IPO initial excess returns.
H15d: The firm’s capacity risk before the IPO is positively related to IPO
initial excess returns.
H15e: The firm’s beta before the IPO is positively related to IPO initial excess
returns.
H15f: The offer size in the IPO is negatively related to the IPO initial excess
returns.
Carter et al., (1998) examined the association between the investment banker
reputation and the level of underpricing. They conclude that the IPO offered by
reputed investment bankers tends to lower the underpriced. Johnson and Miller
(1988), and Carter and Manaster (1990) used different proxies for underwriter
reputations and document a negative association between the underwriter prestige and
initial excess returns. Lowry, officer & Schwert (2010), Kerins, Kutsuna & Smith
(2007), and Beatty & Ritter (1986) investigate that firm age is an important
determinant to explain the variation in the underpricing. They presumed that firm’s
age is a non-financial risk factor which may impact the aftermarket performance.
Older firms are understood to have more experience and market share tend to be less
risky than the younger firms. The portion of share capital retained by the initial
shareholders at the time of an IPO is a major signaling factor. Existing studies on IPO
suggest that the insiders keep large portion of shares capital in the IPO is a positive
99
signal to the other market participants and the portion of shares offered to general
public is used as an inverse proxy of shares retained by insiders.
H16a: The variable for underwriter reputation is negatively related to the IPO
initial excess returns.
H16b: There is an inverse relationship between the firm’s age and the IPO
initial excess returns.
H16c: There is a positive relationship between the percentage of shares
offered to the general public and the IPO initial excess returns.
The initial excess returns model also includes the residual term to unobservable
variables as measured by error terms of the IPO valuation model. A positive residual
point out that the IPOs were offered at higher prices relative to their fundamentals
than to the other IPOs. This implies that there is a negative association between the
residuals and initial excess returns.
H17: The residuals from the IPO valuation model is negatively related to the
IPO initial excess returns.
3.6 The Long-run Performance Model
The rationale to construct the long-run returns model is to determine the cross-
sectional determinants of long-run underperformance.. It is expected that the
investors, who buy shares at offer prices and keep them for three to five years, to be
earn small returns. Furthermore, Ritter (1991) first time reveals that the level of
underpricing is negatively linked to the long run performance of the US IPO firms.
This implies that the IPOs outperformed the market in the short run tend to have
smaller returns in the long-run. At the same time, the literature shows mixed results
about long-run underperformance. But signaling theories suggest that the firms with
good quality signals used underpricing as a signal for the quality product market. It
indicates that the high-quality firms with greater underpricing are expected to do
better in the future and breed larger returns in the long run.
The positive earnings in the preceding years disclosed in the prospectus are
typically observed as a signal of high-quality firms and add value in the evaluation
process. The firms with a strong earnings record assume to keep performing well in
100
future. So, it can be conclude that the earnings disclosed in prospectus has a positive
on aftermarket performance in the long run. This study also examined the impact of
book value of shareholders’ equity on the post-issue performance. An extant literature
shows mixed finding on the association between book to market value ratio and the
aftermarket performance. Fama and French (1995) argue that the book value of
shareholder’s equity over market price is also used to compute the extent of risk and
their findings highlight that book to market value ratio is positively related to the
long-run performance. They conceive that the book value to market value ratio is
insignificant determinant to explain the aftermarket returns for 2 and 3 years.
This study based on the risk-aversion hypothesis, estimates the effect of ex-
ante risk factors on the long-run performance. The financial leverage factor is
commonly used to expose the financial risk to the IPO firm and it is anticipated that
higher the financial leverage depict the IPO is high risky which in results suppose the
superior long-run returns. Khurshed et al., (1999) the first time, investigates the
impact of pre-IPO financial leverage ratio on the aftermarket performance in the long
run and their results do not find the significant relationship between them. On the
same pattern, the impact of other risk factors (e.g., capacity risk, capital availability
risk, efficiency risk and industry risk) on aftermarket performance is positive.
As already discussed in the previous section, firm’s age is viewed as a proxy
for the strength of business operations. Therefore, it is anticipated that the older firms
positively related to the long run returns. Khurshed et al., (1999) examined the
association between firm’s age and the long run performance. They show the
insignificant relationship between firm’s age and the long run returns. Carter et al.,
(1998) examined the impact of underwriter reputation in the long run performance
and find a positive and significant relationship between them. They argue that when
more information becomes available in the market, IPOs offered by less reputed
underwriters perform negatively in the long run. Brav and Gompers (1997) investigate
the impact of underwriter reputation on long-run performance. They find that the
underwriter reputation has explanatory power for long-run performance. Therefore, it
is expected that the underwriter reputation is positively related to the IPO long-run
performance. Koh and Walter (1992) find the positive relationship between the
proportion of shares retained by initial shareholders and the IPO long-run returns.
101
Based on existing literature on signaling theory, the large portion retained by the
initial owners at the time of IPO indicated the quality of the firm and good prospects.
Therefore, it is anticipated that the percentage of shares retained by initial
shareholders is positively related to the IPO long-run performance. Many studies have
examined the association between the IPO initial excess returns (underpricing) and its
aftermarket performance in the long run. The signaling hypothesis suggests that the
high-quality firms offered large offer price discount which performs better in the long
run. Therefore, based on the signaling hypothesis, the underpricing is negatively
related to the aftermarket performance in the long run. Ritter (1991) and Levis (1993)
find the negative and significant relationship between the initial excess returns and the
long run returns. Welch (1989) does not find any relation between them. This study
reexamines the relationship to narrow down the divergence of opinion on this
proposed association.
The IPO long run return model is devised by considering the impact of fundamentals,
ex-ante risk factors and signaling factors on long run performance as follows:
𝑳𝑹𝑹𝒊 = 𝛽0 + 𝛽1
𝐸𝑖
𝑃𝑜+ 𝛽2
𝐵𝑉𝑖
𝑃𝑜+ 𝛽
3𝐹𝑖𝑛𝐿𝑒𝑣𝑖 + 𝛽
4𝐶𝑎𝑝𝑡𝑙𝑅𝑠𝑘
𝑖+ 𝛽
5𝐸𝑓𝑓𝑅𝑠𝑘
𝑖
+ 𝛽6𝐶𝑝𝑐𝑡𝑦𝑅𝑠𝑘
𝑖+ 𝛽
7𝐹𝑟𝑚𝐵𝑒𝑡𝑎𝑖 + 𝛽
8𝑂𝑓𝑓𝑟𝑆𝑖𝑧𝑒
𝑖+ 𝛽
9𝑈𝑛𝑑𝑅𝑒𝑝
𝑖
+ 𝛽10
𝐹𝑟𝑚𝐴𝑔𝑒𝑖
+ 𝛽11
𝑃𝑂𝑆𝑖 + 𝛽12
𝑅𝑒𝑠𝑖 + 𝐼𝐸𝑅𝑖 + 𝜀𝑖
Equation (14)
Where LRR is the long run returns of IPO firms and measured through the: (i)
Buy and Hold abnormal returns (BHAR), (ii) Cumulative abnormal returns (CAR),
and (iii) Fama-French five factors model (FFFF) from day = 1 to x=1, 2, 3,4,5 years.
To estimate the long run IPO performance, this study employs the
methodology of buy and hold adjusted returns used by (Ritter, 1991; Loughran and
Ritter, 1995; Barber and Lyon, 1997) to estimate long-run returns. The mathematical
equation of buy and hold abnormal returns model can be expressed as,
𝐵𝐻𝐴𝑅𝑖,𝑇 = [∏(1 + 𝑅𝑖,𝑡)
𝑇
𝑡=1
− 1] − [∏(1 + 𝑅𝑚,𝑡)
𝑇
𝑡=1
− 1]
Equation (15)
102
Where BHARi,T is the buy and hold abnormal monthly returns of the firm i at t
time periods, where ∏ represents a product over T time periods, T time start from day
one to first, second and third anniversaries respectively, Rit is the monthly stock
returns of the firm i in t time period, Rmt is the market monthly returns measured
through the KSE100 index as a benchmark at t time period.
The average Buy and Holds abnormal returns for t period is defined as;
Equation (16)
Lyon et al., (1999) propose a skewness adjusted t-statistics to check the
significance of whether the mean BHAR is equal to zero, which can be estimated as
follows:
𝑡 = √𝑛 × (𝑆 +1
3ŷ𝑆2 +
1
6𝑛ŷ)
Where
𝑆 =𝐵𝐻𝐴𝑅𝑡
𝜎(𝐵𝐻𝐴𝑅𝑡) 𝑎𝑛𝑑 ŷ =
∑ (𝐵𝐻𝐴𝑅𝑖 − 𝐵𝐻𝐴𝑅)3𝑛𝑖=1
𝑛𝜎(𝐵𝐻𝐴𝑅𝑡)3
Where BHARt is the sample mean BHAR, 𝜎(𝐵𝐻𝐴𝑅𝑡) is the standard deviation of
cross-sectional sample buy and hold abnormal returns and n is the number of IPO
firms in the sample. Ŷ is an estimated value of skewness coefficient. This study
employs the skewness adjusted t-statistics to deal with the issue of skewness as the
critical values of common t-statistics are inappropriate in BHAR method.
This study employs the following methodology to estimate long-run monthly
returns are used by (Lyon, Barber & Tsai, 1999) to estimate cumulative abnormal
returns. The mathematical equation of the cumulative abnormal returns model can be
expressed as,
𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡 − 𝑅𝑚𝑡
Equation (17)
Where Rit is the monthly returns of stock i at t time event and Rmt is the monthly
returns of KSE100 index considered as a benchmark at t time event.
103
𝐴𝑅𝑡 =1
𝑛∑ 𝐴𝑅𝑖𝑡
𝑛
𝑖=1
Equation (18)
The cumulative abnormal return from event month q to event month s is the total of
the average of market adjusted returns.
𝐶𝐴𝑅𝑞,𝑠 = ∑ 𝐴𝑅𝑡
𝑠
𝑡=𝑞
Equation (19)
CAR is calculated from the closing price on the first trading day and the
cumulative mean adjusted returns for month 1 to 60. Since the cumulative adjusted
return method is less skewed than the buy and hold return method, usually t statistics
provides well-specified results. Ritter recommends the following t statistics and
estimated as:
𝑡𝐶𝐴𝑅1,𝑡= 𝐶𝐴𝑅1,𝑡 ∗ √
𝑛𝑡
𝑡 ∗ 𝑣𝑎𝑟 + 2(𝑡 − 1) ∗ 𝑐𝑜𝑣
Where nt are the event firms traded in each month, var is the mean variation of
ARi,t over 60 months and cov is the first order auto-covariance of the ARt series.
3.6.1 Hypotheses about the Long Run Returns Model
Based on the existing studies on IPO aftermarket performance, this study
attempts to investigate the impact of three groups of variables (ex-ante risk, signaling
and fundamentals factors) on long-run returns. Fama and French (1995) show a
positive relationship between the book value of shareholder’s equity to market value
and the aftermarket performance in the long run. They reveal that the book value to
market value may possibly use as a proxy for financial risk in return which is related
to the firm’s financial distress. Therefore, it is anticipated that the book to market
value multiple is positively linked with long-run performance. The earnings disclosed
in prospectus is viewed as a signal of a quality IPO in the market. An extant literature
finds a positive association of earnings published in prospectus and the aftermarket
performance. This implies that the firms reported earnings in the prospectus
outperform the market in the long run.
104
H18a: There is a positive relationship between the book value of shareholder’s
equity over offer prices and the IPO long-run returns.
H18b: There is a positive relationship between the earnings before IPO over
offer prices and the IPO long-run returns.
Fama and French (1992) argue that the firm size and the book to market value of
shareholder’s equity considered as a riskiness of the IPO firm. They used the natural
logarithm of IPO firm’s capitalization as a proxy for firm size and smaller the market
capitalization is the more risky IPO. Therefore, based on the risk and the return
theoretical assumption, the IPO offer size is negatively related to the aftermarket
performance in the long run. The existing literature (Amor and Kooli, 2017;
mumtaz,smith & ahmed, 2016; Michel, oded & shaked, 2014; Kerins, Kutsuna &
Smith, 2007) on IPOs and the risk-return theoretical assumption are used to develop a
number of hypotheses between the ex-ante risk factors and the aftermarket
performance in the long run. This implies that the long run returns model is an
increasing function of the ex-ante risk variables.
H19a: The average pre-IPO financial leverage is positively related to the IPO
long-run returns.
H19b: The firm’s capital availability risk before the IPO is negatively
associated with the IPO long-run returns.
H19c: The firm’s efficiency risk before the IPO is positively associated with
the IPO long-run returns.
H19d: The firm’s capacity risk before the IPO is positively related to the IPO
long-run returns.
H19e: The firm’s beta before the IPO is positively related to the IPO long-run
returns.
H19f: There is a negative relationship between the IPO offer size and the IPO
long-run returns.
As discussed in the previous section about underwriter reputation, the decision to hire
the prestigious underwriter indicates the signal of a quality firm. The high reputed
105
underwriters only provide consultancy to the quality firms in order to maintain their
standing in the financial markets. Therefore, it is expected that the IPO offered by the
prestigious underwriters performs well in the long run. The firm’s age is generally
considered as a proxy for IPO firms’ experience in the industry. It is expected that the
older firms have the ability to generate consistent profits. Therefore, due to the
conservatism principle in risk-return assumption, investors put a high demand an IPO
firms’ shares, which results in a higher aftermarket performance. It is expected that
the most experienced firms in terms of age are positively linked to the IPO long-run
performance. The percentage of shares retained by initial shareholders is usually
considered as a key signal of the quality firm. It is accepted that the information about
the initial shareholder’s capital structure at the time of IPO transaction reflects the
insider information about the prospects of the issuing firm. Therefore, it is anticipated
that the inverse proxy used as a percentage of shares offered for a portion of shares
retained by initial shareholders is negatively related to the IPO long-run returns.
H20a: The underwriter reputation is positively related to the IPO long-run
returns.
H20b: There is a positive relationship between the age of IPO firms and the
IPO long-run returns.
H20c: There is a negative relationship between the percentage of shares
offered to the general public and the IPO long-run returns.
The IPO long-run returns model includes residual error from valuation model and
initial excess returns as explanatory variables to control the impact of IPO mispricing
on the first day and a mean reversion effect takes place in the market respectively. In
this study, it is expected that the market is efficient and any mispricing on the first day
is ultimately corrected in the long run returns. Therefore, it is anticipated that the
positive values of residual are negatively linked to the aftermarket performance in the
long run. An extant literature on signaling theory suggests that the larger offer price
discount in the IPO offer prices by the high-quality firms used as a signal and perform
better in the long run as compared to the low-quality firms. Therefore, it is expected
that the initial excess returns are negatively linked to the IPO long-run returns.
106
H21: There is a negative relationship between the valuation residuals and the
IPO long-run returns.
H22: There is a negative relationship between the initial excess returns and
the IPO long-run returns.
3.6.2 CAPM, Fama-French Three- and Five-Factor Models
This study used the Sharpe-Lintner (1964) capital asset pricing model
(CAPM), Fama-French (1993) three-factor (FF3F) model and Fama-French (2015)
five-factor (FF3F) model to test the long-run abnormal performance of IPO portfolio
using the calendar-time regression approach. The Fama-French (1993) show that the
capital asset pricing model of Sharp (1964) cannot explain the cross-sectional
variation in the expected returns of a security or portfolio, which related to size
(market capitalization) and the value (book to market) factors. Fama-French proposed
a three-factor model that adds the size and book to market equity factors in addition to
the market risk factor. The CAPM is based on the following time series regression:
𝑅𝑝,𝑡 − 𝑅𝑓,𝑡 = 𝛼𝑝 + 𝛽𝑝(𝑅𝑚,𝑡 − 𝑅𝑓,𝑡) + 𝜀𝑝,𝑡
Equation (20)
When Fama-French (1992) introduce two more factors in the equation (20) then new
FF3F could be based on the following time series regression.
𝑅𝑝,𝑡 − 𝑅𝑓,𝑡 = 𝛼𝑝 + 𝛽𝑝(𝑅𝑚,𝑡 − 𝑅𝑓,𝑡) + 𝑠𝑝𝑆𝑀𝐵𝑡 + ℎ𝑝𝐻𝑀𝐿𝑡 + 𝜀𝑝,𝑡
Equation (21)
Where Rp,t is the value-weighted and/or equally-weighted returns on the IPO
portfolio in month t, Rf,t is the 3-months T-bills rate in the month t in Pakistan, Rm,t is
the PSX benchmark index (KSE100) returns in month t, SMBt is the difference
between the returns of value-weighted portfolio of small stocks and the big stocks in
the month t, HMLt is the difference between the returns of value-weighted portfolio of
high book to market equity stocks and the low book to market equity value stocks in
the month t. βp, sp and hp are the coefficients of market, size and value premium
factors. The intercept (αp) term is employed to test the null hypothesis that the mean
monthly excess return equals zero.
After the 20 years, more recent literature shows that the FF three-factor model
is also unable to explain the cross-sectional variation in expected returns of a security
107
or portfolio particularly related to profitability and investment along with other
anomalies. These anomalies include momentum (Jegadesh and Titman, 1993),
maximum daily returns (Bali et al., 2011), accruals (Sloan, 1996), idiosyncratic
volatility (Ang et al., 2006), net share issues (Loughran and Ritter, 1995; Ikenberry et
al., 1995) and liquidity risk (Pastor and Stambaugh, 2003).
Based on the extant literature on the Fama-French three-factor related
anomalies, Fama-French (2015) introduce five-factor model by adding profitability
and investment factors in the existing Fama-French three-factor model as:
𝑅𝑝,𝑡 − 𝑅𝑓,𝑡 = 𝛼𝑝 + 𝛽𝑝(𝑅𝑚,𝑡 − 𝑅𝑓,𝑡) + 𝑠𝑝𝑆𝑀𝐵𝑡 + ℎ𝑝𝐻𝑀𝐿𝑡 + 𝑟𝑝𝑅𝑀𝑊𝑡 + 𝑐𝑝𝐶𝑀𝐴𝑡
+ 𝜀𝑝,𝑡
Equation (22)
Where RMWt is the difference between the returns of the value-weighted
portfolio of robust profitability stocks and weak profitability stocks in the month t,
CMAt is the difference between the returns of the value-weighted portfolio of
conservative investment stocks and aggressive investment stocks in the month t. rp,
and cp are the coefficients of profitability and investment factors. The intercept (αp)
term is employed to test the null hypothesis that the mean monthly excess return
equals zero.
Equation (20), Equation (21) and Equation (22) models are estimated using the
Newey-West HAC (Heteroskedasticity and autocorrelation) consistent standard errors
to compute the t-statistics for the FF3F and FF5F model coefficients. To estimate the
parameters of market, size, value, profitability and investment factors, we collect
monthly data of 225 non-IPO firms listed on the PSX from 2000 to 2017.
3.6.2.1 FF 3-Factor Variables construction
The Fama-French three-factor model is assembled using the six value-
weighted portfolios designed on the basis of size and book-to-market ratio. SMBFF3F
is the average return on the three small portfolios minus the average return on the
three big portfolios. HML is the average returns on the two value portfolios minus the
average return on the two growth portfolios.
𝑆𝑀𝐵𝐹𝐹3𝐹 =1
3(𝑆𝑚𝑎𝑙𝑙_𝑉𝑎𝑙𝑢𝑒 + 𝑆𝑚𝑎𝑙𝑙_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝑆𝑚𝑎𝑙𝑙_𝐺𝑟𝑜𝑤𝑡ℎ) −
1
3(𝐵𝑖𝑔_𝑉𝑎𝑙𝑢𝑒
+ 𝐵𝑖𝑔_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝐵𝑖𝑔_𝐺𝑟𝑜𝑤𝑡ℎ)
108
𝐻𝑀𝐿 =1
2(𝑆𝑚𝑎𝑙𝑙_𝑉𝑎𝑙𝑢𝑒 + 𝐵𝑖𝑔_𝑉𝑎𝑙𝑢𝑒) −
1
2(𝑆𝑚𝑎𝑙𝑙_𝐺𝑟𝑜𝑤𝑡ℎ + 𝐵𝑖𝑔_𝐺𝑟𝑜𝑤𝑡ℎ)
Rm,t –Rf,t is the excess returns of the market, value weighted monthly returns of
KSE100 index and three-month treasury bills rate of SBP. SMBFF3F and HML for July
of year t to June of year t+1 include 225 firms of PSX for which we have market
equity data for December of t-1 and June of t, and book equity data for t-1.
3.6.2.2 FF 5-Factor Variables construction
The FF five-factor are assembled using the six value-weighted portfolios
designed on the basis of size and book-to-market ratio, size and operating
profitability, and size and investment factors each.
𝑆𝑀𝐵𝐵/𝑀 =1
3(𝑆𝑚𝑎𝑙𝑙_𝑉𝑎𝑙𝑢𝑒 + 𝑆𝑚𝑎𝑙𝑙_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝑆𝑚𝑎𝑙𝑙_𝐺𝑟𝑜𝑤𝑡ℎ) −
1
3(𝐵𝑖𝑔_𝑉𝑎𝑙𝑢𝑒
+ 𝐵𝑖𝑔_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝐵𝑖𝑔_𝐺𝑟𝑜𝑤𝑡ℎ)
𝑆𝑀𝐵𝑂𝑃 =1
3(𝑆𝑚𝑎𝑙𝑙_𝑅𝑜𝑏𝑢𝑠𝑡 + 𝑆𝑚𝑎𝑙𝑙_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝑆𝑚𝑎𝑙𝑙_𝑊𝑒𝑎𝑘) −
1
3(𝐵𝑖𝑔_𝑅𝑜𝑏𝑢𝑠𝑡
+ 𝐵𝑖𝑔_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝐵𝑖𝑔_𝑊𝑒𝑎𝑘)
𝑆𝑀𝐵𝐼𝑁𝑉 =1
3(𝑆𝑚𝑎𝑙𝑙_𝐶𝑜𝑛𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑣𝑒 + 𝑆𝑚𝑎𝑙𝑙_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝑆𝑚𝑎𝑙𝑙_𝐴𝑔𝑔𝑟𝑒𝑠𝑠𝑖𝑣𝑒)
−1
3(𝐵𝑖𝑔_𝐶𝑜𝑛𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑣𝑒 + 𝐵𝑖𝑔_𝑁𝑒𝑢𝑡𝑟𝑎𝑙 + 𝐵𝑖𝑔_𝐴𝑔𝑔𝑟𝑒𝑠𝑠𝑖𝑣𝑒)
𝑆𝑀𝐵𝐹𝐹5𝐹 =1
3(𝑆𝑀𝐵𝐵/𝑀 + 𝑆𝑀𝐵𝑂𝑃 + 𝑆𝑀𝐵𝐼𝑁𝑉)
Where SMBFF5F is the average returns on the nine small portfolios minus the
average return on the nine big portfolios. To estimate SMBFF5F (size factor), design a
size portfolio in July of year t for each IPO went public during year t, stocks are
sorted on the basis of market capitalization as at the end of June for each IPO listing
year. The two portfolios are constructed (small and big) based on the market
capitalization is greater or below the median and readjust with average returns
estimated using the value-weighted technique.
𝐻𝑀𝐿 =1
2(𝑆𝑚𝑎𝑙𝑙_𝑉𝑎𝑙𝑢𝑒 + 𝐵𝑖𝑔_𝑉𝑎𝑙𝑢𝑒) −
1
2(𝑆𝑚𝑎𝑙𝑙_𝐺𝑟𝑜𝑤𝑡ℎ + 𝐵𝑖𝑔_𝐺𝑟𝑜𝑤𝑡ℎ)
109
HML is the average returns on the two value (high book to market) portfolios
minus the average return on the two growth (low book to market) portfolios. The B/M
ratio sorted using the market capitalization at the end of December of t-1 and the B/M
ratio for the fiscal year ending in the calendar year t-1 for each IPO firm listed during
the calendar year t-1. Three portfolios designed as using the cut-off points of 30 and
70 percentiles and readjust with average returns estimated using the value-weighted
technique. This study form six portfolios from the intersection of two size and three
value factor portfolios (Small_Low, Small_Neutral, Small_High, Big_Low,
Big_Neutral, Big_High).
𝑅𝑀𝑊 =1
2(𝑆𝑚𝑎𝑙𝑙_𝑅𝑜𝑏𝑢𝑠𝑡 + 𝐵𝑖𝑔_𝑅𝑜𝑏𝑢𝑠𝑡) −
1
2(𝑆𝑚𝑎𝑙𝑙_𝑊𝑒𝑎𝑘 + 𝐵𝑖𝑔_𝑊𝑒𝑎𝑘)
RMW is the average return on the two high (robust) operating profitability
portfolios minus the average return on the two low (weak) operating profitability
portfolios. The profitability factor contains accounting data for the year ended
December t-1 and defined as the difference of annual revenues minus cost of goods
sold, administrative expenses and selling expenses divided by total assets for the year
t-1. Three portfolios designed as using the cut-off points of 30 and 70 percentilesand
readjust with average returns estimated using the value-weighted technique. This
study forms six portfolios from the intersection of two size and three profitability
factor portfolios (Small_Robust, Small_Neutral, Small_Weak, Big_Robust,
Big_Neutral, Big_Weak).
𝐶𝑀𝐴 =1
2(𝑆𝑚𝑎𝑙𝑙_𝐶𝑜𝑛𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑣𝑒 + 𝐵𝑖𝑔_𝐶𝑜𝑛𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑣𝑒) −
1
2(𝑆𝑚𝑎𝑙𝑙_𝐴𝑔𝑔𝑟𝑒𝑠𝑠𝑖𝑣𝑒
+ 𝐵𝑖𝑔_𝐴𝑔𝑔𝑟𝑒𝑠𝑠𝑖𝑣𝑒)
CMA is the average return on the two conservative investment portfolios
minus the average return on the two aggressive investment portfolios. CMA
(investment factor) defined as the percentage change in total assets in the fiscal year t
compared to the fiscal year t-1. Three portfolios designed as using the cut-off points
of 30 and 70 percentilesand readjust with average returns estimated using the value-
weighted technique. This study forms six portfolios from the intersection of two size
and three investment factor portfolios (Small_Conservative, Small_Neutral,
Small_Aggressive, Big_Conservative, Big_Neutral, Big_Aggressive).
110
4. Chapter 4
4. Results and Discussion
111
The chapter 4 splits into five sections. Section 1, explains the power of pre-
IPO valutation estimates to enlighten the cross-sectional variation in the post-IPO
market value of each valuation method and also probe the firm-specific characteristics
and stock market-related factors that had influenced the choice, bias and accuracy of
each valuation method employed by underwriters when valuing IPOs. Section 2,
explains the explanatory power of prospectus information (fundamentals, ex-ante risk
and signaling factors) on the IPO initial prices (valuation) and the cross-sectional
analysis of various valuation models. Section 3, provides the comprehensive analysis
of short-run abnormal returns also known as “underpricing” and to probe the effect of
prospectus information on the short-run returns. Section 4, provides the insights of
long-run aftermarket performance up to five years using Event-time (BHARs and
CARs) approaches to estimate the long-run aftermarket performance also known as
“long-run underperformance” and to probe the effect of prospectus information on the
long-run returns. Section 5, provides the robustness insights of LRR using Jenson’s
Alpha (intercept term) estimated through the Calendar-time approach. In this study,
Jenson’s Alpha has been estimated through capital asset pricing model proposed by
Sharpe-Lintner (1964), Fama-French three- and five factor models proposed by Fama
& French (1993, 2015). This chapter starts with the summary statistics of variables
followed by the quantitative analysis, hypotheses testing and debate on empirical
findings in all the sections are discussed.
4.1 The Choice, Bias and Accuracy of Valuation Methods
This section presents the insights of IPO valuation process used by Pakistani
lead underwriters and to uncover the relevant determinants of valuation methods. This
section explains the power of pre-IPO valuation estimates to explain the cross-
sectional variation in the post-IPO market prices of each valuation method and also
probe the firm-specific characteristics and market factors that had influenced the
choice, bias and accuracy of each valuation method employed by underwriters when
valuing IPOs.
4.1.1 Explaining the Choice of Valuation Methods
In this part, this study addresses the first research objective of “to investigate
the firm-specific characteristics and stock market-related factors that had influenced
the choice of valuation methods when valuing IPOs.”. This study also reviews how
112
lead underwriters add value estimates derived through pre-issue valuation process to
estimate the fair value estimates and determine the preliminary offer price of a given
IPO ( see section 3.3.1).
4.1.1.1 Descriptive Statistics
In relation to firm-specific characteristics used in the binary logit and cross-
sectional regression models, this study provides the descriptive statistics in Table 4.1.
As discussed in the earlier chapter of this study, data have been obtained from the
PSX DataStream and prospectuses (published at the time of formal listing in the
exchange) of IPO firms.
Table 4. 1: Descriptive Statistics of IPO Firm’s characteristics
Variable Name Mean Min Percentiles
Max SD N 25th 50th 75th
Total Assets (millions) 28,25 15.00 1,235 2,715 16,67 691,99 94,57 88
Firm Age (Years) 15.12 1.30 3.50 8.00 19.00 78.00 17.61 88
Property, Plant & Equip.(AIP) 40.08 0.001 8.964 36.98 68.19 99.863 31.46 88
Profitability (%) 20.67 -165.59 4.013 16.16 36.34 137.52 38.21 88
Sales Growth (%) 42.17 -48.338 0.00 18.90 55.89 740.68 95.07 88
Dividend Payout (%) 20.25 0.000 0.00 0.00 41.63 100.00 29.22 88
Market Returns (%) 7.859 -63.807 1.036 9.468 17.70 51.850 19.27 88
Ex-ante Uncertainty (%) 1.297 0.667 0.924 1.159 1.586 2.538 0.503 88
Dilution Factor (%) 23.31 2.500 14.92 25.00 27.87 50.000 10.71 88
Table 4.1 reports the summary statistics of IPO firms’ characteristics
employed in Binary Logit and Cross-sectional ordinary least squares regression
models to estimate the choice, bias and accuracy of each valuation method employed
by underwriters in the IPO valuation process. The financial variables data such as
total assets, firm age, property, plant & equipments, profitability, predicted sales
growth, dividend payout and dilution factor are taken from the latest financial
statements disclosed in the prospectus documents while market returns and ex-ante
uncertainty are calculated using the benchmark index of PSX. The descriptive
statistics of total assets (SIZE) demonstrate that on average total assets before the IPO
year is 28,25 (millions PKR) which symbolize the firm size when decided to go
public. But the values of percentiles and standard deviation depicts that the most of
the firms are smaller in size due to greater variability in the firm size. Another
implication of this lump is due to several privatization IPOs carried out in the sample.
113
The extant literature on IPO enlightens that the youngest firms went public to expand
their business operations and to capitalize their growth opportunities. The firm age is
computed as the difference between the date of incorporation and formal listings for
trading at the exchange. It is witnessed that the minimum IPO firm age is 15 months
at the time of formal listing and the maximum age is 78 years old when it went public.
Due to the immense gap between the young and old firms, the natural logarithm of
age i-e Log(1+age) is employed as a proxy for age risk factor. This study observed
that the average age of newly listed firms is about 15 years. It is recorded that the
most firms added in the sample are young firms. The descriptive statistics of assets in
place (AIP) is estimated as the ratio of property, plant & equipment over total assets
in the latest financial year before the IPO. The average (median) of the AIP is 48.09%
(36.98%) in the latest financial year before the IPO tend to indicate that more than
50% firms’ fixed assets are below the median. In this study, about 40% IPO firms are
linked to the telecommunication and financial services sectors who invest more in
working capital rather than the fixed tangible assets.
Table 4.1 documents the descriptive statistics of forecasted profitability
(PROF) described as most latest year predicted operating earnings over latest years’
predicted sales before the IPO. The average (median) forecasted profitability is equal
to 20.68% (16.16%) before the IPO. Only 15 out of 88 IPO firms are predicted to
document negative earnings in latest financial statement and this study is not lead by
loss-making IPO firms. This indicates that the most profitable firms decide to go
public to raise long-term equity capital to capitalize the future opportunities. Further,
the predicted sales growth (GROW) are used as a proxy for growth opportunities in
most recent year and onwards. The average (median) predicted sales growth in sample
is 42.18% (18.91%) in the latest financial year. The results are consistent with the
argument of rapidly growing firms face challenges of cash imbalances in the short to
medium term because the capital investments are more than the cash inflows.
Pakistani firms that decide to go public as a rule disclosed their historical payouts
and/or associated companies in the prospectuses and this information helps investors
to identify the dividend policy in the future. As per Security and Exchange
Commission of Pakistan (SECP) policies, only asset management companies are
liable to pay above 80% dividend to get tax shield benefits. The average (median)
dividend payout is of 20.26% (0%) indicates that only 36 out of 88 firms have
114
announced the dividends in the latest financial year before the IPOs. In an earlier
discussion, findings show that most firms of sample are young in age and have low
profits.
Table 4.1 also documents the summary statistics of market returns (MktRet)
during six months before the IPO. A persistent rise in aggregate stock returns may
point out the window of opportunity for the growth companies to get the benefits of
high valuations in the market. This study adds market returns during a six month
interval from 185 trading days prior to IPO and 5 days before the formal listing of
IPO firm. The average (median) market returns before the IPO is 7.86% (9.47%),
positive returns before the IPO open the ‘window of opportunity’ to growth
companies. It is observed that, in a high volatile market, the investors are more
indecisive about the intrinsic values of new offerings. The ex-ante uncertainty (SD) is
estimated as a standard deviation of the benchmark index (KSE100) of daily returns
during a six month interval before the IPO. The average (median) ex-ante uncertainty
equals 1.30% (1.16%) in the preceding months before the IPO. Next, this study
examines the dilution factor is estimated through the percentage of shares offered over
total post-issue outstanding shares at the time of listing. The average (median)
dilution factor equals 23.32% (25%) at the time of formal listing. The large fraction of
equity sold to general public indicates that poor quality firms and underwriters, prefer
to use multiples valuation to value firms’ equity as per needed valuations.
4.1.1.2 The Univariate Analysis
The key reason of univariate analysis is to explain the associations between
the variables used in the binary logit and cross-sectional regression models. Table 4.2
demonstrates the coefficients of correlation of variables employed in the choice, bias
and accuracy of valuation models and t-statistic are reported below each correlation
coefficient. The correlation coefficient between the firm age and firm size appears to
be positive and strongly statistically significant at the 99% level of confidence.
However, the correlation coefficient appears to be modest (0.4173). This implies that
the level of maturity tends to increase the size of the firm and disclosed data on age
and size come into view by stakeholders in their pricing decision during the IPO
process.
115
Table 4. 2: Correlation Matrix of Variables used in Binary Logit & Cross-sectional Models
Total Assets Age AIP Profitability Sales Growth Dividends Technology Mkt Return Ex ante Und Rep
Total Assets 1
Age 0.4150 1
4.230***
AIP 0.0882 -0.0017 1
0.821 -0.016
Profitability 0.0827 0.1662 -0.1919 1
0.770 1.563 -1.813*
Sales Growth -0.2449 -0.2374 -0.1890 -0.0994 1
-2.342** -2.266** -1.785* -0.927
Dividends 0.1880 0.2727 -0.1104 0.2624 -0.0892 1
1.775* 2.628** -1.030 2.522** -0.831
Technology -0.0230 0.0731 0.0561 -0.1595 0.2257 0.0417 1
-0.213 0.679 0.521 -1.498 2.148** 0.387
Mkt Return -0.0192 -0.1808 -0.0664 -0.1270 0.0208 0.0270 0.0311 1
-0.178 -1.704* -0.618 -1.188 0.193 0.251 0.289
Ex ante -0.0516 -0.1072 0.0401 0.0261 0.1632 -0.0653 -0.0174 -0.3382 1
-0.479 -0.999 0.372 0.242 1.534 -0.607 -0.161 -3.332***
Und Rep -0.0593 -0.0469 -0.0249 -0.0994 0.1229 0.0696 -0.0254 -0.0327 -0.2312 1
-0.551 -0.435 -0.231 -0.926 1.148 0.647 -0.235 -0.303 -2.204**
Dilution Factor -0.5830 -0.3571 -0.1256 -0.0917 0.1013 -0.1572 -0.1720 0.0052 -0.0531 0.1728
-6.653*** -3.545*** -1.174 -0.854 0.944 -1.476 -1.619 0.048 -0.493 1.627
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
116
The correlation coefficient of property, plant & equipment related to firm size
and age is positive but insignificant. The coefficients demonstrate that there is weak
association between the property, plant & equipment, firm age and size. The
correlation coefficient of operating profitability is positively related to the firm size
and the number of years in the industry, but these coefficients appear to be
insignificant. The amount of association related to size (0.0742) and age (0.1534) are
nominal. At the same time, the coefficient of operating profitability appears to be
negatively related to the property, plant & equipments but insignificant. The
coefficient related to asset intangibility is (-0.1919).
The coefficient of sales growth appears to be negative and significant related
to the size and age of the firm at the 5% level each respectively. The value of
coefficient reveals a moderate association related to size (-0.2445) and age (-0.2366)
as expected. This implies that the pace of sales growth turns down as size and age of
the firms’ increase over time. The coefficient of sales growth also appears to be
negative and insignificant linked with the fraction of property, plant & equipments
over total assets and the operating profitability. The coefficient of dividends payout
are statistically significance and positive linked with size and age of the firms at the
90% and 95% level of confidence respectively. The degree of association and
significance are consistent with the economic theory of corporate payouts as the larger
and mature firms tend to announce more dividends than to retain as capital reserves.
The coefficient of dividends payout is positively linked with operating profitability
and significant at the 1% level. The amount of profitability coefficient demonstrates a
restrained association (0.2624). This implies that the larger profitable firms pay more
net income as dividends than the less profitable firms. This study also investigates the
negative and insignificant association of dividends related to the sales growth and the
property, plant & equipments. This study has investigated that the correlation
coefficient of market returns before deciding to go public is negative and strongly
significantly related to the age of IPO firm. The amount of coefficient related to prior
IPO market returns is (-0.1808). This implies that the younger firms took advantage of
high valuations in pre-issue valuation procvss. The coefficients of market returns are
negative and insignificant to the firm’s size and operating profitability factors.
Table 4.2 presents the association of ex-ante uncertainty related to other
predictors. The coefficients of ex-ante uncertainty appear to be negatively related to
117
the size and age of IPO firms. The amount of association related to firm size is (-
0.0516) and firm age is (-0.1076) but found to be insignificant. The degree of ex-ante
uncertainty linked with property, plant & equipments, operating profitability and sales
growth are (0.0385), (0.0272) and (0.1632) respectively. The coefficient of ex-ante
uncertainty related to market returns is nnegative and significant at the 5% level. The
amount of association reveals a modest correlation between the SD and percentage of
market returns before IPO. This implies that on the average variation in daily market
returns before the IPO decrease when the market returns demonstrate high positive
returns and vice versa. The coefficient of dilution factor is is negative and strongly
significant related to the size and firm age as expected. The degree of association
demonstrates a modest correlation related to firm age (-0.5830) and firm size (-
0.3571). The association of dilution factor appears to be irrelevant related to property,
plant & equipments, operating profitability, sales growth, corporate payouts, prior
IPO market returns and ex-ante uncertainty.
In sum, the correlation matrix presents the disorder findings of the association
between each predictor. The associations of operating profitability come into sight
positive but insignificant. In addition, the results of association of sales growth and
the corporate payouts are consistent with IPO theory. The correlation of dilution
factor appears to be negative and statistically significant to the size of IPO firms and
maturity level of the IPO firms as expected. Though, the purpose of the univariate
analysis is not to examine the hypotheses related to valuation methods. The testing of
hypotheses related to binary logit models is discussed in the next sections.
4.1.1.3 The IPO valuation process
In this section, pre-IPO valuation process is discussed using a graphical and
tabular overview. Figure 4.1 describes the IPO valuation and pricing process involved
in the newly listed securities while Table 4.3 presents the frequencies and percentages
of the valuation methods used by the lead underwriters during the sample period.
118
Figure 4.1 provides an overview of the IPO valuation process and IPO pricing
process when private firms went public. After the submission of application of New
Equity Listing Application in the stock exchange, issuer firm appoints a financial
advisor to conduct a due diligence on a firm’s financial statements and core business
operations. When permission is granted by the relevant exchange, the lead
underwriter has to submit and/or publish a prospectus document which contain all the
important information about the issuing firm such as the shareholding structure of the
company, purpose of IPO proceeds, future prospects, company history, details of
trailing financial statements and valuation methods used to estimate fair value, and
deliberate discount in offer prices.
Figure 4. 1: IPO Valuation and Pricing process involve in New Offerings
Figure 4.1 described that the Pakistani lead underwriters employed dividend
discount model, discounted cash flow and comparable firms based on size and
industry to estimate the fair-value estimates. The fair-value estimates calculated using
an underwriter specific valuation method perceived as an ex-ante estimate of market
value. Then underwriters deliberately offer the discount in preliminary offer prices to
attract more participation in the auction. In Pakistan, only high net-worth individuals
(HNWIs) and Institutions are allowed to participate in the bookbuilding auction while
general investors can subscribe after the completion of bookbuilding auction. The
offer price computed through the demand of bookbuilding participants is called a
strike price and this strike price is followed by the general public portion.
119
Table 4. 3: The Summary of Valuation Methods disclosed in prospectus
Valuation Method Frequency Percentage
Multiples 65 73.864
• Price/Earnings Ratio 36 40.909
• Price/Book Ratio 51 57.955
• Price/Sales Ratio 1 1.136
• Price/EBITDA Ratio 1 1.136
Discounted Cash Flow 20 22.727
Dividend Discount Model 15 17.073
Source: Complied from prospectus documents
Table 4.3 presents the summary of valuation methods disclosed by Pakistani
underwriters in the prospectuses of 88 IPOs listed on PSX in the sample period. In
practice, most underwriters publish only widely accepted valuation models rather than
the unconventional equity valuation methods as highlighted by Fernandez (2001)11. It
has been observed that the multiples valuation approach is used almost 74% of the
times. This similar fraction is observed in the France (Roosenboom, 2007) and
Belgium (Deloof et al., 2009). The most admired multiples based on industry and/or
size is the price-book ratio, followed by the price-earnings ratio, price-sales ratio and
price-EBITDA ratio. The lead underwriters calculate the fair value estimates by
taking the product of average (or median) of specific accounting-based information
multiple with the matching accounting-based information about the IPO firm.
Usually, the investment banks used projected accounting information for ongoing and
subsequent years during multiples estimations. Furthermore, the lead underwriters
also regularly employ other accounting-based equity valuation models. Table 4.3
documents that DCF is used to calculate the fair value estimates of IPO firm’s equity
in 20 times (22.73%) and the dividend discount model is used to calculate the intrinsic
value of newly listed firm by 15 times (17.07%). The percentage of the DCF model
used in emerging economy is greater than the US lead underwriters (Houston et al.,
2006: Asquith et al., 2005) but less than the France lead underwriters (Roosenboom,
2012).
11 Fernandez (2001) talks about the valuation of Terra-Lycos (internet service provider) in 2000 by
several investment banks, which use weighted-averages of inquisitive combination of different
multiples such as number of inhabitants, gross national product (GNP) per capita, enterprise-value
(EV) per page view, capitalization per subscriber and capitalization per page view.
120
Practically, the multiples valuation is frequently used by investment banks to
value IPO firm’s equity while some irregularities have been observed with this
valuation method. First, from a theoretical viewpoint, the economic rationale of
multiples is open to discussion as the analysis is not derived by the fundamentals like
future cash flows, growth opportunities and predicted uncertainties which notify
regarding value independent of market prices. Second, the multiples valuation method
assumes that the market is efficient in order to determine prices for the comparables.
In addition, the multiples valuation method is static in nature in most of the cases,
while DCF is an appropriate measure for a dynamic business environment. From a
conceptual viewpoint, the multiples valuation technique has a problem in
implementation. It is quite difficult to recognize comparable firms having similar
business operations and firm-specific financial characteristics. The multiples
valuation leaves too much room for ‘playing with mirrors’ and open freedom for lead
underwriters to get a needed valuation. Penman (2007) argues that different multiples
offer different valuation estimates and it is difficult to assess which is most unbiased
and accurate. Pratt, Reilly and Schweihs (2000) highlight that the investment banks
fail to make necessary modifications to comparable firms’ financial statements and an
easy dependence on the median or mean of comparable firms multiples to estimate
incorrect results. Lie and Lie (2002) discuss that the DFCF is required sturdy work to
estimate appropriate forward cash flows and discount rates rather than the multiples
valuation estimates. There are some difficulties 12 that recognized related to
underwriter’s real choice of valuation techniques are found from the extant literature.
4.1.1.4 The Analysis of Binary Logit Regression Models
In this section, this study investigates the firm-specific characteristics and
stock market-related factors that had influenced the choice of each valuation method
used by lead underwriters when valuing IPOs. In this analysis, the valuation methods
employed by underwriter are used as dependent variables and t-statistics have been
estimated through Huber/White standard errors. The findings of the binary logit
12 In order to estimate the fair value of IPOs, the lead underwriter is expected to get some consensus
value using the various valuation methods. In emerging markets, the practitioners usually employ
valuation methods for valuing IPOs that are used by developed countries practitioners. The choice of
valuation method is not based on the model’s prerequisite than the desired targeted value. Therefore,
the underwriters deliberately offer discount in fair value estimate (1) to avoid law-suit from outside
investors due to bad-quality IPOs, (2) to avoid the chance of under subscription, (3) the chance of
missing material information disclosure in offering documents, and (4) to create excess demand in the
market
121
model assist to examine the cross-sectional determinants of the chosen valuation
method.
Table 4. 4: Results of Binary Logit of Preferred Valuation Methods
Multiples
Models DDM DCF Multiples P/E Ratio P/B Ratio
Independent Variables (1) (2) (3) (4) (5)
Intercept -10.9957** 9.0024 -2.7813 -13.4749** 4.3583
(-2.025) (1.236) (-0.470) (-2.099) (0.924)
Total Assets 0.7851 -0.9634
0.2420 0.9255 -0.3821
(1.374) (-1.391) (0.435) (1.583) (-0.890)
Firm Age -0.3841 -2.5898**
2.4833** 4.4598*** 0.6692
(-0.394) (-2.323) (2.328) (4.188) (0.943)
Property, plant & Equip. 0.0064 0.0307**
-0.0336** -0.0216* -0.0230**
(0.483) (2.648) (-3.032) (-1.779) (-2.558)
Operating Profitability -0.0006 0.0198*
0.0033 0.0125 -0.0125
(-0.039) (1.782) (0.509) (1.331) (-1.379)
Sales Growth -0.0026 -0.022**
0.0056 0.0001 0.0063
(-0.691) (-2.377) (0.894) (0.029) (1.149)
Dividend Payout 0.0408** -0.0119
-0.0123 -0.0184* -0.0009
(2.922) (-0.946) (-1.344) (-1.836) (-0.101)
Technology 1.0477 0.675
0.0053 -0.1599 -0.0806
(1.295) (0.905) (0.007) (-0.214) (-0.138)
Market Returns 0.0032 0.043**
-0.0317* 0.0321* -0.0244*
(0.143) (2.398) (-1.753) (1.703) (-1.667)
Ex-ante -0.7295 -0.1000
-0.0845 -0.0422 -0.1574
(-0.633) (-0.129) (-0.081) (-0.038) (-0.295)
Underwriter Reputation 1.9519*** 0.3148
0.5258 0.7142 0.3514
(2.631) (0.431) (0.817) (0.984) (0.661)
Dilution Factor -0.0213 -0.0143
0.0179 -0.0266 -0.0010
(-0.726) (-0.386) (0.483) (-0.649) (-0.032)
McFadden R2 0.3797 0.3723 0.2823 0.4349 0.1620
LR-Statistic 26.976*** 32.857** 27.149** 51.322*** 19.225*
Prob(LR-Statistic) 0.0046 0.0005 0.0044 0.0000 0.0572
N 88 88 88 88 88
The Z-Statistics are within parentheses and calculated on the basis of Robust Huber/White standard
errors. ***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
In the dividend discount model (DDM) valuation analysis, Table 4.4 reports
that the Firm age is negatively linked with the choice of DDM valuation method and
these findings contradict with the assumption of the DDM valuation theory that the
underwriters prefer to use DDM when valuing older firms. These findings contrary to
Deloof et al., (2009) argued as the dividend discount model is the best measure to
valuing stable and large firms. The Total Assets used as a proxy for firm size and
122
results unfold that the firm size is positively linked with the preference of DDM
valuation method as expected. The coefficient of total assets appears to be positive but
statistically insignificant. The result of firm size determinant provides the supporting
evidence of the proposed hypothesis of H1 which present the positive association
between the firm size and choice of DDM valuation method. The Property, plant and
equipment is also positively linked with the DDM valuation method and underwriters
more likely to use DDM when valuing large firms. The finding reveals that the
coefficient of Operating Profitability is negative and this indicates that the
underwriters less likely to use DDM when valuing firms which are relatively
profitable in the IPO year. On the similar pattern, the coefficient of Sales Growth
appears to be negative and statistically insignificant. Underwriters do not use the
DDM valuation method when firms’ sales growth is relatively less. The findings
reveal that the lead underwriters prefer to choose the DDM valuation method that
firms had offered cash and/or stock payouts in the trailing years. The coefficient of
Dividend Payout appears to be positive (0.0408) and strongly significant at the 95%
level of confidence. This implies that the investment banks prefer to choose DDM
valuation model when valuing large profitable firms which offered their major income
as corporate payouts. These findings are in sequence with working hypothesis of H6
and comparable with existing literature (Bhattacharya, 1979; Damodaran, 1994;
Roosenboom, 2007; Deloof et al., 2009; Abdulai, 2015). The coefficient of
Technology firms is positive but statistically insignificant; underwriters do not use
direct valuation for technology firms and these results are inconsistent with the DDM
valuation theory and Roosenboom (2007) findings. Bartov et al., (2002) argue that the
technology firms are probably to be assessed using multiples valuation than the direct
estimation methods such as DCF and DDM because these models do not incorporate
the value of growth options in the fair-value estimates. The Market returns are
positively linked with the preference of DDM valuation method but statistically
insignificant. According to valuation theories, relative valuation techniques are more
common when aggregate stock market returns are higher rather than the direct
valuation. The findings conjecture that the possible investors may participate
aggresively in the high dividend yield IPOs when market sentiments are bullish before
IPO. The insignificance of coefficient of Ex-ante uncertainty is appear to be negative
that indicates that the underwriters do not use the DDM valuation method when
aggregate market volatility is large before the IPO. The coefficient of Dilution Factor
123
appeared to be negative. The coefficient of Underwriter Reputation appears to be
positive (1.9519) and strongly significant at the 99% level of confidence. This implies
that the reputed underwriters prefer to use direct valuation than the relative multiples
“to leave less money on the table” during the IPO valuation process.
In the discounted cash flow (DCF) valuation analysis, Table 4.4 reports that
the lead underwriters prefer to choose the DCF valuation method for young firms. The
coefficient of Firm Age appears to be negative (-2.5898) and strongly significant at
the 95% level of confidence. This implies that the lead underwriters prefer to use DCF
valuation method when valuing young firms. The results of younger firms are
consistent with Roosenboom (2007) but contradict with the Abdulai (2015), and Kim
and Ritter (1986) as they argued that it is complicated to predict future cash flows for
young firms without trailing financial fundamentals. The results reveal that the
coefficient of Property, Plant & Equipment appears to be positive (0.0307) and
significant at the 95% level of confidence. This implies that the underwriters more
likely to use the DCF method when firms have placed their more investments as fixed
assets. The firms having more assets as fixed assets expected to generate more
revenue in the future and DCF is considered as a good measure to capitalize predicted
future cash flows to estimate the fair value estimates. These findings are in sequence
with proposed hypothesis of H3 and findings of Lev (2001) as accounting numbers
supposed to be a good estimator of the firm value from the tangible assets than the
intangible assets. The findings reveal that the coefficient of Operating Profitability
appears to be positive (0.0198) and significant at the 90% level of confidence. This
implies that there is a positive association between the operating profitability and the
underwriter’s choice to select DCF to value IPOs. The results are in sequence with the
assumption of the IPO valuation theory that the profitable firms are valued by the
direct valuation techniques. Roosenboom (2007) and Abdulai (2015) also find a
positive relationship between the choice of DCF and extent of operating profitability
before the IPO. The predicted sales growth in the IPO year is used as a proxy for
growth opportunities. The findings show that the coefficient of Sales Growth appears
to be negative (-0.0220) and significant at the 95% level of confidence. This implies
that the rapidly growing firms face challenges of cash imbalances in the short to
medium term because the capital investments are more-than the cash inflows. The
coefficient of Market Returns is (0.043) and statistically significant at 95% level of
124
confidence. These findings are in sequence of Penmen (2001), Roosenboom (2007)
and Demirakos, Strong & Walker (2010). The findings report that the Technology
firms and Underwriter Reputation are positively related to the decision of DCF
valuation method but found statistically insignificant. This study conjectures that
market sentiments before the IPO offer a “window of opportunity” when market
participants are more enthusiastic to purchase risky stocks and more willing to believe
the cash flow and discount rate suppositions factoring the DCF model. The findings
also highlight the Ex-ante uncertainty and Dilution Factor are negatively linked with
the preference of DCF valuation technique. This implies that the market participants
may be more uncertain about fundamental values during high ex-ante uncertainty
because the DCF method offers information which investors required to estimate the
fundamental value of the IPO firms. The findings reveal that the underwriters do not
use the DCF valuation method when valuing larger, older and lower growth firms.
In the multiples valuation analysis, Table 4.4 presents the findings of binary
logistic regression model when lead underwriters used comparable firms as valuation
multiples when valuing IPOs. The findings reveal that the coefficient of Firm Age
appears to be positive (2.4833) and strongly significant at the 95% level of
confidence. This implies that the lead underwriters prefer to choose the multiples
valuation method when valuing older firms. These results are inconsistent with the
proposed hypothesis that older firms are valued by direct valuation methods rather
than the comparable firms approach while findings are also evident from other
existing studies such as Roosenboom (2007), Deloof et al., (2009), Demirakos et al.,
(2010), and Kim & Ritter (1999) as they argued that it is complicated to predict future
cash flows for young firms without trailing financial fundamentals. The results reveal
that the coefficient of Property, Plant & Equipment appears to be negative (-0.0336)
and strongly significant at the 95% level of confidence. This implies that the
underwriters use multiples valuation method when firms have placed less investment
in fixed capital investment. Firms having more assets as fixed are expected to
generate more revenues in the future and vice versa, multiples technique is considered
as a good measure to discount predicted future cash flows to estimate the fair value
estimates. These findings are comparable with Abdulai (2015) but contrary to the
Roosenboom (2007). The findings show that the coefficient of Market Returns
appears to be negative (-0.0317) and weakly significant at the 90% level of
125
confidence. This study conjecture that the aggregate stock market returns before the
IPO offer a “window of opportunity” when market participants are more eager to
purchase risky stocks rather than based on their fundamental estimates. This implies
that the underwriter prefers to use relative comparable firms approach to value IPO
firms when market sentiments are bullish before the IPO. As consistent with the
valuation theories, this study finds that the lead underwriters more likely to use
comparable multiples when IPOs are relatively profitable and expected positive sales
growth in IPO year. The findings reveal that the coefficient of dividends payout and
market returns before the IPOs are negatively linked to the selection of multiples
valuation method. This implies that only mature and large profitable firms offer
corporate payouts and underwriters are prefer to use comparable multiples when firms
are immature in terms of sales and enhancing their business operations. The findings
also reveal that the prestigious underwriters opt multiples when valuing IPOs. The
dilution factor represents the percentage of shares offered by initial sponsors to the
general public and there is a positive association with the dilution factor and the post-
IPO valuation estimates. The investment banks are more likely to use comparable
multiples approach when firms having the large fraction of dilution factor at the time
of valuation process. The findings unfold that investment banks prefer to use
multiples approach when valuing technology firms.
In this study, the findings report that the underwriters use price-book and
price-earnings ratios as multiples valuation measures. This part extend our analysis to
further investigate the determinants of both P/E and P/B ratios. Table 4.4 shows the
findings of binary logit regressions when P/E and P/B used as dependent variables
separately. In the P/E analysis, the statistically significant determinants are firm age,
property plant & equipment, dividend payout and market returns before the IPO. Most
of the findings are consistent with multiples valuation analysis and comparable with
Deloof et al., (2009) and Demirakos, Strong & Walker (2010). The findings reveal
that the underwriters prefere to choose P/E valuation measure when aggregate stock
market returns are positive before IPO, don’t have the history of offer corporate
payouts, have fewer investments in fixed assets and older firms. The findings report
that the P/E multiple is positively linked with the high operating profitability, high
sales growth and the prestigious underwriter factors. On the other side, underwriters
prefer to choose a P/B valuation measure when aggregate stock market returns are in
126
bearish fashion before the IPO and firms having fewer investments as fixed assets.
The findings reveal that the prestigious underwriters prefer to choose both P/E and
P/B multiples during high ex-ante volatility in the market before the IPO. The
findings conclude that the fair value of technology firms and positive sales growth
firms are drawn from predicted growth opportunities which can easily estimate
through multiples valuation approach.
In sum, the findings of binary logit model of DDM analysis report that the
dividends payout is a key factor when underwriters employ DDM as a valuation
method. However, the firm age, asset-in-tangibility, operating profitability and
predicted sales growth are more considered factors when underwriters used DCF as a
valuation method. In multiples valuation, firm age, asset-in-tangibility and aggregate
market returns before IPOs are significant determinants. The most findings are in line
with proposed hypotheses and valuation theories.
4.1.2 Explaining the Bias of Valuation Methods
In this part, this study addresses the second research objective “to investigate
the firm-specific characteristics and market factors that had influenced the bias
associated with each valuation method”. First, the Wilcoxon Sign Rank test and
standard t-statistics on the medians and mean values of signed prediction errors of
each valuation method are used to estimate the bias associated with each valuation
method. Second, the cross-sectional regression models for each valuation method are
carryied out to investigate the firm-specific characteristics and market-related factors
that had influenced the signed prediction errors.
4.1.2.1 The Analysis of Sign Prediction Errors
In this part, the descriptive statistics of SPE of each valuation method
employed by underwriters are discussed. The SPE is computed as (Estimated
valuei,preIPO – Market Valuei) /Market Valuei, where Estimated Value is provided by
the lead underwriters in the prospectuses to compute the fair value estimates before
IPO using several valuation methods and Market Value represents the first day closing
price of IPO firm i. Kim and Ritter (1999), Francis et al., (2000) Liu et al., (2002),
Deloof et al., (2009) and Roosenboom (2012) used the closing prices of the first
trading day is to estimate the signed prediction errors. An extant literature conjectures
127
that the signed prediction errors capitalize the bias associated with each valuation
technique and these errors also used in the cross-sectional analysis of bias regressions
as a dependent variable.
Table 4.5: Analysis of Signed Prediction Errors (at 1st Day Closing Prices)
Valuation Method Mean Min Percentiles
Max SD N 25th 50th 75th
Dividend Discount Model (%) 3.99 -58.50 -22.48 -1.91 27.27 100.06 39.58 15
Discounted Cash Flow (%) -3.25 -76.30 -17.81 -0.50 13.70 36.96 27.16 20
Multiples (%) 11.21** -55.04 -17.92 5.64* 34.68 236.78 43.36 65
P/E Ratio (%) 9.88 -55.04 -23.52 5.26 34.73 236.78 50.84 36
P/B Ratio (%) 10.81** -49.03 -9.61 6.78*** 27.00 100.29 35.13 51
Fair Value Estimates (%) 8.81** -76.30 -14.92 4.18* 25.47 236.78 40.63 88
***significance at 1% level, **significance at 5% level, *significance at 10% level
Table 4.5 presents the descriptive statistics of signed prediction errors of each
valuation method by undertaking first day closing prices of IPOs as market values.
The Wilcoxon Sign Rank test and standard t-statistics on medians and mean values of
sign prediction errors different from zero are used to estimate bias attached with each
valuation method. According to an efficient market hypothesis and an extant
literature, signed prediction errors approach is the best measure to capture the bias
associated with each valuation method. The findings show that the most valuation
methods are associated with positive values of mean and median valuation prediction
errors which are significantly different from zero. The results show that the positive
valuation prediction errors for all valuation methods exceed 50%. The findings reveal
that the DDM and DCF seem to be unbiased value estimators: the median valuation
prediction errors are only (-1.92%) and (-0.50%) respectively. The median values are
not statistically significantly different from zero estimated through the Wilcoxon Sign
Rank test. This implies that the lead underwriters accurately estimate the intrinsic
value of issuing firms’ equity. These findings are consistent with Deloof, Maeseneire
& Inghelbrecht (2009, 2002), and Francis, Olsson & Oswald (2000) but contrary to
Roosenboom (2012), and Cassia, Paleari & Vismara (2004). Multiples valuation
method (combined all separate multiples) produces biased value estimates: the median
valuation prediction error is 5.64% and significant at the 10% level different from
zero. This suggests that the lead underwriters overestimate the market prices ex-ante.
On the other hand, the key issue of comparing the bias for all valuation methods is
128
that certain valuation methods more suitable than others. Furthermore, the multiples
valuation analysis is enhanced to individually analyze the measures of multiples to
examine the bias associated with each multiple measures. The finding shows that the
P/E ratio seems to be an unbiased value estimator while P/B ratio produces biased
value estimates. If market values are considered as equilibrium prices then P/B ratio
tends to overestimate value. These findings are consistent with existing literature
(Deloof, Maeseneire & Inghelbrecht, 2009, 2002; How, Lam & Yeo, 2007; Cassia,
Paleari & Vismara, 2004) while contrary with Roosenboom (2012).
Table 4. 6: Analysis of Signed Prediction Errors (at IPO Offer Prices)
Valuation Method Mean Min Percentiles
Max SD N 25th 50th 75th
Dividend Discount Model (%) 44.01** 2.63 27.27 28.95** 59.00 96.07 27.74 15
Discounted Cash Flow (%) 12.64** 0.00 0.00 0.00** 20.67 78.57 20.77 20
Multiples (%) 39.38*** -30.60 4.56 31.92* 60.33 227.60 53.56 65
P/E Ratio (%) 39.05** -29.20 -2.58 31.46* 49.00 225.38 55.31 36
P/B Ratio (%) 34.82*** -35.00 2.40 22.80* 61.67 227.60 49.07 51
Fair Value Estimates (%) 32.29*** -30.60 0.00 23.56* 46.46 227.60 45.58 88
***significance at 1% level, **significance at 5% level, *significance at 10% level
Table 4.6 presents the descriptive statistics of signed prediction errors of each
valuation method by considering the offer prices of IPO firms as market values. This
study used Wilcoxon Sign Rank test and standard t-statistics on medians and mean
values different from zero of signed prediction errors to estimate the bias associated
with each valuation method. Findings reveal that the no single valuation measure
appears to be insignificant and produces biased value estimates. Comparing each
valuation method, this study finds that the discounted cash flow method is the least
biased estimates of value. These findings are comparable with Cassia, Paleari &
Vismara (2004).
These findings raise the question that why lead underwriters intentionally
overvalue the fair-value estimates with respect to the immediate aftermarket prices
and at preliminary offer prices. One of the reasons that the lead underwriters involved
in both to estimate the fair-value estimates and to set preliminary offer prices in many
cases is to deliberately offer a large discount to the investors. This implies that the
higher offer price discounts are associated with the biased valuations. The findings
conjecture that the biased valuations of lead underwriters may offer opportunities to
the investors to impatiently participate in the subsequent offers.
129
4.1.2.2 Cross-Sectional Analysis of Bias of Valuation Methods
In this part, this study investigates the firm-specific characteristics and market
factors that had influenced the bias related to several valuation methods used by the
lead underwriters when valuing IPOs. In this analysis, the signed prediction errors are
used as dependent variables for each valuation method employed by underwriters and
t-statistics have been estimated through White (1980) heteroscedastic standard errors.
To limit our analysis, this analysis used only three most popular techniques for cross-
sectional regression analysis (DDM, DCF, multiples and fair-value valuations as a
weighted average of all valuation methods).
Table 4.7 shows the results of cross-sectional regressions of bias associated
with dividend discount model, discounted cash flow, comparable multiples and the
fair value estimates methods. In the DDM bias valuation analysis, the number of
observations is very limited and only three variables have been used on the basis of
theoretical support discussed in the research methodology chapter. Table 4.7 reports
that the coefficient of firm size appears to be negative and strongly significant at the
90% level of confidence. The finding reveals that the large firms produce less biased
valuation when followed by the DDM method. The results are comparable with
Beatty and Ritter (1986) who argue that the larger IPO firms can be easily estimated
through direct valuation models such as discounted cash flow model and dividend
discount model as they are more stable in terms of market share, revenue growth,
payout history and forecasted cash flows. The results document that the firms having
large investments as fixed show less biased valuations. This implies that the DDM is
an appropriate value estimator to determine the fair-value estimates for capital-
intensive firms. The results document that the higher market returns before the IPO
show less biased valuations when valuing through DDM method. These findings are
consistent with Roosenboom (2012) bias DDM valuation analysis.
In the discounted cash flow (DCF) bias valuation analysis, the number of
observations is very limited and only four variables have been employed on the basis
of theoretical support discussed in the research methodology chapter. Table 4.7
reports that the coefficient of firm size appears to be negative and significant at the
90% level of confidence. The results show that the large firms produce less biased
valuations when followed by DCF method. The results are consistent with the debate
of Beatty and Ritter (1986). The results document that the firms having a large portion
130
of their investments as fixed show less biased valuations. The coefficient of predicted
profitability in the IPO year appears to be positive and strongly significant at the 99%
level of confidence. This indicates that the more profitable firms produce less biased
valuations. The findings are consistent with existing literature (Cassia, Paleari &
Vismara, 2004; Francis, Olsson & Oswald, 2000). The coefficient of predicted sales
growth in the IPO year is negative and strongly significant at the 99% level of
confidence. This indicates that the higher sales growth firms produce less biased
valuations.
Table 4. 7: Cross-sectional Regressions of Bias of Valuation Methods
Model SPE_DDM SPE_DCF SPE_ Multiples SPE_Fair Value
Independent Variables (1) (2) (3) (4)
Intercept 130.6102* 9.3872 24.1169** 17.9229**
(2.159) (1.342) (2.055) (2.332)
Size -11.9973* -5.5221* -3.8731** 0.8149
(-2.1874) (2.051) (-2.105) (0.332)
Firm Age -3.9101 -6.0302
(-0.303) (-1.150)
Property, plant & Equip. 0.2575* -0.1738 0.1093 -0.0245
(2.141) (-1.432) (0.763) (-0.416)
Operating Profitability -0.0736 -0.1172
(-0.655) (-1.277)
Sales Growth 0.3084*** -0.0867 -0.0563
(4.442) (-1.449) (-1.256)
Dividend Payout -0.3675*** -0.1651 -0.0855
(-3.585) (-0.895) (-0.607)
Technology -8.8599** -8.8025
(-2.038) (-1.115)
Market Returns -1.7162** 1.3571** -0.5044**
(-8.503) (2.339) (-2.337)
Ex-ante -3.7309 16.0768
(-0.303) (1.357)
Underwriter Reputation -10.6093 2.0360
(-1.049) (0.245)
Dilution Factor 0.9944** 1.9418**
(2.039) (2.153)
Adj. R-Square 0.8223 0.6461 0.2355 0.1383
F-Statistic 13.8867*** 5.932*** 1.428 1.079
Prob(F-Statistic) 0.001 0.006 0.189 0.3893
N 15 20 65 88
t-statistics using White(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
131
In the multiples and fair value estimates bias valuation analysis, Table 4.7
reports that the coefficient of offer size is positive and statistically significant at the
95% level of confidence. This implies that smaller size firms produce large valuation
bias when valued by multiples valuation approach. The coefficient of technology is
negative and statistically significant at the 95% level of confidence, revealing that
technology firms produce more valuation biases when valued using multiples
valuation technique. The coefficient of prior market returns appears to be positive and
strongly significant at the 95% level of confidence. The findings conjecture that the
firms offered in non-crisis effect periods, bullish market fashions, produce more
valuation bias than the one offered in bearish market fashions. The coefficient of
dilution factor appears to be positive and strongly significant at the 95% level of
confidence. The results show that the more shares offered in the IPO produce more
biased valuations when followed by the multiples valuation method. The results are
consistent with the findings of Roosenboom (2012). The results document that the
higher market returns before the IPO show less biased valuations. This implies that
the lead underwriters choose higher multiples firms’ to capitalize the impact of bullish
sentiments in the fair-value estimates.
Table 4.7 presents the results of bias valuation methods estimated using the
closing prices on the first trading day as market values. This study also estimates the
results of bias valuation using IPO offer prices as market value. The results are
reported in the Appendix Table A.5. The findings of the bias DDM valuation method
is same as reported in Table 4.7 while the results of the bias DCF valuation method
are different as reported in Table 4.7. The findings show that the operating
profitability and property, plant & equipments are key drivers to produce biased
valuations when followed by the DCF method. The results report that higher predicted
sales growth firms show less biased valuations, higher pre-IPO aggregate stock
market returns show more biased valuations and prestigious underwriters show less
biased valuations when followed by the multiples valuation method.
4.1.3 Explaining the Accuracy of Valuation Methods
In this part, this study addresses the third research objective “To investigate
the firm-specific characteristics and market-related factors that had influenced the
valuation accuracy associated with each valuation method”. First, the Wilcoxon Sign
132
Rank test and standard t-statistics on medians and mean values of absolute prediction
errors of each valuation method are used to estimate the valuation accuracy of each
valuation method. Second, to explain the value relevance of pre-IPO value estimates,
perform a Wald-test to examine the joint hypothesis that the intercept and slope are
equal to zero and one respectively. If the pre-IPO value estimates are the unbiased
predictor of post-IPO market prices then the joint hypothesis of intercept and slope
should be zero and one respectively. At last, the analysis of cross-sectional regression
models for each valuation method is performed to investigate the firm-specific
characteristics and market factors that had influenced the valuation accuracy.
4.1.3.1 Analysis of Absolute Prediction Errors
In this part, the descriptive statistics of absolute prediction errors of each
valuation method are discussed. The absolute prediction errors of each valuation
method is computed as |(Estimated valuei,preIPO – Market Valuei) /Market Valuei|. The
absolute prediction errors are the absolute values of signed prediction errors as used in
4.1.2.1 subsection. Kim and Ritter (1999), Francis et al., (2000), Berkman, Bradbury
& Ferguson (2000), Deloof, Maeseneire & Inghelbrecht (2002), Cassia, Paleari &
Vismara (2004), Schreiner and Spremann (2007), Deloof et al., (2009), Demirakos,
Strong & Walker (2010), Roosenboom (2012) and Goh, Rasli, Dziekonski & Khan
(2015) used first day closing prices to estimate the valuation accuracy of each
valuation method. An extant literature conjectures that the absolute prediction errors
capture the valuation accuracy associated with each valuation technique and these
errors are also used in the cross-sectional value relevance regressions as a dependent
variable.
Table 4. 8: Analysis of Absolute Prediction Errors (at 1st Day Closing Prices)
Valuation Method Absolute Prediction Errors Percentage
within 15% of
actual estimates
Obs.
Mean Median
Dividend Discount Model (%) 28.442** 24.218** 30.769 15
Discounted Cash Flow (%) 20.279** 16.813** 45.000 20
Multiples (%) 29.880*** 23.940*** 35.385 65
P/E Ratio (%) 34.277*** 27.207*** 25.000 36
P/B Ratio (%) 27.156*** 23.940*** 41.176 51
Fair Value Estimates (%) 28.121*** 22.459*** 36.364 88
***significance at 1% level, **significance at 5% level, *significance at 10% level
133
Table 4.8 presents the mean, median and percentage of valuation errors within
15% or less of actual estimates of absolute prediction errors of each valuation method
by undertaking the first trading day closing price of IPO firms as market values. In
this analysis, the Wilcoxon Sign Rank test and standard t-statistics on medians and
mean values different from zero of absolute prediction errors respectively are used to
estimate the valuation accuracy associated with each valuation method. According to
existing literature, an absolute valuation prediction errors approach is the appropriate
measure to capture the valuation accuracy of each method by estimating the degree of
central tendency and as the percentage of observations with absolute prediction errors
within 15% or less. The findings reveal that the mean absolute errors of valuation
methods are between 20.28% and 34.28%. The results document that the t-statistics of
mean absolute errors statistically different from zero for DCF estimates are
significantly smaller (20.28% of the sample) and the degree of central tendency of
percentage within 15% is highest (45.00%) than the mean absolute errors of other
methods. This implies that the valuation accuracy of DCF is highest. The results of
DCF highest valuation accuracy are consistent with Deloof et al., (2009, 2002) and
Berkman, Bradbury & Ferguson (2000). This study also observed that the mean
absolute errors for P/E ratio estimates are significantly larger (34.28% of the sample)
and the degree of central tendency of percentage within 15% is lowest (25.00%) than
the mean absolute errors of other methods. This implies that the valuation accuracy of
P/E is smallest. The results of P/E lowest valuation accuracy are consistent with Goh,
Rasli, Dziekonski & Khan (2015), Deloof et al., (2009, 2002), Cassia, Paleari &
Vismara (2004), Berkman, Bradbury & Ferguson (2000), and Kim & Ritter (1999). In
sum, the mean absolute errors are range from 20.28% for the DCF method to 34.28%
for the P/E ratio multiples. The valuation accuracy of fair value estimates is equal to
28.12%.
Table 4. 9: Analysis of Absolute Prediction Errors (at IPO Offer Prices)
Valuation Method Absolute Prediction Errors Percentage
within 15% of
actual estimates
Obs.
Mean Median
Dividend Discount Model (%) 44.011** 28.947** 7.692 15
Discounted Cash Flow (%) 12.639** 0.000** 65.000 20
Multiples (%) 46.349*** 31.920*** 23.077 65
P/E Ratio (%) 47.467*** 31.460*** 16.667 36
P/B Ratio (%) 42.770*** 30.600*** 27.451 51
Fair Value Estimates (%) 37.189*** 25.932*** 31.818 88
***significance at 1% level, **significance at 5% level, *significance at 10% level
134
Table 4.9 presents the mean, median and percentage of valuation errors within
15% or less of actual estimates of absolute prediction errors of each valuation method
by undertaking the IPO offer prices as market values. The section estimate the degree
of central tendency of pre-IPO value estimates at IPO offer prices and defined as the
percentage of observations with absolute prediction errors within 15% or less. The
findings reveal that the mean absolute errors of valuation methods are between
12.64% and 47.47%. The results of mean absolute errors estimated through the IPO
offer prices are similar to the findings of mean absolute errors estimated through the
first day closing price of IPO firms. The findings show that the mean absolute errors
are ranging from 12.64% for the DCF method to 47.47% for the P/E ratio multiples.
The valuation accuracy of fair value estimates is equal to 37.19%
These findings raise the question that why do lead underwriters intentionally
overvalue the fair value estimates by using P/E and P/B multiples instead of using
DCF valuation method. This may be one of the reasons that the lead underwriters
involve in both to estimate fair value estimate and set preliminary offer prices in many
cases and deliberately offer a large discount to the investors. The findings conjecture
that the lead underwriters may give opportunities to the investors to impatiently
participate in the subsequent offers in the future.
4.1.3.3 The Analysis of Value Relevancy of each Valuation Model
This study examines the predicting power of pre-IPO value estimates of each
valuation method to explain the cross-sectional variation in the market values. To
explain this value relevancy, this study employed the Wald-test to examine the joint
hypothesis that the intercept equals to zero and the slope coefficient is equal to one. If
the valuation models give unbiased estimates then the intercept should be equals zero
and slope coefficient equals one. In the value relevancy models, the natural logarithm
of Market Value of IPO firms is used as the dependent variable and the natural
logarithm of Estimated Value disclosed in the prospectuses for each valuation model
as the independent variable. In Table 4.10, the t-statistics estimated through White
(1980) heteroscedastic standard errors are reported in the parentheses and test whether
the intercept term is different from zero and slope coefficient is different from one or
not.
135
Table 4. 10: The Analysis of Value Relevancy Through Regressions
Indep. Variable Parameter Adj. R2
(%) N Wald test
Intercept Slop
Dividend Discount Model 0.0354 0.9856** 73.00 15 0.0710
(0.102) (4.317)
Discounted Cash Flow 0.2869 0.7846*** 47.94 20 1.2122
(1.535) (5.670)
Multiples Valuation 0.0423 0.9561*** 83.25 65 0.7378
(0.619) (18.985) P/E Ratio
0.0992 0.9346*** 72.60 36 0.3521 (0.741) (11.534)
P/B Ratio 0.1014 0.9084*** 86.41 51 2.5404*
(1.624) (19.965)
Fair Value Estimate 0.1047 0.9168*** 81.58 88 1.5652
(1.570) (18.990) ***significance at 1% level, **significance at 5% level, *significance at 10% level
Table 4.10 documents the finding of Wald-test and the explanatory power for
each valuation method. On the investigation of ability to explain the predicting power
of each valuation method, the study tests the hypothesis that the slope coefficient
equals one. The findings reveal that the slope coefficient is different from one. The
results report that the multiples (importantly P/B measure) valuation method has
highest predicting power and the discounted cash flow method has lowest predicting
power to market values. These findings are consistent with Cassia, Paleari & Vismara
(2004), and Kim and Ritter (1999). On the other side, Wald-statistic findings reject
the joint hypothesis of the intercept that equals zero and slope coefficient equals one
for each valuation method. The findings of Wald-test of joint hypothesis, none of the
valuation method produces unbiased estimates of market value. The findings of Wald-
statistic are consistent with the Roosenboom (2012).
This study also conduct the value relevancy regression analysis at IPO offer
prices used as market values. Table A.6 (see appendix) reports the finding of the
explanatory power of each valuation method and Wald-statistic to test whether the
intercept term and slope coefficients are different from zero and one respectively. The
findings are similar as observed in the Table 4.10.
136
4.1.3.3 Cross-Sectional Analysis of Accuracy of Valuation Methods
In this part, this study investigates the firm-specific characteristics and market
factors that had influenced the valuation accuracy of each valuation method used
during the IPO valuation process. In this analysis, the absolute prediction errors are
used as dependent variables with respect to the valuation method employed by
underwriters and t-statistics are estimated through White (1980) heteroscedastic
standard errors. To limit our analysis, only three most popular techniques are used for
cross-sectional regression analysis.
Table 4.11 presents the results of the cross-sectional regression of valuation
accuracy of the dividend discount model, discounted cash flow, multiples valuation
and fair value estimates methods. In the dividend discounted model (DDM) valuation
accuracy analysis, Table 4.11 reports that the coefficient of firm size appears to be
negative and strongly significant at the 95% level of confidence. The finding reveals
that the large firms produce high valuation accuracy when followed by the DDM
method. The results are comparable with Beatty and Ritter (1986) who argue that the
larger IPO firms can be easily estimated through direct valuation models. The results
document that the firms having large investments as fixed show high valuation
accuracy. This implies that the DDM seems to be an appropriate value estimator for
capital-intensive firms. The results document that the higher market returns and large
market volatility before the IPO show less valuation accuracy. These findings are
consistent with Roosenboom (2012) accuracy of DDM valuation analysis.
In the discounted cash flow (DCF) valuation accuracy analysis, Table 4.11
reports that the coefficient of firm size appears to be negative but statistically
insignificant related to the accuracy of DCF method. The results show that the larger
firms produce high valuation accuracy as compared with Beatty and Ritter (1986).
The results document that the firms having large investments as fixed show less
valuation accuracy. This implies that the DCF seems to be an inappropriate value
estimator for capital-intensive firms. The coefficient of operating profitability before
the IPO year is negative and strongly significant at the 95% level of confidence. This
indicates that the higher profitable firms produce less valuation accuracy. The
coefficient of predicted sales growth is found to be positive but statistically
137
insignificant. This implies that the high sales growth IPOs followed by DCF produce
modest valuation accuracy.
Table 4. 11: Cross-sectional Regressions of Accuracy of Valuation Methods
Model APE_DDM APE_DCF APE_Multiples APE_Fair Value
Independent Variables (1) (2) (3) (4)
Intercept 140.1700** 104.2097 63.1419*** -2.9339
(2.699) (1.361) (2.820) (-0.051)
Total Assets -12.5947** -15.548 -4.4939* 3.6154
(-2.481) (-1.742) (-2.019) (0.634)
Firm Age 1.1136 6.6711
(0.123) (0.870)
Property, plant & Equip. -0.0755 0.7089** -0.0866 -0.0305
(-0.344) (2.333) (-0.703) (-0.349)
Operating Profitability -0.2313** -0.107** -0.0766*
(-2.224) (-2.579) (-1.907)
Sales Growth 0.0171 -0.1009** -0.0516*
(0.383) (-2.147) (-1.729)
Dividend Payout -0.1216 -0.113
(-0.861) (-0.929)
Technology 6.894 0.36167
(0.693) (0.063)
Market Returns -1.1312*** 0.1657 -0.1496
(-5.382) (1.101) (-0.976)
Ex-ante 6.9779** -10.1482
(2.604) (-1.677)
Underwriter Reputation 5.6632 2.4605
(0.850) (0.373)
Dilution Factor 0.1221 0.4086
(0.538) (1.153)
Adj. R-Square 0.7422 0.4467 0.2119 0.11183
F-Statistic 7.679 4.624 1.247 0.869
Prob(F-Statistic) 0.009 0.008 0.281 0.5727
N 15 20 65 88
t-statistics using White(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
In the multiples and fair value estimates valuation accuracy analysis, Table
4.11 reports that the coefficient of offer size appears to be negative and significant at
the 90% level of confidence. The coefficient of predicted sales growth in the IPO year
appears to be negative and strongly significant at the 95% level of confidence. The
results show that the firms with a positive trend in sales growth produce more
valuation accuracy when followed by the multiples valuation method. The results are
138
consistent with the findings of Roosenboom (2012) and Demirakos et al., (2010). The
results document that the higher operating profitability predicted sales growth and
large volatility in the market returns before the IPO show high valuation accuracy
when followed by the fair value estimates. The coefficient of predicted operating
profitability appears to be negative and statistically significant at the 95% level of
confidence. This implies that the firms more operating profitable before the IPO are
accurately valued during pricing process. The coefficient of ex-ante risk is found to be
positive and statistically significant at the 95% level of confidence. This implies that
the more aggregate stock market volatility before the IPO produce less valuation
accuracy of firms followed by multiples valuation technique.
Table A.7 (see appendix) presents the results of valuation accuracy of each
valuation method using IPO offer prices as market values and these results are
estimated for robustness. The findings of the accuracy of the DDM valuation method
is same as reported in the Table 4.11 while the results of valuation accuracy of the
DCF method reveals that operating profitability and AIP are additional significant
factors that impact the intensity of valuation accuracy of DCF. The findings show that
the large firms, higher sales growth and bullish market sentiments increase the
valuation accuracy of the multiples valuation approach.
4.2 The IPOs Initial Prices (Valuation) Analysis
In this section, the data is used for various valuation and aftermarket
performance models, and the results are discussed to unfold the research questions of
“whether the prospectus information is useful to predict the IPO offer prices, initial
returns and long-run returns?” As documented in figure 4.2 of post-IPO performance
analysis, the preliminary offer prices are related to prospectus information. As
discussed in the literature review chapter, the prospectus information is classified into
three groups: the fundamental, signaling and the ex-ante risk factors. In this section,
the basic valuation models have been used as set out in equation 9 and equation 10.
The graphical overview of this section is presented in figure 4.2.
139
Figure 4. 2: Aftermarket IPO Valuation and Performance Analysis
Figure 4.2 presents the graphical overview of short and long-run performance
determinants, and irregularities involved in aftermarket performance such as short-run
underpricing and long-run underperformance anomalies. During the pricing process,
the underwriters deliberately offer discount in preliminary offer prices to attract more
participation in the bidding auction. In Pakistan, only high net-worth individuals
(HNWIs) and institutions are allowed to participate in the bookbuilding auction while
general investors can subscribe after the completion of bookbuilding auction process.
The prospectus information plays the vital role in the aftermarket performance
because the prospectus document contains the information of the shareholding
structure of the IPO company, the purpose of IPO proceeds, proposed investment
plans, future prospects, company history, details of trailing financial statements and
valuation methods used to estimate fair value estimates. The prospect information is
divided into three groups namely as fundamental, signaling and ex-ante risk factors.
The key objective of this study is investigate the impact of prospectus information on
the IPOs aftermarket performance and also irregularities that prevail in the primary
market.
140
4.2.1 Descriptive Statistics
The descriptive statistics of fundamental, signaling and ex-ante risk factors
used as a proxy for prospectus information are reported in the Table 4.12. As already
discussed earlier, data has been obtained from the PSX DataStream and prospectuses
(published at the time of formal listing in the exchange) of the IPO firms.
Table 4. 12: Descriptive Statistics of Variables used in Performance Models
Variable Name Mean Min Percentiles
Max Std.
Dev N
25th 50th 75th
Offer Price (PKR) 26.67 10.00 10.00 14.00 30.00 235.00 32.20 86
Book Value (PKR) 20.63 6.82 10.29 13.67 24.02 124.50 19.34 86
Earnings Per Share (PKR) 3.14 -4.43 0.12 1.24 4.53 26.51 5.12 86
D (Dummy for Negative Earnings) 0.21 0.00 - - - 1.00 0.41 86
Dividend Per Share (PKR) 1.02 0.00 0.00 0.00 1.50 10.00 1.81 86
Financial Leverage (%) 49.41 1.40 32.51 49.65 67.80 96.60 24.34 86
Capital Availability Risk (%) 81.60 12.50 66.80 100.00 100.00 100.00 26.86 86
Efficiency Risk (%) 68.08 0.07 52.07 74.52 85.68 248.03 32.64 86
Capacity Risk (%) 61.20 0.00 24.34 70.17 100.00 100.00 40.47 86
Firm Beta (%) 6.71 0.26 1.27 3.61 6.89 50.60 9.55 86
Offer Size (Millions PKR) 992.40 40.00 150.00 330.80 1,150.00 1,2161.25 1,677.23 86
Underwriter Reputation (Dummy) 0.66 0.00 - - - 1.00 0.48 86
Firm Age (Years) 15.23 1.30 3.50 8.00 19.00 78.00 17.79 86
Portion of Shares Offered (%) 23.28 2.50 14.77 25.00 27.73 50.00 10.81 86
Table 4.12 presents the descriptive statistics of IPO firm’s characteristics and
market data used in the valuation and performance models to estimate the impact of
prospectus information on aftermarket IPOs performance. The financial variables data
such as earnings per share, book value per share, retained earnings, dividends payout,
total liabilities, total assets, net sales, cost of goods sold, proposed investment plan,
firm age and the portion of shares offered are taken from the latest financial
statements available in the prospectus documents, while IPOs offer prices, gross IPO
proceeds, market returns and ex-ante uncertainty are taken from the official website of
Pakistan Stock Exchange. In this part, this study excludes two more firms on the basis
of incomplete information about concerned factors.
The descriptive statistics of IPO offer prices and book values that on average
the IPO sample takes a value of 26.67 and 20.63 respectively. The descriptive
statistics of forecasted earnings before the IPO report that on average the IPO sample
141
takes a value of 3.14. This study also used a dummy variable for negative earnings
reported before the IPO. This study observes about 20.0% IPO firms have reported
negative earnings. Rees (1999) argues that the inclusion of negative earnings dummy
in the valuation model is to discriminate the effect of loss building firms on the IPO
pricing. This implies that most of the firms in the sample are young firms. The
average value of proposed dividends at the time of IPO takes a low value. This study
observes that about 41.0% firms (36 out of 86 IPOs) offered cash dividends to its
shareholders. This implies that the sample of this study contains both large and small
firms while many large firms have the track record of paying dividends before IPO
and release the projected dividends in the prospectus and, on the other side, many
small firms did not offer any payout before IPO and not promise for any payout in
near future.
Table 4.12 reports the descriptive statistics of Financial Leverage described as
the ratio of total liabilities to total assets in the recent financial statements disclosed in
the prospectuses. The average (median) financial leverage of IPO sample is equal to
49.41% (49.65%) which is greater than the US, UK and China countries. This implies
that the excessive use of debt financing increases its financial leverage in terms of
high-interest payments, which negatively impact on firms’ core earnings. According
to Loughran & McDonald (2013) and Thomas H.T. (2011) pre-IPO higher financial
leverage is a proxy for ex-ante risk, which deflates equity values. Modiliani & Miller
(1966) argue that higher leverage increases the level of insolvency. Further, we look
at the Capital Availability Risk defined as the ratio of retained earnings over net
income using latest year financial statements data disclosed in the prospectuses. The
availability of internal capital is very important to capitalize the growth opportunities
because most IPO firms are young. The average (median) capital availability risk of
IPO sample is equal to 81.6% (100.0%). This indicates that a large portion of net
income retained in the firm means higher capital availability and lower financial risk.
Table 4.12 also reports the descriptive statistics of four other non-financial risk
factors such as efficiency risk, capacity risk, firms’ beta and IPO gross proceeds. The
efficiency risk is defined as the cost of goods sold over the net sales and used as a
proxy for production efficiency. The average (median) efficiency risk is equal to
68.08% (74.52%). This implies that the high operating efficiency risk means higher
142
cost of goods sold and less production efficiency. As we already discussed earlier that
it is mandatory for issuing firms to disclose the purpose of IPO proceeds and their
proposed investment plans in the prospectus documents. The Capacity Risk is defined
as the ratio of the proposed investment plan over IPO proceeds disclosed in the
prospectus. If the large portions of IPO proceeds are allocated for investment
activities tend to large uncertainty of the returns indicate the higher capacity risk. In
the IPO sample, on average 61.2% of IPO proceeds are proposed for investment
purposes. Our results indicate that the higher proportion of utilization plan over IPO
proceeds leads higher capacity risk. Leone et al. (2003) and Espenlaub et al., (1999)
find the positive relation between the impact of IPO proceeds on under-pricing or
aftermarket performance. Beaver et al,. (1970) first time in the history argues that the
firm beta is an important factor to evaluate the riskiness of the firm’s equity value.
Firm beta is a proxy for price volatility and measured by the standard deviation of
IPO aftermarket prices for 180 trading days from the date of formal listing on the
market. The sample mean of firm beta is 6.7% and the median is 3.61%. This implies
that the higher value of firm beta indicates negative aftermarket performance. The
Offer Size is broadly used as a proxy for the level of risk of the IPO firms. Bessler and
Thies (2007) and, Agarwal, Liu and Rhee (2008) find that offer size is a significant
determinant of aftermarket performance. Sohail & nasr (2007) and Loughran & Ritter
(2002) conjecture that the offer size is negatively linked with the firm valuation. The
offer size is measured as the product of the number of shares offered in an IPO and
the offer price. The average (median) offer size is equal to 992.4 million (330.8
million) and the standard deviation is 12,161.25 million. The findings indicate that the
approximately 70% firms offered capital is less than the average offered capital and
support the evidence of most IPO firms are young.
The Underwriter Reputation is a first signaling variable in the model. Baron
(1982) enlightens the significant role of underwriter reputation in determining the
subscription and allocation of shares. Dominique et al., (2013) finding reveals that the
selection of prestigious underwriters in the IPO process leave less money on the table
because they offered shares at high prices. In this study, underwriter reputation is
measured by how many times the underwriter participated in the IPO process. This
variable takes a value of 1 for prestigious underwriters and 0 for less reputed
underwriters. In our sample, 66% firms (57 out of 86 IPOs) are sponsored by
143
prestigious underwriters. Based on an extant literature and aforementioned reputation
measure, most IPO firms went public in their early business cycle years. It is
conjectured that the Firm Age reflects the level of maturity and size gained in the
product market. Ritter (1999) argues that it is difficult to predict expected cash flows
and payouts (dividends) of younger firms without establishing their trail records and
their value estimated based on their expected future growth rates. The findings report
that the youngest firm in the study is 15 months old and the oldest firm is 78 years
old. The average (median) firm age is 15 years (8 years). The percentile results show
that the majority firms are young as most firms fall under 15 years. The Percentage of
Shares offered during IPO is a signaling factor in the valuation model. Beatty and
Ritter (1986) describe the phenomenon that smaller offering firms are on average
more speculative (higher uncertainty) than their larger offering firms. The descriptive
statistics of the percentage of shares offered show that on average; IPO firms offered
23% of the total outstanding shares to the public.
4.2.2 The Univariate Analysis
The key purpose of the univariate analysis is to explain the association
between the variables used in the valuation and aftermarket performance models.
Table 4.13 demonstrates the coefficients of correlation (Pearson) between variables
used in the valuation models with initial offer prices while t-statistics are reported
below the each correlation coefficient. In this section, the study only focused are the
discussion of correlations between the independent variables (fundamental, ex-ante
risk and signaling factors) and the dependent variable of IPO offer prices.
The univariate analysis of fundamental factors reports expected correlations.
The finding reveals that the correlation coefficient between the predicted earnings and
initial offer prices appears to be positive and strongly significant at the 95% level of
confidence. The coefficient value (0.2204) shows a moderate association between the
predicted earnings and initial offer prices. The correlation coefficient of dividends
payout is positively linked with initial offer prices and strongly significant at the 95%
level of confidence. The coefficient value (0.2409) demonstrates a modest association
between the dividends disclosed in the prospectuses and the initial offer prices. The
predicted earnings and proposed dividends disclosed in the offering documents
strongly contribute in the valuation process.
144
Table 4. 13: Correlation Matrix of Variables used in Valuation and Aftermarket Performance Models
OP/BV EPS/BV D DPS/BV Fin Lev Captl Rsk Eff Rsk Cpcty Rsk Firm Beta Offr Size Und Rep Firm Age
EPS/BV 0.2204
2.0708**
D -0.0725 -0.6196
-0.6661 -7.2364***
DPS/BV 0.2409 0.6323 -0.2951
2.2748** 7.4803*** -2.8302**
Fin Lev -0.0852 -0.1727 0.0941 -0.0456
-0.7837 -1.6067 0.8663 -0.4186
Captl Rsk -0.0748 -0.4702 0.3546 -0.7734 0.1168
-0.6879 -4.8825*** 3.4762** -11.1807*** 1.0779
Eff Rsk -0.0329 -0.3776 0.3400 -0.2404 0.1133 0.2305
-0.3020 -3.7379** 3.3137** -2.2698** 1.0447 2.1711**
Cpcty Rsk -0.2030 -0.1769 0.2881 -0.1931 -0.2188 0.1995 0.0550
-2.9003* -1.6472 2.7577** -2.8036* -2.0552** 2.8661* 0.5052
Firm Beta 0.4803 0.5503 -0.1642 0.5039 0.0710 -0.2234 -0.1392 -0.1715
5.0187*** 6.0401*** -1.5261 5.3471*** 0.6526 -2.1006** -1.2888 -1.5963
Offr Size 0.3891 0.2458 -0.1373 0.2984 0.1287 -0.1663 -0.0925 -0.2759 0.3183
3.8709** 2.3238** -1.2704 2.8649** 1.1891 -1.5453 -0.8517 -2.6308** 3.0773**
Und Rep 0.0427 -0.0101 0.0042 -0.0573 -0.1493 -0.0096 0.0953 0.1567 -0.0473 0.0164
0.3918 -0.0923 0.0387 -0.5259 -1.3836 -0.0875 0.8783 1.4547 -0.4346 0.1509
Firm Age 0.1849 0.2565 -0.3211 0.1985 0.0972 -0.2785 -0.0355 -0.4436 0.2804 0.4713 -0.0406
2.7240* 2.4320** -3.1072** 1.8562 0.8949 -2.6578** -0.3261 -4.5371*** 2.6781** 4.8981*** -0.3724
POS -0.0952 -0.1699 0.2778 -0.2157 -0.1592 0.1443 0.1260 0.3531 -0.0495 -0.3406 0.1886 -0.3141
-0.8762 -1.5798 2.6499** -2.0245** -1.4783 1.3373 1.1641 3.4595** -0.4547 -3.3210** 2.7601* -3.0319**
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
145
The findings reveal that the association between the ex-ante risk factors and
the offer prices has expected correlations. The correlation coefficient between the
financial leverage and the offer prices appears to be negative. This indicates that the
higher financial leverage has a negative impact on the offer prices as consistent with
the existing theoretical and empirical literature. The correlation coefficient of capital
availability risk is negatively linked to offer prices but statistically insignificant. The
association between the efficiency risk and the initial offer prices appears to be
negative but statistically insignificant. The correlation coefficient of capacity risk
appears to be negative relate to offer prices and significant at the 90% level of
confidence. This implies that the firms allocate more proceeds as long-term
investments produce large capacity risk tend to price lower in the pricing decision
process. The correlation coefficient of firm beta found to be positive and significant at
the 90% level of confidence to initial offer prices. The findings of firm beta are
contradicted with the theoretical literature because firms having large volatility priced
are lower in the market. The association between the offer size and the offer prices
appears to be positive and statistically significant at the 90% level of confidence. The
univariate analysis of the offer size with initial offer prices is contradicted by the
findings of Sohail & Nasr (2007) and Loughran & Ritter (2002).
The findings of signaling factors are consistent with the literature and
proposed hypotheses. The correlation coefficient of underwriter reputation is
positively related to initial offer prices. The coefficient of firm age appears to be
positive and statistically significant at the 90% level of confidence to the initial offer
prices. This indicates that the level of maturity in the product market creates
additional value in the initial offer prices decision process. The association between
the percentage of shares offered in the primary market and the initial prices appear to
be negative. This implies that the initial shareholders retain a large part of their initial
shareholdings and offer smaller fractions to the market which produce a positive
signal. These findings are consistent with an extant literature.
In sum, Table 4.13 of correlation matrix shows mixed findings of each
predictor with initial offer prices. The findings of fundamental factors link to initial
offer prices are correlated. From the risk factor analysis, only financial leverage,
capital availability risk, efficiency risk and capacity risk are correlated to the initial
146
offer prices. From the signaling factors, all predictors (underwriter reputation, firm
age and the percentage of shares sold) are consistent factors as expected to be.
However, the discussion of univariate analysis is not intended to test the proposed
hypotheses related to initial offer prices.
4.2.3 Accounting Based Valuation Models Analysis
In this part, the multivariate analysis is employed to address the research
objective of “to investigate the usefulness of prospectus information to price the IPOs
in the ex-ante pricing decision process”. The prospectus information is classified into
three clusters as fundamental, ex-ante risk and the signaling factors.
McCarthy (1999) argues that the lead underwriters used accounting
information such as book value, dividends and earnings to set the preliminary IPO
offer prices. In the literature review chapter, this study documents that the
fundamentals such as earnings, dividends and book values are key driving forces for
equity valuation. Equation (9) presents the basic valuation model for the offer and
first-day closing prices. Table 4.14 presents the findings of basic valuation model for
initial offer prices and first trading day closing prices. The findings are divided into
two panels, Panel A presents the finding by using full IPO sample while Panel B
presents the finding by using non-privatization IPOs sample because the existing
studies document that the state-owned firms are somehow valued differently.
Table 4. 14: Empirical Findings of Basic Valuation Models
Panel-A: Full IPO Sample
Variable Model 1 (OP/BV) Model 2 (FDCP/BV)
Intercept (BV/BV) 1.2416*** 1.5454***
(12.0243) (11.8874)
Earnings (EPS/BV) 0.8279 1.7577**
(1.5493) (2.1102)
D 0.2866** 0.5065**
(2.3324) ( 2.1038)
Dividends (DPS/BV) 2.6327** 5.0129**
(3.1175) (3.1955)
Adj. R2 0.12805 0.24559
F-Statistic 3.96523** 8.79001***
Wald-test 144.58510*** 141.31150***
N 86 86
147
Panel-B: Non-Privatization IPO Sample
Variable Model 1 (OP/BV) Model 2 (FDCP/BV)
Intercept (BV/BV) 1.2519*** 1.5656***
(12.4276) (11.6504)
Earnings (EPS/BV) 0.6294 1.1885
(1.2050) (1.3655)
D 0.2129** 0.3261
(2.7512) (1.8281)
Dividends (DPS/BV) 2.5193** 4.795801**
(2.2605) (2.3273)
Adj. R2 0.11477 0.19455
F-Statistic 3.11188** 5.79727**
Wald-test 154.44560*** 135.73330***
N 78 78
t-statistics using White(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
Table 4.14 presents the results of basic valuation models and t-statistics have
been estimated through White (1980) heteroscedastic standard errors in both panels.
The first column of each panelist the accounting variables, model 1 and model 2
exhibits the regression coefficients using initial offer prices and initial market prices
scaled by book value as dependent variables respectively.
Model 1 demonstrates the findings of basic valuation model using initial offer
prices scaled by book value as the dependent variable. The coefficient of intercept
appears to be positive and strongly significant at the 99% level of confidence to the
initial offer prices. The intercept coefficient indicates the impact of book value on
initial offer prices and confirms the H11a hypothesis that proposed a positive
relationship between the initial offer prices and book values. This indicates that the
book value has a positive impact on the pricing valuation decision and the findings are
consistent with Firth (1998). The regression coefficient of earnings scaled by book
value appears to be positive but statistically insignificant to the initial offer prices.
The finding reveals that the lead underwriters and issuers put a greater weight to
earnings before the IPO when valuing IPOs. As already discussed in the literature
review chapter, Klein (1996) and Beatty et al., (2002) used trailing earnings disclosed
in the prospectuses while Kim & Ritter (1999) and How & Yeo (2001) used
forecasted earnings disclosed in the prospectuses, the empirical findings reveal similar
conclusions that pre-IPO earnings (predicted earnings) have a larger predicting power
to the initial offer (market) prices. According to signaling theory, good quality firms
148
produce signals to disclose its true value in the market. When lead underwriters
decide to reveal this information means it separated the good firms from bad ones.
The findings of this study report the robust relationship between the initial offer prices
and the earnings disclosed in the prospectuses. The coefficient of negative
earnings dummy appears to be positive and strongly significant at the 99% level of
confidence. The negative earnings dummy variable weighting parameter to firms
reporting losses in the prospectus and these findings are similar to trailing earnings
results. The results indicate that, for loss-making firms before the IPOs, the lead
underwriters put more attention to the book value of equity when valuing IPOs. The
findings also support the working hypothesis of H11c that proposed a positive
relationship between the negative earnings dummy and initial offer price. The
coefficient of dividends disclosed in the offering documents appears to be positive
and statistically significant at the 95% level of confidence. The lead underwriters
considered dividend information as a positive signal about future cash flows and a key
driving force of valuation model. The empirical findings reveal that the dividends
disclosed in the prospectuses have positive and significant effect on the preliminary
offer prices.
The value of adjusted-R squared of the fundamentals valuation model looks to
have a less but statistically significant explanatory power to initial offer prices. This
implies that the book value of equity, earnings and dividends have the significant role
in the valuation process. The findings of the Wald test support the evidence of
fundamentals valuation model findings that document the model validity and the
robust joint impact of fundamentals on the IPO offer prices.
Third column of Table 4.14 demonstrates the findings of basic valuation
model related to initial market prices. The findings reveal that all variables of
fundamentals are positive and statistically significant. Their regression coefficients
are slightly higher than the regression coefficients of model 1. The predicting power
of accounting information of model 2 is greater than the predicting power of model 1.
The finding of Wald test is similar to initial offer prices valuation model, which report
the validity of the model that proposed the positive impact of fundamentals on the
initial market prices. In Panel B of Table 4.14, report the empirical findings of basic
valuation model of non-privatization 78 IPOs. The findings unveil that the
coefficients of non-privatization IPOs are slightly lower than the full IPO sample. The
149
statistical results of non-privatization IPOs are similar for both the initial offer prices
and the initial market prices to the earlier full IPO sample results. This implies that the
non-privatization IPOs are priced similarly as in the private IPOs.
This study also estimate the impact of underwriting contract with offerer about
the unsold equity during IPO implementation process. In Pakistan, for fixed price
auction, according to rule 4 of the Companies (Issue of Capital) Rules, 1996, it is
mandatory for underwriters to assign a underwriting contract about the commitment
of fully subscription in cash under the specified period of time but SECP has only
relaxed this clause to privatization IPOs (state-owned enterprises) due to the large
offer size. For book building auction, according to Clause 5 of appendix 2 of the
Listing of companies and securities regulations of KSE, the offer price should be
determined through book building process and the offer size should be underwritten
by book runner of the said IPO. The shares allocated for general public should be
underwritten under Clause 6 of appendix 2 of the Listing Companies and Securities
Regulations of KSE and rule 4(iii) of Companies (Issue of Capital) Rules, 1996. The
findings of this analysis are unable to explore the significance role of underwriting
contract during initial valuation process. This analysis is performed using full sample
data as well as for non-privatized IPOs sample but the findings are consistent in both
cases (see Appendix Table A.8).
One of the key objective of this study is to investigate the impact of
fundamental, risk and signaling factors to IPO valuation. Table 4.14 only analyzes the
explanatory power of the fundamental factors in the valuation model. However, there
are few other factors such as ex-ante risk and signaling factors that explain the
variation in the initial offer and initial market prices.
Table 4.15 and Table A.9 (see appendix) report the findings of the cross-
sectional analysis of valuation models using full IPO sample, both on the initial IPO
offer prices scaled by book value of equity and the initial market prices (1st day
closing prices) scaled by book value of equity respectively. Model 1 includes
fundamental, ex-ante risk and signaling factors while model 2 also includes the
privatization dummy variable to examine the impact of privatizations on initial prices.
150
Table 4. 15: Cross-sectional Analysis of Valuation Models using Full Sample
Variable Model 1 (OP/BV) Model 2 (OP/BV)
Fundamental Factors
Intercept (BV/BV) -2.0982 -2.7758
(-0.9129) (-1.1751)
Earnings (EPS/BV) 0.7325 0.6340
(0.9950) (0.8403)
D 0.4120* 0.3539
(2.9413) (1.5418)
Dividends (DPS/BV) -3.8275** -3.5840**
(-2.9787) (-2.1882)
Risk Factors
Financial Leverage -0.0113* -0.0111*
(-2.9028) (-2.9069)
Capital Availability Risk -0.0027 -0.0036
(-0.7809) (-1.0305)
Efficiency Risk 0.0009 0.0001
(0.1738) (-0.0040)
Capacity Risk -0.0089** -0.0096**
(-2.1816) ( -2.2898)
Firm Beta 0.0632*** 0.0673**
(3.5315) (4.2925)
Offer Size 0.4137* 0.5205**
(2.6727) (2.9574)
Signal Factors Underwriter Reputation 0.1719 -0.0062
(0.5431) (-0.0177)
Firm Age -0.0195** -0.0183**
(-2.1366) (-2.0376)
Percentage of Shares Offered -0.0225 -0.0228
(-1.3430) (-1.3588)
Privatization - -0.9284*
(-2.7478)
Adj. R-square 0.34345 0.368765
F-Statistic 3.13868*** 3.190606***
N 86 86
t-statistics using White(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
The findings of fundamental factors discussed in Table 4.14 are similar as
investigated in Table 4.15. However, the explanatory power of full sample models
(discussed in Table 4.15 and Table A.9) improves significantly when more variables
are added in the models. The regression coefficients have mixed sign while their
magnitudes differ by the addition of more independent variables. The coefficients of
151
the book value of equity and dividends are contrary to the basic valuation model
analysis.
As discussed in the methodology framework chapter, the proposed hypotheses
related to ex-ante risk factors are developed on the basis of risk-aversion assumption.
It is conjectured that the initial offer prices are diminishing function of risk factors.
Therefore, IPO firms are priced lower at the initial offer prices because investors
demand compensations for riskier investments.
In the Table 4.15, Model 1 shows the impact of ex-ante risk factors on the
initial offer prices scaled by book value of equity. The coefficient of financial
leverage appears to be negative and statistically significant at the 90% level of
confidence. This finding is consistent with Fama and French (1998), who investigate
an inverse relationship between the financial leverage and the firm valuation. This
implies that the excessive use of debt financing increases their debt paying ability due
to high-interest payments, which negatively impact on IPOs prices. According to
Loughran & McDonald (2013) and Thomas (2011) pre-IPO higher financial leverage
as a proxy for ex-ante risk, increases the deflation of equity valuation. Modiliani &
Miller (1966) argue that higher leverage increases the level of insolvency. The results
are consistent with the proposed H12a hypothesis that the financial leverage is
negatively related to IPO prices. The findings of financial leverage are similar by the
inclusion of privatization dummy in model 2. Table A.9 (see appendix) presents the
findings of valuation model by taking first day closing prices scaled by book value of
equity as a dependent variable. The coefficient of financial leverage retrieved from
Table A.9 is not in line with the proposed hypothesis of H12a. The capital availability
is the next risk factor. Fama and French (1998) argue that the IPO firms follow the
pecking order theory as IPOs prefer to finance their investments first with the internal
resources such as retained earnings, then with the external resources such as debt and
issuing equity. Therefore, the greater the internal capital availability considered as
beneficial and lower business risk. The coefficient of capital availability risk appears
to be negative but statistically insignificant. This implies that lower capital availability
indicates higher business risk and this finding is contras to the proposed H12b
hypothesis that a large portion of net income retained in the firm means higher capital
availability and lower financial risk. The univariate analyses presented in Table 4.13
also support the findings of capital availability risk. As a robustness measure, in Table
152
A.9, the coefficient of capital availability risk is found to be positive but statistically
insignificant to the initial market prices.
In this study, the efficiency risk and capacity risk factors are viewed as non-
financial risk factors. The efficiency factor estimates the operational efficiency of
IPOs which hypothesized that more efficient firms generate more profit for the
shareholders and are priced higher during the IPO pricing process. The efficiency risk
factor is estimated as the ratio of cost of goods/services sold (CGS) over total revenue
from the latest financial statements disclosed in the prospectuses. The coefficient of
efficiency risk is found to be positive but statistically insignificant. This implies that
the less operating efficient firms are priced higher during IPO transactions. This
finding contradicts with the proposed hypothesis of H12c that the less operating
efficient firms have the negative impact on the initial prices. The correlation matrix
(see Table 4.13) also reports the negative and insignificant association between the
efficiency risk and the initial offer prices. As a result, no multicollinearity issue
detected about efficiency factor. The results of efficiency risk (reported in Table A.9)
on initial market prices are also contradicted with the proposed hypothesized
relationship between the operating efficiency and the initial valuations.
The capacity risk illustrates the risk associated with issuing firm decision of
the proposed investment plan of IPO proceeds. The coefficient of capacity risk
appears to be negative and strongly significant at the 95% level of confidence. This
indicates that the issuing firms proposale of a larger part of gross IPO proceeds to be
utilized in investment activities produce more risk. As a result, firms having a larger
fraction of IPO proceeds as investment priced lower by market participants.
Therefore, the capacity risk factor is negatively linked with prices, which in turn,
supports the proposed hypothesis of H12d. Keasey and Short (1997) investigate a
positive relationship between the IPO market prices and the IPO proceeds. They
unveil that the utilization of IPO proceeds disclosed in the prospectus are viewed as a
signal to the IPO valuation. The correlation matrix (see Table 4.13) also produces the
negative association between the capacity risk and the initial offer prices.
153
Table 4. 16: Cross-sectional Analysis of Valuation Models using non-PIPO Sample
Variable Model 1 (OP/BV) Model 2 (FDCP/BV)
Fundametal Factors
Intercept (BV/BV) -3.0814 -2.7660
(-1.2791) (-1.0908)
Earnings (EPS/BV) 0.9368 1.9578
(1.2412) (1.2894)
D 0.5534** 0.8663**
(2.2383) (2.3533)
Dividends (DPS/BV) -4.0636 -10.9249**
(-1.3649) (-2.0941)
Risk Factors Financial Leverage -0.0103* -0.0015
(-2.1671) (-0.2469)
Capital Availability Risk -0.0055 -0.0185*
(-0.6577) (-2.3864)
Efficiency Risk 0.0032 0.0040
(0.7670) (1.5463)
Capacity Risk -0.0092** -0.0016
(-2.0839) (-0.2735)
Firm Beta 0.0836*** 0.1556**
(5.1690) (4.2234)
Offer Size 0.5637* 0.5740*
(2.9512) (2.9395)
Signal Factors
Underwriter Reputation 0.0126 -0.0880
(0.0335) (-0.2189)
Firm Age -0.0177 -0.0130
(-1.5299) (-1.1275)
Percentage of Shares Offered -0.0224 -0.0296
(-1.3697) (-1.6820)
Adj. R-square 0.38769 0.48605
F-Statistic 3.37696*** 4.96503***
N 78 78
t-statistics using White(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
The firm beta is used as a proxy of price volatility before the IPO and
measured as the standard deviation of IPO aftermarket prices for 180 trading days
from the date of formal listing on the market. The firms having larger volatility are
riskier and consequently, valued lower by the lead underwriters. The coefficient of
firm beta is found to be positive and statistically significant at the 90% level of
confidence. This implies that the finding of this risk factor is contradicted with the
154
working hypothesis of H12e. The coefficient of firm beta with the initial market
prices is found to be positive but statistically insignificant. This finding also
contradicts with the proposed hypothesis. The last risk factor is offer size which is
measured through the product of the number of shares offered and the IPO offer
prices. The offer size is generally used as a proxy for the level of risk of the IPO firms
(Aggarwal, Liu and Rhee, 2008: Bessler and Thies, 2007). If firms offere a larger
portion of their paidup capital in the IPOs is considered to be more riskier and priced
lower by the lead underwriters. The coefficient of offer size is found to be positive
and statistically significant at the 90% level of confidence. This implies that the firms
offered more capital seems to be priced higher. This finding contradicts with the prior
literature and doesn’t support the proposed hypothesis of H12f. Table 4.15 shows
mixed results to the relationship between the ex-ante risk factors and the initial offer
prices. Therefore, it has been concluded that the pre-IPO risk factors have small but
weak significant influence on the initial IPO valuations.
Regardless of numerous methods used to estimate the lead underwriter
reputation, existing literature summarized that the prestigious underwriters reduce the
uncertainty of the IPO firms that are priced higher at the IPO pricing decision and
lower the initial excess returns in the early days. In this study, the prestigious
underwriter reputation is measured as when it undertook more than 6 IPOs in the
sample. The underwriter reputation as a dummy variable is employed in the valuation
model which gets 1 for the prestigious underwriters and 0 for the less reputed
underwriters. Therefore, the investors are willing to pay higher prices that IPOs are
sponsored by the prestigious underwriters. Yung (2011) argues that prestigious
underwriters should have an advantage in information production because of a large
network with high net-worth institutions and/or individuals resulted in greater price
revision in the auction. The coefficient of underwriter reputation is found to be
positive but statistically insignificant to the initial offer prices. This implies that the
IPOs underwritten by prestigious underwriters are priced higher on the initial offer
prices. In this analysis, the firm age is used as another signaling variable. The existing
literature conjectures that the mature firms (older firms) have less uncertainty because
they contain the consistent market share in the product market. The coefficient of firm
age appears to be negative and statistically significant at the 95% level of confidence.
This implies that mature firms are priced lower at the IPO listing. This finding is
155
inconsistent with the extant literature and proposed hypothesis of H13b. However, the
correlation matrix (see Table 4.13) reports a positive association of firm age to the
initial offer prices. The key signaling factor of this valuation model is the percentage
of shares offered in the IPO. The percentage of shares offered in the IPO signals the
dissatisfaction of the initial shareholders on the firm’s future prospects. Based on the
theoretical and empirical literature, it is conjectured that the percentage of shares
offered is positively linked to the initial valuation as proposed in the H13c hypothesis.
The coefficient of the percentage of shares offered is found to be negative and
statistically insignificant. This implies that the lead underwriters priced higher the
IPOs that offered smaller fractions of outstanding shares. The findings of this analysis
is completely consistent with the proposed hypothesis of H13c and also can be
verified from the univariate analysis. But the coefficient of the percentage of shares
offered with the initial market prices (reported in Table A.9) is found to be positive
but statistically insignificant. This finding is contrary to the working hypothesis
related to initial market prices.
In Table 4.15 and Table A.9 (see appendix), the privatization dummy variable
is used to control the impact of privatization on the valuation model because
Dewenter et al., (1998) argue that the state-owned IPO firms are priced differently
than other private IPO firms. However, the overall results without privatization
dummy model show similar results as presented in the model 2. The privatization
dummy variable model (model 2) overall slightly improves the prediction power from
34.35% to 36.88%. The coefficient of privatization dummy is found to be negative
and significant at the 90% level of confidence.. A sensitivity analysis has been
undertaken in the Table 4.16 using non-privatization IPOs sample. The findings of the
first model of Table 4.16 are similar as presented in the first model of Table 4.15. In
the second model of Table 4.16, all results are in line with earlier findings presented
in the Table A.9 while negative earnings dummy and the capital availability risk
factors are other significant determinants.
4.3 The IPOs Initial Excess Returns Analysis
The empirical findings of IPO initial excess returns (IER) also known as IPO
underpricing have been discussed in literature. As discussed in the research
methodology chapter, the initial excess returns can be estimated as the rate of return
156
earned by market participants on 1st trading day closing or a week or a month closing
prices. This IER analysis attempts to deal with the research objective “To investigate
the usefulness of prospectus information on the IPOs short-run excess returns”. The
IER analysis also investigates the underpricing phenomenon which results in
abnormal returns in early trading days as discussed in the literature. An empirical
model has been employed to investigate the effect of prospectus information,
privatization dummy and initial valuation residuals on the IER.
Similar to the initial valuation analysis (section 4.2), the data of 86 IPOs from
prospectus information (fundamental, risk and signaling factors) have been extracted
from the offering documents. Furthermore, the standardized residual series extracted
from the valuation analysis is added to the IER analysis as a proxy for the
unobservable determinants of the IPO prices.
4.3.1 Descriptive Statistics of IER
The summary statistics of initial excess returns, fundamentals and residual
series extracted from the valuation model are reported in the Table 4.17. As already
discussed in the earlier the share prices and financial variables data of IPO firms has
been obtained from the PSX DataStream and the prospectuses respectively.
Table 4. 17: Descriptive Statistics of Initial Excess Returns Analysis
Variable Name Mean Min Percentiles
Max SD N 25th 50th 75th
Initial Excess Returns (IER %) 32.85*** -35.76 0.00 13.84 37.10 322.0 59.88 86
Book Value (BV/OP) 1.02 0.18 0.67 0.95 1.14 6.42 0.73 86
Earnings Per Share (EPS/OP) 0.09 -0.44 0.01 0.08 0.16 0.55 0.15 86
Residuals (Resi_Val) 0.00 -3.71 -0.43 0.03 0.58 3.85 1.08 86
***Significant at 1% level
Table 4.17 presents the average (median) underpricing of IPO sample is equal
to 32.85% (13.84%) which is greater than the US, UK and other developed countries
(Loughran and Ritter, 2003; Ljungqvist, 2009; Bodnaruk, et al., 2008; Lee, et al.,
2012). However, these IER are lesser relative to China, Jordan, India and Sri Lank
(Yu and Tse, 2006; Marmar, 2010; Shelly and Singh, 2008; Peter, 2007). The finding
implies that the excessive returns highlighting the underpricing anomaly exist in PSX
market. This indicates that the initial participants who buy shares in primary market
157
and sell them on first trading day earn abnormal returns by 32.85% from their
investments. It has been observed from the data that the most IPOs are underpriced
(57 out of 86 IPOs), 17 are overpriced where 12 are priced accurately.
To examine the association between the size of IPO firms and IER on the first
trading day, The scholar categorized the sample into four groups based on market
capitalization of IPOs at the offer prices. PKR 600 million, PKR 2,100 million and
PKR 6,500 millions are taken as cut-offs closet to 1st, 2nd and 3rd quartiles
respectively. The cutoff market capitalization for small group firms is less than PKR
2,100 million.
Table 4. 18: Descriptive Statistics of IERs in different Issue Proceeds
IPO Proceeds N %age of IPOs Mean Median Min Max SD
≤ 600 mn 23 26.74 39.46** 8.45 -20.00 322.00 72.80
> 600 and < 2,100 mn 20 23.256 34.09** 18.70 -22.40 228.00 228.00
> 2,101 and < 6,500 mn 22 25.581 37.39** 16.14 -35.76 270.74 66.68
≥ 6,500 mn 21 24.419 19.64** 13.70 -15.03 99.91 29.60
Small IPO Firms 43 50.00 36.96*** 10.00 -22.40 322.00 67.10
Big IPO Firms 43 50.00 28.73*** 13.98 -35.76 270.74 52.16
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4.18 presents the descriptive statistics of IER related to different size
based IPO groups. The average initial excess returns for different size groups are
significantly different from zero. The IER decrease as the size of firm is increases.
The IPOs that have highest market capitalization (greater than 6,500) produce of
19.64% IER, while firms with lowest market capitalization (less than 600 million)
produce an IER 39.46%. These findings are in sequence with the assumption of ex-
ante uncertainty hypothesis. Table 4.18 also reports the initial excess returns of small
IPO firms (market capitalization below 2,100 million used as the breakpoint) and big
IPO firms separately. The empirical findings highlight that the average IER (36.96%)
of small firms is greater than the average IER (28.73%) of big firms. These findings
point out that the IER are indirect by proportional to the size of issuing firms.
To investigate the impact of the IPOs issued in Hot and Cold time periods and
if they are significantly linked with the IER on the 1st trading day, The periodic cycles
of high volume (number of IPOs) with high IERs are referred as “Hot Issue” market
158
while relatively low volume with lower initial excess returns are referred as “Cold
Issue” market. According to Signaling Theory (Allen and Faulhaber, 1989; Jegadeesh
et al., 1993; Nanda and Yun, 1997), only high-quality firms went public during hot
issue market and deliberately offer the discount in terms of large underpricing to pass
a signal of high-quality firms. Colak and Gunay (2011) inspected that the good quality
firms intentionally wait and decide to go public when they make sure that the market
is hot. According to Risk Composition Hypothesis, Ritter (1984) argued that the more
risky firms went public when they observe hot issue market phenomenon and initial
returns are leading. Peterle and Berk (2016) inspected that generally riskier firms
prefer to issue IPOs during a “hot issue” market when initial excess returns are higher.
This study follows the statistical technique for identification of cyclical
patterns of “hot- and cold-issue” markets as followed by the Ritter (1984), Ghosh
(2004) and Agathee et al., (2012a) as a positive correlation between the number of
IPOs offered and IERs on early days. Figure 4.3 displays the IPOs listed and IER per
year in PSX from 2000 to 2016. Based on the IPO activity and initial returns
displayed in figure 4.3, periods from 2000-2003 and 2008-2013 could be viewed as
cold issue market while periods from 2004-2007 and 2014-2016 could be viewed as
hot issue markets
Figure 4. 3: Yearly Listed IPOs and IER in Hot-Cold Issue Market
32
43
8
14
2
109
3
6
43
1
5
7
4
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
0
2
4
6
8
10
12
14
16
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
IPOs IER %
159
Table 4. 19: Descriptive Statistics of IERs in Hot- & Cold-issue Periods
No. of years N %age of IPOs Mean Median Min Max SD
Hot Issue Periods
2014-2016 3 16 18.60 10.89 13.84 -20.00 35.43 17.96
2004-2007 4 33 38.37 59.46 36.20 -22.40 322.0 74.91
Cold Issue Periods
2008-2013 6 26 30.23 16.54 0.12 -35.76 228.0 52.47
2000-2003 4 11 12.79 23.50 8.45 -18.00 83.00 35.55
Hot IPOs 7 49 56.98 43.59*** 23.98 -22.40 322.0 66.12
Cold IPOs 10 37 43.02 18.61** 2.00 -35.76 228.0 47.68
Wilcoxon-Mann-Whitney Rank Test 2.5513**
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4.19 presents the findings of IER during “hot- and cold-issue” markets
in the PSX. The findings reveal that the weighted average initial returns (43.59%) in
the “hot-issue” markets are more than the weighted average initial returns (18.61%) in
“cold-issue” markets. The average returns of hot and cold issue markets are
significantly different from zero at 99% and 95% level of confidence respectively.
The Wilcoxon-Mann-Whitney rank test (nonparametric test) is used to examine for
equality medians. The difference between medians is significant at the 95% level of
confidence. The findings conclude hat the average IERs in the “hot-issue” market are
significantly higher than in “cold-issue” market. The findings of the hot-cold issues
analysis are inline with the signaling hypothesis and the changing risk composition
hypothesis (Ritter, 1984). The findings also report that the most IPOs (57% offered in
seven years) and a smaller number of IPOs (43% offered in 10 years) decide to go
public in “hot-issue” and “cold-issue” markets respectively. The empirical findings
are consistent with the existing literature (Agathee et al., 2012a; Moorman, 2010;
Helwege and Liang, 2004).
From the existing literature on IPOs offering mechanism, Sherman (2005)
argues that the institutional investors discourage all other methods than the
bookbuilding because of free rider and winner’s curse issues. The institutional
investors, who are more informed, prefer bookbuilding mechanism because they get
more shares at a better price due to their active participation in the price discovery
process. However, Lee (1999), aggarwal (2002), Ljungqvist (2003), and Jenkinson
160
and Jones (2009) argued that the preference of bookbuilding is improved auction to
reduce the level of underpricing in many countries. Many researchers stimulate the
causes of privatization as budget constraints especially during the economic recession,
the inefficiency of state-owned ventures, high political intervention and deprived
financial disciplines. Farinos et a., (2007), Aussenegg and Jelic (2007), Florio and
Memzoni (2004) and Paudyal et al., (1998) investigate the impact of privatization
IPOs on the initial returns and found that the privatization IPOs (PIPOs) outperform
in the short run because they offer large discount due to large offer size.
Table 4. 20: Descriptive Statistics of IERs of various sub-samples
IPO sub-Sample N %age of IPOs Mean Median Min Max SD
Panel A
Fixed Price 64 74.42 41.23*** 21.20 -35.76 322.00 66.85
Book-Building 22 25.58 8.46** 2.57 -20.00 40.79 16.48
Wilcoxon-Mann-Whitney Rank Test **2.1082
Panel B
Privatization 8 9.30 58.76** 65.08 -20.00 131.00 46.31
Non-Privatization 78 90.70 30.19*** 9.23 -35.76 322.00 60.72
Wilcoxon-Mann-Whitney Rank Test 2.2036**
Panel C
Survivors 78 90.70 34.14*** 13.84 -35.76 322.00 61.56
Non-Survivors 8 9.30 20.20 11.67 -31.30 98.50 40.72
Wilcoxon-Mann-Whitney Rank Test 0.6913 ***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4.20 presents the descriptive statistics of initial excess returns of firms
that went public through Bookbuilding and Fixed Price mechanism, Privatization and
non-Privatization IPOs, and the firms survive after the listing as compared to non-
surviving firms. Panel A reports the initial excess returns of IPOs issued through fixed
prices and bookbuilding during the sample period in Pakistan. Most of the IPOs (64
out of 86) used fixed price offering mechanism and had underpricing of 41.23%,
while the IPOs offered by the bookbuilding mechanism had underpricing of 8.46%.
The results of this analysis indicate that due to the large participation of institutional
investors are allotted more shares at a better price during the IPO bookbuilding
bidding process. The findings are in sequence with extant literature that bookbuilding
is used to reduce the degree of underpricing. The Wilcoxon-Mann-Whitney rank test
161
(nonparametric test) is used to examine for equality medians. The difference between
medians is significant at the 95% level of confidence. This implies that initial returns
which followed by the fixed price mechanism, are significantly greater than the
followed by bookbuilding mechanism.
Panel B of Table 4.20 presents the summary statistics of IERs of privatization
and non-privatization IPOs. During the 2000-2016 sample period, eight firms went
public due to the privatization process in the Pakistan. The underpricing of
privatization IPOs is higher than the underpricing of non-privatization IPOs. These
returns are significantly different from the zero. The findings of the Wilcoxon-Mann-
Whitney rank test unveil that the privatization IPOs are priced differently than non-
privatization IPOs during ex-ante valuation. The IPO firms that delist from the PSX
during five years from the date of formal listing are known as non-survivor firms.
Panel C of Table 4.20 reports the descriptive statistics of initial returns of survivor
and non-survivor IPOs during the sample period. The underpricing of survivor IPOs is
higher than the underpricing of non-survivor IPOs. However, thier median returns are
not statistically significantly different from each other when Wilcoxon-Mann-
Whitney rank test is employed.
Table 4. 21: Year-wise Initial Excess Returns Analysis
Year No. of IPOs %age of IPOs Initial Returns
2016 4 4.65 0.29
2015 7 8.14 7.00
2014 5 5.81 24.80
2013 1 1.16 0.14
2012 3 3.49 1.50
2011 4 4.65 1.26
2010 6 6.98 1.20
2009 3 3.49 11.10
2008 9 10.47 42.20
2007 10 11.63 71.03
2006 2 2.33 114.75
2005 13 15.12 44.10
2004 8 9.30 56.13
2003 3 3.49 56.48
2002 3 3.49 26.67
2001 2 2.33 1.88
2000 3 3.49 1.78
Total 86 Average Underpricing 32.85
Table 4.21 presents the year-wise initial excess returns also known as
underpricing of IPO firms listed in PSX during 2000-2016. The average underpricing
of IPOs offered during the sample period is 32.85%. During 2004 to 2008 and 2013 to
162
2016, market performance was terrific and touched its peak levels due to high GDP
growth rate, low inflation rate, stable exchange rate and healthy FDI (Sohail and
Raheman, 2010). The large underpricing is observed often in the hot issue phases
while less underpricing is observed in the cold issues. A smaller underpricing during
2000-2002 and 2009-2013 were observed due to the impact of the US internet bubble
crisis and the US subprime mortgage crisis respectively.
Table 4. 22: Sector-wise Initial Excess Returns Analysis
Sector Name No. of IPOs %age of IPOs Initial Returns
Automobile & Electrical Goods 4 4.65 27.37
Cement 5 5.81 16.46
Chemicals 6 6.98 65.88
Commercial Banks 9 10.47 52.59
Engineering 7 8.14 40.94
Fertilizer 2 2.33 12.51
Foods & Allied 2 2.33 -7.90
Insurance & Leasing 1 1.16 0.10
Investment Sec/Banks 12 13.95 38.87
Modaraba 3 3.49 -3.33
Oil & Gas 5 5.81 88.03
Power Gen. & Distribution 7 8.14 20.94
Property & Investment 3 3.49 35.48
Technology & Communication 12 13.95 20.07
Textile 6 6.98 -3.87
Transporation & Communication 2 2.33 50.67
Total 86 Average Underpricing 32.85
Figure 4. 4: Sector-wise Initial Excess Returns Analysis
0
27.3716.46
65.88
52.5940.94
12.51
-7.900.10
38.87
-3.33
88.03
20.94
35.48
20.07
-3.87
50.67
-20
0
20
40
60
80
100
163
Figure 4.4 presents the sector-wise under and overpricing of IPO firms listed in PSX
during 2000-2016. The higher underpricing observed in the Oil & Gas, Chemicals,
Commercial Banks and Transportation & Communication sectors are 88.03%,
65.88%, 52.59% and 50.67% respectively. On the other side, the overpricing is
observed in the Foods & Allied, Textile and Modaraba sectors. However, the firm
listed in Insurance & Leasing sector is accurately placed in the market.
4.3.2 The Univariate Analysis
Table 4.23 demonstrates the correlation (Pearson) coefficients of variables
used in the initial excess returns analysis. In this part, one-to-one relationship between
the initial excess returns and each predictor is discussed while t-statistic is reported
below each correlation coefficient.
The univariate analysis of fundamental factors reports expected correlations.
The findings reveal that the correlation coefficient between the book value of equity
scaled by offer prices and the initial excess returns appears to be positive (0.0114).
The second fundamental factor is the earnings-to-price ratio and indicates the
information of shares risk. The correlation matrix reveals that the coefficient between
the earnings-price ratio and IER is found to be positive (0.1085). This indicates that
the lead underwriters used earnings disclosed in the prospectus, producing a signal of
good quality firms, valued higher by the market participants on the 1st trading date.
The correlation matrix of ex-ante risk factors related to initial excess returns
reports expected associations are developed on the basic assumption of a risk-return
hypothesis. The findings reveal that the investment in IPOs is considered as riskier
and investors demand compensations in terms of higher underpricing in early days.
On the other hand, only financial leverage and firm beta are significantly linked with
IERs. The univariate analysis of signaling factors related to initial excess returns
reports expected associations. The finding reveals that the coefficient between the
underwriter reputation and initial offer prices appears to be negative (-0.2022). The
coefficient of firm age is found to be negative (-0.0607). The findings indicate that the
rapidly growing and young firms offered discount in offer prices and resulted in high
underpricing on the first trading day. The correlation coefficient of the percentage of
shares offered in IPOs related to initial excess returns is found to be positive (0.1104).
164
Table 4. 23: Correlation Matrix of Variables used in Initial Excess Returns Analysis
IER BV/OP EPS/OP Fin Lev Captl Rsk Eff Rsk Cpcty Rsk Firm Beta Offer Size Und Rep Firm Age POS Priv
BV/OP 0.0114
0.1049
EPS/OP 0.1085 0.4101
1.0000 4.1218**
Fin Lev 0.1989 0.1563 -0.0697
1.8602* 1.4503 -0.6410
Captl Rsk 0.0556 0.0149 -0.3581 0.1168
0.5106 0.1368 -3.5148** 1.0779
Eff Rsk 0.1731 -0.1607 -0.4302 0.1132 0.2305
1.6109 -1.4929 -4.3686** 1.0447 2.1711**
Cpcty Rsk -0.0088 -0.1519 -0.2128 -0.2188 0.1995 0.0550
-0.0809 -1.4090 -1.9963** -2.0552** 1.8661* 0.5051
Firm Beta 0.3529 -0.2608 0.2284 0.0710 -0.2234 -0.1392 -0.1715
3.4573*** -2.4763** 2.1507** 0.6526 -2.1006** -1.2888 -1.5963
Offer Size -0.0601 -0.2577 0.0587 0.1286 -0.1663 -0.0925 -0.2759 0.3183
-0.5517 -2.4449** 0.5397 1.1891 -1.5453 -0.8517 -2.6308** 3.0773**
Und Rep -0.2022 -0.1629 -0.0885 -0.1493 -0.0095 0.0953 0.1567 -0.0473 0.0164
-1.8919* -1.5137 -0.8150 -1.3836 -0.0875 0.8783 1.4547 -0.4346 0.1509
Firm Age -0.0607 -0.0897 0.1664 0.0972 -0.2785 -0.0355 -0.4436 0.2804 0.4713 -0.0406
-0.5581 -0.8257 1.5474 0.8949 -2.6578** -0.3261 -4.5371*** 2.6782** 4.8981** -0.3724
POS 0.1104 -0.0740 -0.1211 -0.1592 0.1443 0.1260 0.3531 -0.0495 -0.3406 0.1886 -0.3141
1.0187 -0.6806 -1.1182 -1.4783 1.3373 1.1641 3.4595** -0.4547 -3.3211** 1.7601** -3.0319**
Priv 0.1394 0.1815 0.2318 0.1272 -0.3127 -0.2263 -0.3233 0.3058 0.3645 -0.3643 0.3222 -0.2540
1.2902 1.6918** 2.1841** 1.1755 -3.0175** -2.1301** -3.1314** 2.9443** 3.5886** -3.5853** 3.1194** -2.4072**
Resi -0.1168 -0.2939 -0.1975 -0.0003 0.0017 -0.0064 0.0013 -0.0001 -0.0001 0.0001 -0.0001 0.0000 -0.1543
-1.0785 -2.8189** -1.8468** -0.0024 0.0162 -0.0058 0.0011 -0.0008 -0.0006 0.0009 -0.0009 0.0002 -1.4316
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
165
An extant literature reports the evidence of large underpricing of privatization
(state-owned firms) IPOs. The correlation coefficient between the privatization and
the initial excess returns appears to be positive but insignificant. As discussed earlier,
the standardized residuals are extracted from valuation model is added to capture the
effect of unobservable factors influencing the initial evaluation process. The
correlation coefficient of standardized residuals related to initial excess returns is
found to be negative (-0.1168). The findings imply that IPOs are accurately placed in
the PSX and the market does not misprice them.
In sum, the findings of univariate analysis conclude that IER are shaped by the
investors’ sentiments. The univariate analysis implies that only a few variables such
as financial leverage, underwriter reputation and firm beta are significantly link firm
betad with IERs. This analysis also validates the evidence of privatization IPOs
underpricing.
4.3.3 The Analysis of Initial Excess Returns Models
this section investigates the explanatory power of prospectus information
(fundamental, risk and signaling factors) on the initial excess returns estimated as the
percentage difference between the IPOs offer prices and the first day market values.
Table 4.24 exhibits the OLS findings of two IER models. In model 1, only the
fundamental, ex-ante risk and signaling factors are used as predictor variables.
However, in the second IER model, the standardized residuals factor along with all
previous variables is used as predictor variables. In this analysis, the initial excess
returns (IERs) are used as dependent variables and t-statistics have been estimated
through the Huber/White standard errors. The findings of IER models assist to
investigate the cross-sectional determinants of initial returns. Finally, a sensitivity
analysis of the findings of IER models is also displayed in Table 4.25.
166
Table 4. 24: Regression Analysis of IER Models using Full IPO Sample
Variable Model 1 (IER) Model 2 (IER)
Fundamental Factors
Intercept 0.7489 2.0278
(0.7615) (1.1900)
Book Value (BV/OP) 0.0229 0.0216
(0.3880) (0.3114)
Earnings (EPS/OP) 0.8076* 0.8548
(2.0249) (1.1966)
Risk Factors Financial Leverage 0.0045* 0.0043
(2.1006) (1.1998)
Capital Availability Risk -0.0011 -0.0015
(-1.3442) (-1.6165)
Efficiency Risk 0.0061*** 0.0061**
(4.0253) (2.6704)
Capacity Risk 0.0012 0.0030
(0.6281) (1.2807)
Firm Beta 0.0271** 0.0233*
(2.0812) (2.0975)
Offer Size -0.1535 -0.3478*
(-1.2979) (-2.1763)
Signal Factors
Underwriter Reputation -0.0893* -0.1329*
(-2.06855) (-2.1827)
Firm Age -0.0051 -0.0047
(-1.3355) (-0.7723)
Percentage of Shares Offered -0.0014 -0.0003
(-0.7852) (-0.1509)
Valuation Residuals - -0.047*
- (-2.7601)
Adj. R-square 0.31609 0.25928
F-Statistic 3.025212** 3.071128**
N 86 86
t-statistics using White(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
The intercept coefficient in both IER models is found to be positive that
validate the underpricing phenomenon exist in the PSX as discussed in the descriptive
statistics of IER analysis. The IER models show the diligence of underpricing even
after amending and controlling additional variables in the analysis.
The research methodology chapter document that the higher book value to
offer price (BV/OP), the higher the underpricing relative to the fundamental factors.
167
In addition, Fama and French (1995) discuss that the higher the BTM ratio exhibits
the riskiness of equity firms. Beatty & Ritter (1986) argue that firms having more ex-
ante uncertainty leads to greater underpricing during early trading days. On the basis
of the aforementioned arguments, the proposed hypothesis hypothesizes a positive
relationship between the BV/OP ratio and IPO initial excess returns. Table 4.24
reports that the regression coefficient of the book value of shareholder’s equity scaled
by offer prices related to initial excess returns is positive but insignificant. The
findings of this analysis are consistent with the existing literature, univariate analysis
and support the evidence of the proposed hypothesis of H14a that documented the
direct relationship of BV/OP related to initial returns. The finding of book value
fundamental is also consistent with Beneda and Zhang (2009) but inconsistent with
Mumtaz, Smith & Ahmed (2016), Banerjee, Dai & Shrestha (2011). The findings are
same as reported in the model 2. Although, the association between the book-values
over offer prices and standardized residuals was found to be negative and statistically
significant. The earnings disclosed in offering documents is positively linked with the
aftermarket performance is generally viewed as a signal as discussed earlier. The
regression coefficient of an earnings-price ratio is appeared to be positive and
significant in the first IER model, however, the regression coefficient appears to be
positive but insignificant in the second IER model. How and Yeo (2000) and Firth
(1998) found the positive association between the accuracy of earnings disclosed in
the prospectus documents and degree of underpricing observed in the market. Sohail
(2015) investigates the negative association of earnings related to underpricing of 83
IPOs listed in Pakistan.
As reported in the correlation matrix, the relationship of all ex-ante risk factors
except capital availability risk and capacity risk factors are related to the initial
returns. The regression coefficient of financial leverage is found to be positive and
statistically significant at the 90% level of confidence related to the initial returns.
This implies that the highly leveraged firms considered as risky and produce higher
returns in the market. This finding provides the evidence to validate the proposed
hypothesis of H15a. This finding is in line with Mumtaz, Smith & Ahmed (2016) but
is contradict with Hedge and Miller (1996), who investigate the impact of financial
leverage on IPO valuation and find an inverse relation between the degree of financial
leverage and underpricing. The second financial risk factor is capital availability risk;
168
the regression coefficient between the capital availability and the initial returns
appears to be negative but statistically insignificant in both models. The findings
conjecture that the market participants raise higher demand for firms that hold their
major earnings as retained earnings. The firms with greater capital availability
generate high IER in the primary market.
The regression coefficients of efficiency risk related to initial excess returns
appear to be positive and significant. This finding supports the proposed hypothesis of
H15c. The findings reveal that less operating efficient firms before the IPOs are
considered riskier firms and generate lower initial returns and vice versa. The capacity
risk is estimated through the percentage of proceeds utilization plan over total
proceeds as disclosed in the prospectuses. The coefficient of capacity risk related to
initial excess returns appears to be positive but statistically insignificant. Firms
proposing more funds for investment activities are viewed as riskier IPOs and
generate higher initial excess returns. The correlation matrix also validates the finding
of regression analysis. The findings contradict with Reber & Vencappa (2016) who
investigate the positive association of the uses of IPO proceeds as investment related
to initial excess returns. The firms having higher returns volatility are considered as
riskier firms and consequently, generates excess returns during early trading days.
The coefficient of firm beta is found to be positive and statistically significant at the
95% level of confidence related to IERs. This implies that the finding of this risk
factor is in line with the working hypothesis of H15e. The findings of the correlation
matrix (see Table 4.23) also support the theoretical and empirical literature of firm
beta. The finding of firm’s beta is in line with existing literature (Mumtaz, Smith &
Ahmed, 2016; Afza, Yousaf & Alam, 2013, Banerjee, Dai & Shrestha, 2011; Beneda
& Zhang, 2009; Sohail & Raheman, 2009; Sohail & Nasr, 2007) but is contradicted
by Javid & Malik (2016), Kafayat & Farooqi (2014). The last risk factor is offer size,
which generally used as a proxy for the level of risk of IPO firms (Bessler and Thies,
2007; Aggarwal, Liu and Rhee, 2008). The firms offere large offer size in the IPOs is
considered as riskier and generate higher initial excess returns. The coefficient of
offer size is negative but statistically insignificant. This implies that the firms offered
more capital seems to be priced higher in early trading days. This finding is consistent
with Reber & Vencappa (2016), Javid & Malik (2016), Banerjee, Dai & Shrestha
169
(2011) and Kerins, Kutsuna & Smith (2007). Therefore, it has been concluded that the
pre-IPO risk factors have the influence on the initial excess returns.
The high underwriters prestige dilute the impact of uncertainty of IPOs that
priced higher at the IPO pricing decision and lower the initial excess returns in the
early trading days. Therefore, the investors are intentionally to pay more prices for
IPOs when launched by prestigious underwriters. The coefficient of underwriter
reputation is found to be negative and statistically significant at the 90% level of
confidence related to the initial excess returns. This implies that the IPOs
underwritten by prestigious underwriters are priced higher on the first day of trading
and support the evidence of working hypothesis of H16a. These findings are
marginally consistent with Jenkinson, Jones & Suntheim (2016), Reber & Vencappa
(2016), and Wang & Yung (2011) who argue that prestigious underwriters have more
inside information because of large network with high net-worth institutions and/or
individuals resulting greater price revision in auction and lower price volatility on first
trading day. On the other side, Banerjee, Dai & Shrestha (2011), Lowry, officer &
Schwert (2010) and Kerins, Kutsuna & Smith (2007) found a positive association of
underwriter reputation with initial excess returns. In this analysis, the firm age is used
as another signaling variable. The existing literature conjectures that mature firms
(older firms) have less uncertainty because they contain persistent market share in the
product market. The coefficient of firm age appears to be negative but statistically
insignificant related to the initial returns. Reber & Vencappa (2016) and Lowry,
officer & Schwert (2010) use IPOs data floated in US primary markets and assert
negative association between firm age and short-run underpricing, while in Pakistan,
Afza, Yousaf and Alam (2013) find positive association when they used corporate
governance factor as a mediator between the underpricing and firm-specific
characteristics. The greater % of shares offered in the new offering is considered as a
signal of dissatisfaction of the initial shareholders on the firm’s future prospects.
Based on the theoretical and empirical literature, it is conjectured that the offer size is
positively related to initial valuation as proposed in the H16c hypothesis. The
coefficient of the percentage of shares offered is found to be negative but statistically
insignificantly related to IERs. The finding of shares retention at the time of formal
listing is inline with the existing literature of Pakistan Javid and Malik (2016), Afza,
Yousaf and Alam (2013) and Sohail and Raheman (2009) but contrary with developed
markets (Reber & Vencappa, 2016).
170
In model 2, the standardized residual series extracted from the IPO valuation
model is added to control the unobservable effect of the initial valuation on the initial
returns. The coefficient of residual is found to be negative and statistically significant
at the 90% level of confidence related to initial returns. The finding implies that
higher values of residuals indicate that IPOs are valued higher relative to the IPOs
fundamentals and consequently lower the IERs.
4.3.4 The Sensitivity Analysis
Table 4.25 report the findings of IER models for the non-privatization IPO
sample. On the same pattern as in Table 4.24, the standardized residuals series from
the non-PIPO valuation model is added in model 2. In this section initial excess
returns (IER) of non-PIPOs used as a dependent variable and t-statistics have been
estimated through Huber/White standard errors. The findings of IER models assist to
estimate the cross-sectional determinants of the initial returns.
The intercept value shows a robust finding of underpricing of the non-
privatization IPOs but the underpricing level is lower than the full IPOs sample. The
justification of lower underpricing is that the full sample is included privatization
IPOs as well and PIPO is confirmed to have higher IER as discussed in the descriptive
statistics part (see Table 4.20). The coefficient of the book value of
shareholder’s equity over offer prices is found to be positively related to IERs and this
finding is similar to the main analysis. However, the coefficient earnings-price ratio
appears to be negatively related to IERs but statistically insignificant (found
significant in model 2). These resuls are not consistent with main analysis and
proposed hypothesis of H14b.
As discussed in methodology chapter, the proposed hypotheses related to ex-
ante uncertainty factors are developed on the basis of risk-aversion assumption. It is
conjectured that initial excess returns are the increasing function of risk factors.
Therefore, the IPO firms are priced higher on the 1st trading date because investors
demand compensations for riskier investments. Except the capital availability risk
factor, the coefficients of all ex-ante risk factors are consistent with the proposed
hypotheses related IERs and with main analysis of a non-PIPOs sample. In the model
1, the capital availability risk, efficiency risk, firm beta and offer size factors are
significant determinants. However, in the model 2, the residuals series is added to
171
control the effect of unobservable factors on the IERs whereas the financial leverage,
efficiency risk and offer size are significant determinants related to IERs.
Table 4. 25: Regression Analysis of IER Models using Non-PIPO Sample
Variable Model 1 (IER) Model 2 (IER)
Fundamental Factors
Intercept 0.9078 1.5893*
(0.9698) (1.7057)
Book Value (BV/OP) 0.6105 1.0235**
(1.0790) (2.0436)
Earnings (EPS/OP) -0.0045 -0.3145*
(-0.0381) (-1.9308)
Risk Factors
Financial Leverage 0.0040 0.0051*
(1.6341) (2.1012)
Capital Availability Risk 0.0031* 0.0019
(2.0765) (1.0550)
Efficiency Risk 0.0049** 0.0060***
(3.3124) (3.7052)
Capacity Risk 0.0009 0.0011
(0.4427) (0.5051)
Firm Beta 0.0327* 0.0099
(2.2137) (1.5962)
Offer Size -0.1977* -0.2879*
(-2.7419) (-2.4254)
Signal Factors
Underwriter Reputation -0.1527 -0.1558
(-0.8938) (-1.3821)
Firm Age -0.0016 0.0012
(-0.4479) (0.3720)
Percentage of Shares Offered 0.0052 0.0078
(0.6329) (0.8799)
Valuation Residuals - -0.1720*
- (-2.4921)
Adj. R-square 0.33602 0.33113
F-Statistic 3.03646** 3.16402**
N 78 78
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
Except the percentage of shares offered factor, the findings of signaling factors
remain unchanged. In this analysis, the coefficients of underwriter reputation, firm
age and the percentage of shares offered related to initial excess returns are as
172
expected and provide supporting evidence of proposed hypotheses H16a-H16c. The
coefficient of a residual factor related to initial excess returns appears to be negative
and statistically significant at the 90% level of confidence. The finding implies that
the higher values of residuals indicating that IPOs are valued higher relative to the
IPOs fundamentals and consequently, in turn, lowers the IERs. The finding of
residuals is consistent with the working hypothesis of H17.
4.4 The IPOs Long-run Returns Analysis
This section reports the findings of long-run aftermarket performance of IPOs
floated during 2000-2012. As already discussed in methodology chapter, the long
term performance can be estimated as the adjusted returns of loyal investors who buy
shares on second trading day and keep them over 5 years. This LRR analysis attempts
to address the objective of “To investigate the usefulness of prospectus information on
the IPOs long-run adjusted returns” The LRR analysis investigates the long-run
performance phenomenon which results in a poor performance observed in longer
period as discussed in literature. The long-run aftermarket price performance is
estimated by using event-time approach (the commutative adjusted returns proposed
by Ritter (1991) and the BHAR as proposed by Barber & Lyon (1997)) and calendar-
time approach (the CAPM as proposed by Sharp-Lintner (1964), and three- and five-
factors asset pricing models as proposed by Fama-French (1993,2015)). The long-run
performance estimated through Calendar-time approach has been discussed in Section
4.5. An empirical model has been employed to estimate the effect of prospectus
information, IERs, privatization dummy and valuation residuals on the LRR. The
initial excess returns and the valuation residuals are used as a proxy for ‘mispricing’
on the first trading day.
In the long run analysis part, the sample of 65 IPOs has been used for the LRR
analysis. The inclusion of IPOs based only those firms that survive for more than five
years from the listing date in the stock exchange. There may be a survivorship bias
because new sample dropped the firms under delistings as highlight by (Shunway,
1997). The data of prospectus information such as fundamental, risk and signaling
factors have been extracted from the offering documents. Furthermore, the
standardized residual series is extracted from the valuation analysis and the IERs are
added in the LRR analysis as a proxy for the unobservable determinants of the IPO
prices and ‘mispricing’ experienced on the 1st day of trading respectively.
173
4.4.1 Descriptive Statistics of LRR
This section provides the summary statistics of equal-weighted and value-weighted
LRR up to five years. These long-run returns are estimated using BHARs and CARs
methods. As discussed earlier, the share prices and the financial variables data of IPO
firms have been obtained from the PSX DataStream and the prospectuses
respectively. The month-wise equally-weighted and value-weighted monthly returns
up to 60 months using BHARs and CARs respectively, are reported in Table A.10 and
Figure A.1. The results of value-weighted returns analysis are used as robustness
measure to validate the results estimated through equally-weighted approach.
Table 4. 26: Descriptive Statistics of equally-weighted Long-run Returns
Variables Mean Min Percentiles
Max SD N
25th 50th 75th
BHAR-EW Y1 -0.0771 -0.9148 -0.6603 -0.2184 0.0916 3.3323 0.8043 65
BHAR-EW Y2 -0.1721 -1.6264 -0.8831 -0.3368 0.2763 3.7589 1.0645 65
BHAR-EW Y3 -0.2352 -2.0876 -0.9933 -0.5671 -0.0145 4.1963 1.3782 65
BHAR-EW Y4 -0.3388 -2.7575 -1.3599 -0.7431 -0.2158 12.6947 2.2289 65
BHAR-EW Y5 -0.6522** -4.6458 -1.6259 -0.9507 -0.1985 8.6673 2.0733 65
CAR-EW Y1 -0.1305 -1.2664 -0.6440 -0.1968 0.1896 1.9937 0.6668 65
CAR-EW Y2 -0.1833* -2.0810 -0.8594 -0.2590 0.4044 2.2881 0.8445 65
CAR-EW Y3 -0.2462* -2.4753 -1.0043 -0.3726 0.3457 4.5127 1.1153 65
CAR-EW Y4 -0.3188** -2.2126 -1.0449 -0.6382 0.3171 3.3329 1.1251 65
CAR-EW Y5 -0.2937** -2.2540 -1.2552 -0.2175 0.4433 3.5519 1.1436 65
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4. 27: Descriptive Statistics of value-weighted Long-run Returns
Variables Weighted Average Min Percentiles
Max SD N
25th 50th 75th
BHAR-VW Y1 -0.1028 -0.0478 -0.0027 -0.0011 0.0004 0.0199 0.0090 65
BHAR-VW Y2 -0.1745* -0.0413 -0.0043 -0.0018 0.0010 0.0449 0.0121 65
BHAR-VW Y3 -0.4338*** -0.0836 -0.0057 -0.0029 -0.0001 0.0281 0.0185 65
BHAR-VW Y4 -0.3557* -0.1038 -0.0086 -0.0037 -0.0009 0.0838 0.0230 65
BHAR-VW Y5 -0.5843** -0.1908 -0.0104 -0.0040 -0.0009 0.0947 0.0321 65
CAR-VW Y1 -0.0622 -0.0462 -0.0024 -0.0006 0.0013 0.0428 0.0112 65
CAR-VW Y2 -0.0154 -0.0373 -0.0031 -0.0011 0.0027 0.0531 0.0122 65
CAR-VW Y3 -0.2583** -0.0655 -0.0070 -0.0014 0.0018 0.0333 0.0147 65
CAR-VW Y4 -0.2501* -0.0669 -0.0069 -0.0015 0.0013 0.0475 0.0158 65
CAR-VW Y5 -0.3206* -0.1395 -0.0056 -0.0007 0.0014 0.0574 0.0229 65
***Significant at 1%, **Significant at 5% and *Significant at 10%
174
Table 4.26 and Table 4.27 present the descriptive statistics of equally-weighted and
value-weighted long-run returns using BHAR and CAR respectively. Table 4.26
portrays that the average returns using BHAR for year 1 to year 5 are -7.71%, -
17.21%, -23.52%, -33.88% and -65.22% respectively. However, the average returns
using CAR for year 1 to year 5 are -13.05%, -18.33%, -24.62%, -31.88% and -29.37%
respectively. These findings show that the LRR found a negative trend when moving
from year 1 to year 5 in the both BHAR and CAR techniques. However, the results
estimated through the BHARs are more negative than those estimated through the
CAR method. Various researchers Barber and Lyon (1996), Eckbo and Norli (2005)
and Choi, Lee and Megginson (2010) argue that the accuracy of LRRs estimated
through BHAR is more than the accuracy of CAR. These results are in sequence with
some Pakistani authors Sohail and Nasr (2007), Mumtaz and Smith (2015), Javid and
Malik (2016) and Mumtaz, Smith and Ahmed (2016) and with international studies
Loughran and Ritter (1995), Stehle (2000), Kooli & Suret (2004), Chorruk &
Worthington (2010), and Bossin & Sentis (2014). The LRRs estimated over five years
are greater than the other emerging countries (Malaysia, India, Taiwan, and China)
and all developed countries except the South Africa and Switzerland (See Table 2.4-
2.7). In this study, the value-weighted returns are used as a robustness measure for the
equally-weighted returns and Table 4.27 document the value-weighted long-run
underperformance using BHARs and CARs. The findings of value-weighted long-run
performance found a similar trend that is observed in the equally-weighted long-run
returns.
Table 4. 28: Year-wise Long-run Returns Analysis using BHAR and CAR
Year BHAR-EW BHAR-VW CAR-EW CAR-VW
Year 1 -0.0771 -0.1028 -0.1305 -0.0622
Year 2 -0.1721 -0.1745* -0.1833* -0.0154
Year 3 -0.2352 -0.4338*** -0.2462* -0.2583**
Year 4 -0.3388 -0.3557* -0.3188** -0.2501*
Year 5 -0.6522** -0.5843** -0.2937** -0.3206*
***Significant at 1%, **Significant at 5% and *Significant at 10%
175
Figure 4. 5: Year-wise Long-run Returns Analysis using BHAR and CAR
Table 4.28 and Figure 4.5 present the summary view of long-run equal- and value-
weighted returns measured by the BHAR and CAR methods. These findings affirm
the long-run underperformance anomaly exist in Pakistan. The investors who
purchase shares on day 2 and keep them over five years, in turns, negative returns.
Table 4. 29: Year-wise Long-run Returns Analysis using BHAR
Year N BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
2000 3 -0.0925 -0.3068 -0.4362 -1.5848 -3.1674
2001 2 -0.2861 0.7083 1.4276 -0.3587 0.1158
2002 2 0.4224 0.8050 2.7676 7.2426 5.7697
2003 2 0.0499 -0.4312 -0.5035 -0.8528 -0.4154
2004 7 -0.2582 -0.4070 -0.9482 -0.4727 -1.0147
2005 13 -0.1991 0.0324 0.3734 -0.1984 -0.6096
2006 2 -0.5461 -0.6100 -0.3312 -0.6278 -0.8539
2007 10 0.2927 0.1001 -0.3556 -0.5274 -0.3384
2008 9 -0.1961 -0.5392 -0.6619 -0.8810 -1.3411
2009 3 1.1459 -0.1139 -0.2040 0.4460 0.6642
2010 5 -0.5795 -0.6096 -0.9924 -1.6142 -2.0962
2011 4 0.2604 0.6856 0.1644 0.7480 0.1468
2012 3 -0.7867 -1.4282 -1.4957 -1.7483 -1.2479
-0.7000
-0.6000
-0.5000
-0.4000
-0.3000
-0.2000
-0.1000
0.0000
Year 1 Year 2 Year 3 Year 4 Year 5
BHAR-EW
BHAR-VW
CAR-EW
CAR-VW
176
Table 4. 30: Year-wise Long-run Returns Analysis using CAR
Year N CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
2000 3 -0.0953 -0.2155 -0.2885 -0.5187 -1.1058
2001 2 -0.2755 0.4327 0.4542 0.2685 0.4924
2002 2 0.2760 0.3613 0.7250 1.0853 0.9175
2003 2 0.3978 0.0926 0.1084 0.1941 0.1179
2004 7 0.0128 0.0368 -0.1320 -0.0194 -0.2668
2005 13 -0.1366 0.0942 0.4175 0.2404 0.2560
2006 2 -0.7642 -0.6106 -0.8579 -1.0471 -1.4254
2007 10 0.2847 0.0537 -0.5260 -0.9141 -0.2984
2008 9 -0.4661 -0.9118 -1.0979 -1.0627 -1.0896
2009 3 0.0878 -0.0477 -0.1469 -0.0607 -0.2708
2010 5 -0.6015 -0.5657 -0.6408 -0.6837 -1.0211
2011 4 0.1120 0.1469 0.1493 0.3475 0.2581
2012 3 -0.7119 -1.0834 -0.7713 -0.7463 0.0802
Figure 4. 6: Year-wise Long-run Returns using BHAR and CAR
Table 4.29 and Table 4.30 present the LRRs of IPOs listed in PSX during 2000-2012.
In the LRRs analysis using BHAR, the firms went public during 2001 and 2002
outperform the market up to year 3 and year 5. However, the firms went public during
2003 and 2008 show underperformance thorough 3- and 5-year periods. Again, firms
went public in 2009 and 2011 show positive long-run returns. The large
underperformance is observed often in the hot issue phases while over performance is
observed in the cold issues. This implies that the firms outperform the market that
went public after the US internet bubble crisis in 1999-2001 and the US subprime
-4.0000
-3.0000
-2.0000
-1.0000
0.0000
1.0000
2.0000
3.0000
4.0000
5.0000
6.0000
7.0000
BHAR1
BHAR3
BHAR5
CAR1
CAR3
CAR5
177
mortgage crisis in 2008-2009 periods. The results of CARs analysis, the IPOs listed
during 2001-2003 outperform the market. However, the IPOs listed during 2004-2009
show underperformance over 3- to 5-year periods. More or less, the pattern of over-
and underperformance is similar to the long-run returns estimated through BHAR.
Table 4. 31: Sector-wise Long-run Returns Analysis using BHAR
Sector Name N BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
Automobile & Elect. 3 -0.5611 -1.1443 -0.7297 -1.5179 -2.1050
Cement 5 -0.0441 0.0914 -0.0356 -0.2082 0.1950
Chemicals 6 -0.1927 -0.5323 -1.0067 -1.4450 -1.8961
Commercial Banks 9 -0.3305 -0.4743 -0.2750 0.5728 -0.1976
Engineering 4 -0.4344 -0.8717 -0.9765 -1.0695 -0.8371
Fertilizer 1 -0.3496 0.4142 0.0038 -0.2158 -0.0531
Foods & Allied 1 1.6601 3.7589 1.9279 3.7598 2.6410
Insurance & Leasing 1 -0.2429 -0.8968 0.0417 -0.5164 -1.6259
Investment Sec/Banks 12 0.1652 -0.1657 0.0748 -0.4106 -1.1435
Modaraba 1 0.1440 2.5865 1.3535 1.5573 4.2396
Oil & Gas 3 -0.0218 -0.0647 -0.3905 0.0581 -0.2582
Power Gen. & Dist. 4 -0.1369 0.1878 -0.1514 1.2111 1.6856
Property & Investment 1 0.1660 -0.2046 -0.5891 -0.9194 -1.0491
Technology & Comm. 6 0.1715 0.3408 0.1242 -0.8700 -1.4548
Textile 6 -0.1742 -0.6016 -0.5300 -0.9710 -1.1948
Transportation & Comm. 2 0.0158 0.0381 0.0385 -0.6742 -0.5265
Table 4. 32: Sector-wise Long-run Returns Analysis using CAR
Sector Name N CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
Automobile & Elect. 3 -0.4813 -0.7756 -0.3882 -0.4932 -0.7972
Cement 5 -0.1323 -0.0684 -0.1857 -0.2942 -0.0975
Chemicals 6 -0.0730 -0.4544 -0.6614 -0.7940 -0.9473
Commercial Banks 9 -0.2901 -0.2698 -0.4167 -0.4977 -0.8801
Engineering 4 -0.3836 -0.7995 -0.9685 -1.1914 -0.5213
Fertilizer 1 -0.2522 0.4738 0.2111 0.1143 0.2092
Foods & Allied 1 0.9787 1.2584 0.8263 1.1800 0.9835
Insurance & Leasing 1 -0.1810 -0.6204 0.2138 0.0546 -0.1622
Investment Sec/Banks 12 -0.0246 -0.3164 -0.2678 -0.3944 -0.0127
Modaraba 1 0.1586 1.6904 0.9451 0.8830 1.6863
Oil & Gas 3 0.3845 0.4696 0.4524 0.5550 0.5514
Power Gen. & Dist. 4 -0.1721 0.1551 -0.0605 0.4499 0.4316
Property & Investment 1 0.2492 -0.1608 -0.6859 -1.5646 -1.5647
Technology & Comm. 6 -0.1748 0.2379 0.1292 0.0097 -0.5317
Textile 6 -0.3250 -0.5539 -0.4009 -0.4395 -0.1759
Transportation & Comm. 2 0.0405 0.0634 -0.0507 -0.2910 -0.3268
178
Figure 4. 7: Sector-wise Long-run returns Analysis using BHAR and CAR
Table 4.31 and Table 4.32 present the sector-wise over- and under-performance of
IPOs listed during 2000-2012. In the long-run returns analysis using BHAR, the
underperformance is observed in all the sectors other than the Cement, Foods &
Allied, Modaraba and Power Generation & Distribution sectors for the year 5. The
highest on average underperformance (-210.5%) is observed in the Automobile &
Electrical Goods sector while on the other side, the highest over-performance
(423.96%) is observed in Modaraba sector, In the LRRs analysis using CAR, the
underperformance is observed in all the sectors other than the firms went public from
the Fertilizer, Foods & Allied, Modaraba, Oil & Gas, and Power Generation &
Distribution sectors. The highest on average underperformance (-156.47%) is
observed in the Real Estate Investment sector while on the other side, the highest
over-performance (168.63%) is observed in the Modaraba sector,
The findings of Table 4.33 helps to investigate the relationships of LRRs using
BHAR & CAR related to privatization and non-privatization IPOs respectively. The 7
out of 65 IPOs went public due to the privatization process in Pakistan during the
sample period.
-3.0000
-2.0000
-1.0000
0.0000
1.0000
2.0000
3.0000
4.0000
5.0000
BHAR1
BHAR3
BHAR5
CAR1
CAR3
CAR5
179
Table 4. 33: Privatization and Non-Privatization IPOs Long-run Returns Analysis
Panel A
Sub-Sample N %age of IPOs BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
Privatization 7 10.77 -0.1955 -0.1514 0.1870 1.5377 0.8130
Non-Privatization 58 89.23 -0.0628 -0.1746 -0.2862 -0.5653*** -0.8291***
Wilcoxon-Mann-Whitney Rank Test 0.4127 0.9205 1.1322 2.2961** 2.2114**
Panel B
Sub-Sample N %age of IPOs CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
Privatization 7 10.77 -0.0544 0.1042 0.0015 0.1063 0.0143
Non-Privatization 58 89.23 -0.1397 -0.2180* -0.2761* -0.3701** -0.3308**
Wilcoxon-Mann-Whitney Rank Test 0.6454 1.1957 1.2591 1.3861 1.0263
***Significant at 1%, **Significant at 5% and *Significant at 10%
Figure 4. 8: Privatization and Non-Privatization IPOs Long-run Returns Analysis
Panel A of Table 4.33 presents the LRRs using BHAR of privatization and
non-privatization IPOs that took place during 2000-2012. The privatization IPOs
outperform the market from 3- to 5-year periods while non-privatization IPOs show
underperformance throughout the period. The results reveal that the PIPOs produce
negative returns in short-run periods, but outperform the market in the long term
because the PIPOs are mature enough in the product market than the non-privatization
IPOs. The findings of the Wilcoxon-Mann-Whitney rank test unveil that PIPOs are
priced differently than non PIPOs in early periods. Panel B of Table 4.33 presents the
LRRs using CAR of privatization and non-privatization IPOs that took place during
-1.0000
-0.5000
0.0000
0.5000
1.0000
1.5000
2.0000
Year 1 Year 2 Year 3 Year 4 Year 5
BHAR-PIPO
BHAR-Non PIPO
CAR-PIPO
CAR-Non PIPO
180
2000-2012. The privatization IPOs outperform the market from 2- to 5-year periods
while non-privatization IPOs show underperformance in all the time periods. These
results are consistent as estimated through the BHAR.
To examine the effect of firm size (market capitalization basis) on the long-run
returns using BHAR and CAR, we categorized the sample IPOs into four types based
on market capitalization at the offer prices. PKR 600 million, PKR 1,800 million and
PKR 3,750 millions are taken as cut-offs closet to the 1st, 2nd & 3rd quartiles
respectively. The cutoff market capitalization for small group firms is less than PKR
1,800 million.
Table 4. 34: Firm’s Size-wise Long-run Returns Analysis using BHAR
IPO Proceeds N BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
< 600 mn 17 -0.1156 0.0391 0.2916 -0.3014 -0.6047
≥ 600 and < 1,800 mn 16 0.1866 -0.2836 -0.4059 -0.9544** -1.5037**
≥ 1,800 and < 3,750 mn 16 -0.2914** -0.3679** -0.2955 0.3474 0.2000
≥ 3,750 mn 16 -0.0855 -0.0893 -0.5639** -0.4493 -0.7036**
Small Firms 33 0.0309 -0.1173 -0.0466 -0.6180** -1.0406**
Big Firms 32 -0.1884* -0.2286 -0.4297** -0.0509 -0.2518
Wilcoxon-Mann-Whitney Rank Test 0.0853 0.2165 0.2034 0.8988 1.5812
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4. 35: Firm’s Size-wise Long-run Returns Analysis using CAR
IPO Proceeds N CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
< 600 mn 17 -0.1520 -0.0170 0.1772 0.1458 0.1215
≥ 600 and < 1,800 mn 16 -0.0185 -0.2832 -0.2790 -0.5067** -0.4554*
≥ 1,800 and < 3,750 mn 16 -0.2409* -0.2628 -0.4031** -0.4370 -0.3530
≥ 3,750 mn 16 -0.1093 -0.1807 -0.5065** -0.5062* -0.5137*
Small Firms 33 -0.0873 -0.1461 -0.0440 -0.1706 -0.1582
Big Firms 32 -0.1751* -0.2217* -0.4548*** -0.4716** -0.4333**
Wilcoxon-Mann-Whitney Rank Test 0.0853 0.0853 0.9513 0.7414 0.6889
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4.34 and Table 4.35 present the long-run returns as categorized on the
basis of market capitalization using BHAR and CAR methods respectively. Table
4.34 reports the long-run returns using BHAR for the period over the year 1 to the
year 5. The findings reveal that the long-run underperformance brings up when
181
increases the firm size over all time periods. In sum, the long-run returns of small size
firms’ are less than the long-run returns of large firms. This implies that smaller size
firms are riskier than the big firms over the longer time horizons. Table 4.35
demonstrates the on average returns using CAR for the period year 1 to the year 5.
The findings report that smaller firms outperform the market over 3- to 5-year
periods. On the other side, medium to large size firms shows underperformance
throughout the time periods. In sum, the long-run underperformances of smaller size
IPOs is less than the underperformance of larger IPOs. It is conclude that smaller size
IPOs are less risky than the large size IPOs and these results are contradicted related
to BHAR results. In both cases, the Wilcoxon-Mann-Whitney rank test is insignificant
for all the time periods.
It is important to discuss IPOs long-run performance is inclined by the degree
of IERs. In general, existing literature argues that the ‘mispricing’ on day 1 may have
the lowest aftermarket performance in the long term. We categorized the IPOs sample
into four groups based on initial excess returns, IER less than 0.00% , 20.00% and
68.00% are taken as cut-offs closet to 1st, 2nd & 3rd quartiles respectively.
Table 4. 36: Initial Returns-wise Long-run Returns Analysis using BHAR
Initial Excess Returns N BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
IER ≤ 00.00 % 18 -0.3100** -0.2756 -0.6100*** -0.7858** -1.2541***
> 0.00 IER < 20.00 % 16 0.1690 -0.0142 0.1418 -0.1582 -0.5950
≥ 20.00 IER < 68.00 % 16 0.1245 0.1572 0.3114 0.3765 0.0485
IER ≥ 68.00 % 15 -0.2750** -0.5676*** -0.7706*** -0.7582*** -0.7384**
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4. 37: Initial Returns-wise Long-run Returns Analysis using CAR
Initial Excess Returns N CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
≤ 0.00 % 18 -0.3533*** -0.4282** -0.5711*** -0.4996** -0.6652***
> 0.00 and < 20.00 % 16 -0.0878 -0.1048 0.0717 -0.0688 0.1280
≥ 20.00 and < 68.00 % 16 0.1629 0.2293 0.2307 -0.0078 0.1205
≥ 68.00 % 15 -0.2217 -0.4135** -0.7044*** -0.7001** -0.7393**
***Significant at 1%, **Significant at 5% and *Significant at 10%
Table 4.36 and Table 4.37 present the LRRs as categorized on the basis of
IERs using BHAR and CAR methods respectively. In Table 4.36, this study explores
the long-run performance using BHAR for the period over year 1 to year 5. The
findings reveal that the IPOs that are underpriced and overpriced more in early days
produce high underperformance in longer period. The IPOs underpriced between the
182
20-68% perform better than the large under- and over-priced in the long run. The
results in Table 4.37 of long-run CAR performance related to IERs are consistent but
the magnitudes vary over different time periods.
The findings of Table 4.38 employed to investigate the relationships of LRRs
using BHAR & CAR related to financial and non-financial IPOs respectively. Out of
65 IPOs, 24 ( 37% approx.) firms went public in the financial sectors mainly as
commercial banks and investment/securities companies while 41 (63%) firms were
from the non-financial sectors.
Table 4. 38: Financial and Non-Financial IPOs Long-run Returns Analysis
Panel A
Sub-Sample N %age of IPOs BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
Financial 24 36.92 -0.0385 -0.1988 -0.0322 0.0146 -0.5807
Non-Financial 41 63.08 -0.0996 -0.1565 -0.3541* -0.5457** -0.6942**
Wilcoxon-Mann-Whitney Rank Test 0.7272 0.0612 0.7136 0.0748 0.4146
Panel B
Sub-Sample N %age of IPOs CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
Financial 24 36.92 -0.1116 -0.2215 -0.2704 -0.4099 -0.3381
Non-Financial 41 63.08 -0.1416 -0.1610 -0.2321* -0.2654* -0.2676*
Wilcoxon-Mann-Whitney Rank Test 0.2651 0.3874 0.6593 1.3661 0.9447
***Significant at 1%, **Significant at 5% and *Significant at 10%
Figure 4. 9: Long-run Performance Analysis of Financial and Non-Financial IPOs
-0.8000
-0.7000
-0.6000
-0.5000
-0.4000
-0.3000
-0.2000
-0.1000
0.0000
0.1000
Year 1 Year 2 Year 3 Year 4 Year 5
BHAR-Fin
BHAR-Nfin
CAR-Fin
CAR-Nfin
183
Panel A of Table 4.38 shows that non-financial IPOs produce more negative
BHAR returns from 1- to 5-year periods than the financial IPOs. The findings of the
Wilcoxon-Mann-Whitney rank test unable to unveil that the financial IPOs did not
perform differently from the non-financial IPOs in the long run. Panel B of Table 4.38
presents that the financial IPOs produce more negative returns from 1- to 5-year
periods than the non-financial IPOs. These results are not in line as estimated through
the BHAR.
4.4.2 The Univariate Analysis
Table 4.39 demonstrates the correlation (Pearson) coefficients of variables
used in the long-run performance models. To avoid the complexity, only year 1, year
3 and year 5 long-run returns are added in this analysis. In this part, the one-to-one
relationship of BHAR1Y, BHAR3Y & BHAR5Y related to each predictor is
discussed while t-statistic is reported below each correlation coefficient. The
correlation matrix related to CAR1Y, CAR3Y & CAR5Y with each predictor is
attached in appendices (see Table A.11). From the Table 4.39, it has been observed
that the correlation matrix does not report any multicollinearity problem.
The finding reveals that the correlation coefficient between the book value of
equity scaled by offer prices as well as earnings scaled by offer prices and the BHARs
(only with BHAR3Y and BHAR5Y) appears to be positive and statistically
significant. The earnings coefficient related to BHAR1Y is also found positive.
The correlation coefficient of financial leverage is significantly positive. The
coefficients of capital availability risk are found to be negatively linked with
BHAR1Y, BHAR3Y & BHAR5Y each. The efficiency risk coefficients are linked
with BHARs appear to be negative. The coefficients of capacity risk show mixed and
statistically insignificant results related to BHARs. The coefficient of firm beta is
significantly positive. The coefficient of offer size is found to be negative but
statistically insignificant. The correlation coefficient of underwriter reputation dummy
appears to be negative and significant (only with BHAR1Y) related to BHARs. This
implies that the IPOs sponsored by prestigious underwriters produce negative LRRs.
The firm age coefficient is also found to be negative means the younger firms
outperform in the long run.
184
Table 4. 39: Correlation Matrix of Variables used in LRR Models
Variables
BHAR
1Y
BHAR
3Y
BHAR
5Y
BV/
OP
EPS/
OP
Fin
Lev
Captl
Rsk
Eff
Rsk
Cpcty
Rsk
Firm
Beta
Offer
Size
Und
Rep
Firm
Age POS Priv IERs
BHAR
3Y 0.3211
2.691***
BHAR
5Y 0.0684 0.5757
0.544 5.589***
BV/ OP -0.0225 0.363 0.4276
-0.179 3.092*** 3.755***
EPS/ OP 0.1883 0.514 0.2737 0.4558
1.522 4.757*** 2.258** 4.06**
Fin Lev -0.0552 0.0338 0.2261 0.1802 -0.1018
-0.439 0.269 1.842* 1.454 -0.812
Captl
Rsk -0.0394 -0.0587 -0.0381 -0.0421 -0.3649 0.1305
-0.313 -0.467 -0.302 -0.334 -3.1*** 1.044
Eff Rsk -0.1279 -0.2176 -0.1131 -0.1782 -0.5103 0.1424 0.2498
-1.023 -1.770* -0.903 -1.438 -4.71** 1.142 2.048**
Cpcty
Rsk 0.1148 -0.0639 -0.0935 -0.1860 -0.2778 -0.0848 0.2465 0.1131
0.917 -0.508 -0.746 -1.502 -2.29** -0.675 2.019** 0.903
Firm
Beta 0.2456 -0.0561 -0.0448 -0.2665 0.2367 0.0435 -0.2812
-
0.1811 -0.1484
2.011** -0.446 -0.356 -2.1** 1.933* 0.346 -2.326** -1.462 -1.191
Offer
Size -0.0001 -0.1461 0.0495 -0.213 0.1147 0.1841 -0.1415
-
0.1694 -0.2962 0.3951
-0.001 -1.172 0.393 -1.73* 0.917 1.487 -1.134 -1.364 -2.461** 3.414***
Und Rep -0.2993 -0.1784 -0.1437 -0.142 -0.1174 -0.151 0.0088 0.1069 0.2010 -0.0573 -0.1610
-2.489** -1.439 -1.153 -1.142 -0.939 -1.208 0.070 0.853 1.628 -0.455 -1.294
185
Firm Age -0.0170 -0.1331 -0.0965 -0.068 0.1055 0.1012 -0.2612
-
0.1165 -0.5883 0.3294 0.4889 -0.2087
-0.142 -1.066 -0.770 -0.541 0.842 0.807 -2.148** -0.931 -5.77*** 2.769*** 4.448*** -1.693*
POS 0.0150 -0.0401 -0.1213 -0.082 -0.0662 -0.1792 0.1343 0.1527 0.3836 -0.0558 -0.4135 0.2131 -0.3907
0.119 -0.318 -0.970 -0.654 -0.527 -1.446 1.076 1.226 3.297*** -0.444 -3.60*** 1.731* -3.36***
Priv -0.0515 0.1072 0.2474 0.1752 0.3020 0.2193 -0.3189
-
0.2573 -0.4035 0.3464 0.5779 -0.3241 0.5205
-
0.3198
-0.409 0.856 2.026** 1.412 2.515** 1.784* -2.67*** -2.1** -3.50*** 2.931*** 5.621*** -2.7*** 4.83*** -2.6**
IERs 0.0239 -0.0368 0.0814 -0.024 0.0570 0.1656 0.0323 0.1729 0.0161 0.3544 0.0169 -0.1857 -0.0892 0.1460 0.157
0.190 -0.292 0.648 -0.193 0.453 1.333 0.256 1.393 0.128 3.008*** 0.134 -1.500 -0.711 1.172 1.271
Resi -0.0061 -0.0102 0.0394 -0.278 -0.1890 -0.0734 0.0355 0.0030 0.0480 -0.0144 -0.0314 -0.0105 0.0022 -0.024
-
0.151
-
0.112
-0.048 -0.081 0.313 -2.3** -1.527 -0.584 0.282 0.023 0.382 -0.115 -0.249 -0.083 0.018 -0.198
-
1.217
-
0.898
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
186
The coefficient of the percentage of shares offered finds to be negatively
linked with BHARs (only with BHAR3Y & BHAR5Y). The privatization IPOs
coefficient is significantly positive meant the state-owned enterprises perform better
in the longer periods. The findings of this analysis is inlinet with the descriptive
statistics of PIPOs (see Table 4.33). The coefficient of IERs and the residuals
extracted from valuation model show mixed and insignificant results.
4.4.3 The Analysis of Long-run Returns (LRR) Models
In this part, this study investigates the explanatory power of prospectus
information (fundamental, risk and signaling factors), IERs and residuals linked with
LRRs estimated through the BHAR and CAR respectively. It is widely discussed in
the descriptive statistical analysis that the long run, up to 5 years from the listing day,
underperformance in Pakistan is in line with the literature of emerging and developed
markets (see Table 2.4-2.7). Table 4.40 exhibits the regression results of BHARs long
run models for the year 1 to the year 5. In these models, BHAR1Y, BHAR2Y,
BHAR3Y, BHAR4Y and BHAR5Y are used as a dependent variable and t-statistics
have been estimated through the Huber/White standard errors. The findings of LRR
models assist to investigate the cross-sectional determinants of LRRs. Finally, a
sensitivity analysis of the findings of LRR models is also reported in the Table 4.41.
The intercept coefficient in LRR models is found to be negative and validate that the
underperformance phenomenon exist in the PSX as discussed in the descriptive
statistics of LRR analysis. The LRR models show the diligence of the
underperformance even after amending and controlling more variables in the analysis.
Table 4.40 reports that the regression coefficient of book value of
shareholder’s equity scaled by offer prices related to BHAR1Y, BHAR2Y and
BHAR3Y are found to be negative and significant, which is not inline with the
proposed hypothesis of H18a. However, in year 4 and year 5, the book value to offer
prices coefficients are found to be positive and statistically significant, which are in
sequence with extant literature and hypothesis of H18a that documented the direct
relationship of BV/OP related to LRRs. Fama and French (1995) find a positive
association between the book value of shareholder’s equity over market values and
aftermarket performance in the long run.
187
Table 4. 40: Regression Analysis of LRR Models using Full Sample
Variable BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
Fundamental Factors
Intercept 1.5978 -2.4392 1.4789 -7.3272 -2.3201
(0.741) (-0.456) (0.415) (-0.947) (-0.286)
Book Value (BV/OP) -0.3538*** -0.8128** -0.0043 1.0413** 0.7456***
(-2.84) (-3.271) (-0.025) (2.164) (3.955)
Earnings (EPS/OP) 1.5733** 4.1749*** 4.4721*** 4.2894** 2.8338**
(2.432) (2.721) (3.569) (2.271) (2.121)
Risk Factors
Financial Leverage -0.0067 -0.0129** -0.0061 -0.0003 -0.0027
(-1.102) (-2.083) (-1.309) (-0.045) (-0.713)
Capital Availability Risk 0.0079* 0.0319** 0.0258** 0.0218* 0.0163
(2.0608) (2.203) (3.469) (2.807) (1.282)
Efficiency Risk -0.0013 -0.0083 -0.0071 -0.0118* -0.0010
(-0.324) (-1.231) (-1.257) (-2.075) (-0.152)
Capacity Risk 0.0063* 0.0065 -0.0017 0.0115 0.0124*
(2.108) (0.956) (-0.276) (1.444) (2.269)
Firm Beta 0.0248** -0.0606** -0.0309** -0.0271** -0.0190*
(2.101) (-2.621) (-2.703) (-2.033) (-2.281)
Offer Size -0.1733 -0.0184 -0.2886 0.8303 0.1505
(-0.696) (-0.029) (-0.741) (0.962) (0.164)
Signal Factors
Underwriter Reputation -1.1688*** -0.0617 -0.3995 -0.1952 0.0973
(-3.89) (-0.101) (-0.834) (-0.305) (0.182)
Firm Age 0.0010 -0.0030 -0.0055 -0.0306 -0.0002
(0.338) (-0.346) (-0.345) (-1.163) (-0.011)
Percentage of Shares Offered 0.0021 -0.0500** -0.0124 -0.0570** -0.0607
(0.129) (-2.091) (-0.628) (-2.069) (-1.608)
Initial Excess Returns -0.0058*** 0.0033 0.0004 0.0089 0.0079*
(-3.31) (0.763) (0.099) (1.574) (2.793)
Residuals -0.2121 -0.2580* 0.0818 0.4302* 0.4109**
(-1.661) (-2.067) (0.581) (2.192) (2.216)
Adj. R-square 0.42465 0.35608 0.46598 0.55270 0.34054
F-Statistic 2.782*** 2.084** 3.288*** 4.657*** 1.946**
N 65 65 65 65 65
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
Their findings reveal that the book-value over market-value may possibly use
as a proxy for financial risk in return which is linked with firm’s financial distress.
The results of this analysis reveal that the BV/OP is robustly and significantly related
to LRR. The earnings disclosed in prospectus is viewed as a signal of quality firm in
188
the market and extant literature also has similar findings. The regression coefficients
of earnings over offer prices are positive and statistically significant related to year
from BHAR1Y to BHAR5Y. The findings of this analysis are consistent with the
proposed hypothesis of H18b. A sensitivity analysis is reported in Table 4.41 as
excluded the privatization IPOs from the sample based on the literature and
descriptive statistics analysis support as (1) privatization IPOs are priced differently
than the private IPOs (2) aftermarket performance inconsistent than the private IPOs,
and (3) state-owned enterprises hold a major market share in the product market due
to governmental involvement. The findings of Table 4.41 reveal that the regression
coefficients of book value and earnings scaled by offer prices are found to be negative
and positive respectively related to BHAR1Y to BHAR5Y.
In this LRR analysis, the CAR for year 1 to the year 5 also used as dependent
variable to investigate the cross-sectional determinants of LRRs. The findings of LRR
analysis using full sample and non-privatization sample are mentioned in the
appendix (see Table A.12-A.13). The findings of fundamentals (BV/OP and EPS/OP)
are same as estimated in the BHARs long-run returns models. This implies that the
book value of shareholder’s equity no more important in the longer period, however,
earnings disclosed in prospectus is a significant factor.
The proposed hypotheses related to ex-ante risk factors are based on the
positive relationship supposition between the risk and return. The regression
coefficient of financial leverage is negative and statistically insignificant linked with
BHARs for BHAR1Y to BHAR5Y. The finding is in line with Amor and Kooli
(2017) and Hedge and Miller (1996). This study also estimates the determinants of
long-run returns using non-privatization IPOs sample. The findings of financial
leverage related to BHARs are consistent with the findings by using full IPOs sample
(see Table 4.41). As a robustness measure, long-run returns using cumulative
abnormal returns method are used as a depended variable in the long-run models to
probe the impact of cross-sectional determinants on the LRR. The findings of
financial leverage coefficients are inconsistent with the findings of BHARs long-run
models, however, consistent with the related hypothesis (see Table A.12 and Table
A.13). The second financial risk factor is capital availability risk; the regression
coefficients between the capital availability and the long-run models appear to be
positive and statistically significant for year 1 to the year 5. The finding is robust to
189
the BHARs when long-run models were used using non-privatization IPOs sample.
The results of non-privatization IPOs are similar to the full IPOs sample, which are
contradicted to the proposed hypothesis of H19b. The findings conjecture that the
firms which retain less part of their net income as retained earnings generate high
returns in the long term period. The results estimated through CARs models also same
as with the BHARs models.
The regression coefficients of efficiency risk related to BHARs long-run
returns appear to be negative but statistically insignificant. The regression coefficients
extracted from the other BHARs and CARs long run models are similar, which
contradicts to the proposed hypothesis of H19c. The findings reveal that less
operating efficient firms before the IPOs generate high returns and outperform the
market in the long run. The coefficients of capacity risk related to long-run returns
appear to be positive and statistically significant in each BHARs model for year 1 and
year 5. The firms proposed more funds for investment activities are viewed as riskier
IPOs and in turn earn lower long-run returns. The findings are consistent with the
proposed hypothesis of H19d. The findingss of non-privatization IPOs sample are in
line with the full IPOs sample. The CARs long run models also produce similar
results as estimated through BHARs models. Leone et al., (2003) and Klein (1996)
explore the impact of IPO proceeds linked with aftermarket performance and
conclude the value relevance of IPO transaction proceeds information. The findings
reveal that the IPO firms; point of view, higher the utilization plans as investments
means fewer chances that the IPO firm is under their optimal production capacity
reflects lower capacity risk. On the investor view point, a large part of IPO proceeds
assigned as investment means more funds allocated to risky ventures reflects the
positive long-run returns anticipated. Amor and Kooli (2017) investigate the
association between the utilization of IPO proceeds as capital investments and
aftermarket long-run performance of newly listings. They find that the firms, which
invest more in fixed assets produce negative returns in the long run.
The firms with higher returns volatility are considered as riskier firms and
consequently, generate excess returns in the long term. The coefficient of firm beta is
found to be positive and statistically significant only for BHAR1 but appear to be
negative for BHARs for year 2 to year 5. The finding of prior market volatility is in
line with existing studies of Pakistan (Mumtaz, Smith & Ahmed, 2016; Javid &
190
Malik, 2016). The findings from non-privatization IPOs sample are in line with
estimates from the full IPOs sample. This implies that the finding of this risk factor
shows mixed association with different time periods. The last risk factor is offer size,
the coefficient of firm size is negatively related to BHARs for year 1 to the year 3.
However, for year 4 and year 5, BHARs appear to be positive and contradicts with the
theoretical base. The findings are consistent with Javid & Malik (2016) but are
contradicted with Mumtaz, Smith & Ahmed (2016), and Michel, oded & shaked
(2014).
As discussed in methodology chapter about the underwriter reputation, the
decision to hire the prestigious underwriter indicates the signal of a quality firm. The
high reputed underwriters only provide consultancy to the quality IPOs due to
maintain their standing in the financial markets. Amor and Kooli (2017) and Kerins,
Kutsuna & Smith (2007) argue that the IPOs offered through high underwriters
prestige perform well in the long run. The regression coefficient of underwriter
reputation is positive only for BHAR5Y while during year 1 to year 4 appears
negative. The results of BHARs from non-privatization IPOs sample are similarly
estimated through the full sample. The CARs long-run models also show mixed
results. This study conjectures that the underwriter prestige is not a key important
factor of long-run returns. However, it has the significant impact on initial excess
returns.
In this analysis, firm age is used as another signaling variable. The firm age is
generally considered as a proxy for IPO firm’s experience in the industry. It is
expected that older firms have the ability to generate consistent profits. Therefore, due
to the conservatism principle in risk-return assumption, investors put the high demand
of IPO firms’ shares, which results in a higher aftermarket performance. The
coefficient of firm age appears to be positive but only for early years related to the
BHARs. However, the coefficients are negative for BHARs for year 3 to year 5. This
implies that the mature firms produce positive returns for initial years but produce
negative returns are in longer periods. On the other side, the CARs long-run models
produce positive returns for year 2 to year 5, which are consistent with the extant
literature (Mumtaz, Smith & Ahmed, 2016) and proposed hypothesis of H20b.
191
Table 4. 41: Regression Analysis of LRR Models using Non-PIPO Sample
Variable BHAR1Y BHAR2Y BHAR3Y BHAR4Y BHAR5Y
Fundamental Factors
Intercept 2.9615 3.3978 5.2120 9.5279 14.6045
(1.178) (0.525) (1.016) (0.827) (1.052)
Book Value (BV/OP) -1.0416*** -0.8899 -0.0753 -1.9248* 0.0432
(-3.11) (-1.589) (-0.182) (-1.761) (0.064)
Earnings (EPS/OP) 1.9110*** 4.2626** 4.5190*** 6.7542** 4.4549**
(2.738) (2.423) (2.876) (2.137) (2.321)
Risk Factors
Financial Leverage -0.5708 -0.0171* -0.0038 -0.0114 -0.0005
(-0.004) (-1.751) (-0.518) (-1.188) (-0.051)
Capital Availability Risk 0.0087 0.0506*** 0.0279*** 0.0671*** 0.0451*
(1.361) (2.867) (3.080) (2.997) (1.882)
Efficiency Risk -0.0045 -0.0184** -0.0043 -0.0216 -0.0127
(-1.314) (-2.131) (-0.723) (-2.106)** (-1.450)
Capacity Risk 0.0046 0.0128* 0.0007 0.0257** 0.0333**
(1.124) (1.693) (0.082) (2.236) (2.558)
Firm Beta 0.0197 -0.0467 -0.0002 -0.0322 -0.0491*
(1.404) (-1.300) (-0.011) (-1.185) (-1.77)
Offer Size -0.3223 -0.8973 -0.7843 -1.8286 -2.1844
(-0.982) (-1.198) (-1.426) (-1.392) (-1.364)
Signal Factors
Underwriter Reputation -1.2539*** 0.2892 -0.4004 -0.1717 0.6955
(-4.14) (0.355) (-0.616) (-0.168) (0.945)
Firm Age 0.0111* 0.0121 -0.0169 -0.0048 0.0533
(1.715) (1.097) (-0.783) (-0.103) (1.395)
Percentage of Shares Offered 0.0093 -0.0497* -0.0078 -0.0709*** -0.1198**
(0.575) (-1.980) (-0.340) (-2.732) (-2.289)
Initial Excess Returns -0.0072*** 0.0033 -0.0029 0.0042 0.0161***
(-4.52) (0.584) (-0.562) (0.713) (2.775)
Residuals -0.4412** -0.1979 -0.0126 -0.3528 0.3603
(-2.509) (-0.910) (-0.076) (-0.950) (0.920)
Adj. R-square 0.55043 0.41844 0.39350 0.44853 0.33907
F-Statistic 3.955*** 2.3246** 2.096152** 2.627707*** 1.657
N 58 58 58 58 58
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
The large percentage of shares offered in the IPOs generate signals of
dissatisfaction of initial shareholders on the firms’ future prospects. Based on the
theoretical and empirical literature, it is conjectured that the % of shares offered is
negatively linked with LRRs as proposed by H20c hypothesis. The % of shares
192
offered coefficient is negative and statistically significant linked with BHARs from
BHAR2Y to BHAR5Y. The finding of this analysis is consistent with the proposed
hypothesis of H20c and also can be verified from the univariate analysis (see Table
4.39).
It is hypothesized that any ‘mispricing’ on the first trading day, in the efficient
markets, is corrected in the longer period when more IPOs related information has
taken place. The regression coefficient of IERs related to BHARs appears to be
positive for all years except year 1. These findings are also similar to non-
privatization IPOs sample BHARs. On the other side, the CARs long-run models
produces negative coefficients related to IERs, which is consistent with the working
hypothesis of H22 and with Kerins, Kutsuna & Smith (2007). These findings are
consistent with signaling theory, which suggests that the large offer price discount in
the offer price by the quality IPOs used as a signal and perform better in the longer
period than the low-quality firms. Therefore, it is expected that IERs are negatively
linked with IPO long-run returns.
The standardized residual series extracted from the IPO valuation model is
added to control the unobservable effect of the initial valuation on the LRRs. The
coefficient of residual is negative and statistically significant at the 90% level of
confidence related to BHARs. The findings of CARs long-run models are robust as
concluded in the BHARs long-run models. The higher values of residuals indicates
that IPOs are valued higher in early years and adjust these prices in the long run when
more information becomes accessible, in turn lowers the LRRs.
In this section, BHAR and CAR are employed to estimate the LRRs in order
to investigate the accuracy of measuring long run returns. Based on the empirical
findings of intercepts and adjusted R-squared of BHARs and CARs (see Table 4.40
and Table 4.41), the BHAR is more prudent valuation estimator than the CAR
because the BHAR is a geometric product of the spread in compound returns of
returns vs compound returns of what expected returns over a time period.
4.5 IPOs Long-run Performance Using Calendar-Time Approach
This section provides the insights of long term aftermarket performance of
firms that went public during 2000-2012. As already discussed in methodology
193
chapter, the LRRs can be estimated as the adjusted returns of loyal investors who
purchase shares in primary market and hold them up to 5 years. This LRR analysis
attempts to deal with objective: “To investigate the robustness of IPOs long-run
underperformance using asset pricing models (calendar-time approach)”. This long-
run returns analysis also validates the long-run underperformance anomaly which is
observed in section 4.4 and discussed in the literature as well. In the calendar time
analysis, the long term price performance is estimated by using the capital asset
pricing model proposed by the Sharpe-Lintner (1964) and, Fama-French three- and
five-factor models proposed by Fama & French (1993,2015). In the long run analysis
part, the sample of 65 IPOs has been used for the asset pricing models analysis. The
inclusion of IPOs is based only on those firms that at least survive for five years from
the date of formal listing. The data of IPO prices and asset pricing models’ factors
have been extracted from PSX DataStream and annual reports of 225 non-IPO firms
respectively.
4.5.1 Descriptive Statistics
This part provides the summary statistics of variables used in equally-weighted and
value-weighted (see Appendix Table A.14) long-run returns analysis. These LRRs are
estimated by approaches of calendar-time such as CAPM and Fama-French three &
five factor models. As already discussed in the earlier, the share prices and financial
variables data of IPOs and non-IPO firms have been obtained from the PSX
DataStream and annual reports. The descriptive statistics of determinants of asset
pricing models used in equally-weighted and value-weighted analysis are reported in
Table 4.42 and Table A.14 respectively. The results of value-weighted returns used as
robustness check and to validate the results estimated through equally-weighted
approach.
194
Table 4. 42: Descriptive Statistics of Variables Used in Asset Pricing Models
Variable Mean Min Percentiles
Max SD Obs. 25th 50th 75th
IPO_Rf 0.0043 -0.8236 -0.0818 -0.0105 0.0685 2.3180 0.1795 3,900
Rm_Rf 0.0071*** -0.3853 -0.0180 0.0067 0.0516 0.3074 0.0780 3,900
SMBFF3F 0.0086*** -0.2933 -0.0210 0.0049 0.0322 0.2787 0.0534 3,900
SMBFF5F 0.0086*** -0.2911 -0.0204 0.0049 0.0327 0.2819 0.0529 3,900
HML 0.0072*** -0.5069 -0.0243 0.0020 0.0279 0.5269 0.0551 3,900
RMW 0.0000 -0.3183 -0.0225 0.0005 0.0228 0.3315 0.0454 3,900
CMA -0.0043*** -0.3676 -0.0240 -0.0021 0.0168 0.4529 0.0446 3,900
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
The median excess return on IPOs portfolio over risk-free rates is found to be positive
irrespective of whether it is value-weighted or equally-weighted. The difference
between the minimum and maximum values is large for both value-weighted and
equally-weighted IPOs adjusted returns indicative the evidence of extreme values in
the sample. However, the author didn’t try to winsorize or truncate the data due to
limited sample of IPO firms as compared to other similar studies. The standard
deviation of value-weighted returns is greater than the equally-weighted returns.
Ediriwickrama & Azeez (2017) and Fama (1998) employ th Fama (1998) and value-
weighted returns since it represents the total wealth effect. The median market risk
premium is observed positive for the sample period during 2000-2012 indicates that
more investment in a risk-free asset such as treasury bills is a profitable venture than
the investing in a risky-asset such as IPOs. The mean and median excess returns of
SMBFF3F, SMBFF5F, HML and RMW appeared to be positive and statistically
significant at the 99% level of confidence except CMA.
4.5.2 The Univariate Analysis
Table 4.43 demonstrates the correlation (Pearson) coefficients of variables used in
asset pricing models. To avoid the repetition, this study only document the equally-
weighted correlation matrix in Table 4.43, while value-weighted correlation matrix is
reported in Table A.15 (see appendix). In this part, one-to-one relationship of IPO
portfolio excess returns related to the market risk premium, SMB, HML, RMW and
CMA factors is discussed while t-statistic is reported below each correlation
195
coefficient. It has been observed that the correlation matrix does not report any
multicollinearity problem.
Table 4. 43: Correlation Matrix of Variables Used in Asset Pricing Models
IPO_RF RM_RF SMBFF3F SMBFF5F HML CMA RMW
IPO_RF 1
RM_RF 0.4002*** 1
(27.268)
SMBFF3F -0.1625*** -0.4169*** 1
(-10.284) (-28.641)
SMBFF5F -0.1666*** -0.425*** 0.9965*** 1
(-10.554) (-29.319) (743.751)
HML -0.0211 -0.0891*** 0.3891*** 0.3834*** 1
(-1.3207) (-5.586) (26.368) (25.923)
CMA 0.0247 0.0196 -0.1965*** -0.1881*** -0.2504*** 1
(1.5485) (1.2273) (-12.518) (-11.963) (-16.151)
RMW -0.0601*** -0.1114*** -0.0573*** -0.0532*** -0.2173*** 0.3110*** 1
(-3.753) (-7.001) (-3.583) (-3.326) (-13.900) (20.431)
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
The univariate analysis of asset pricing models’ factors reports expected
correlations. The finding reveals that the correlation coefficient of the market risk
premium appears to be positive (0.4002) to IPOs portfolio excess returns and
statistically significant at the 1% level. This finding is consistent with the theorey and
the empirical literature on asset pricing models. The SMB is measured by the
difference between the portfolio returns of small stocks and big stocks in the month t.
The correlation coefficient between the size factors (i.e. SMBFF3F and SMBFF5F) and
IPO portfolio adjusted returns is found to be negative and statistically significant. The
results are in sequence with the theorey of size effect because small (low market
capitalization) firms produce higher returns in the market. The HML is measured by
the difference between portfolio returns of a high BTM stocks and the low BTM value
stocks in month t. The correlation coefficient of the value factor (HML) is found to be
negative by (-0.0211) related to IPO portfolio adjusted returns. The findings are in
line with the theorey of value premium effect because the growing (low book-market
ratio) firms produce higher returns in the market. The RMW is measured by the
difference between portfolio returns of robust profitability and weak profitability
196
stocks in the month t. The correlation coefficient of profitability factor (RMW) is
found to be negative by (-0.0601) related to IPO portfolio adjusted returns. The results
are in sequence with the underlying theoretical assumptions of profitability effect
because firms having low operating profitability produce higher returns as compared
to the comparable firms. The CMA is measured by the difference between the
portfolio returns of conservative investment stocks and aggressive investment stocks
in the month t. The coefficient of investment factor (CMA) is found to be positive
(0.0247).
4.5.3 The Analysis of Asset Pricing Models
In this part, the study investigates the long-run performance using the Jenson’s
alpha (coefficient of intercept term) estimated through the CAPM model proposed by
the Sharpe-Lintner (1964), FF3F & FF5F models proposed by Fama & French
(1993,2015). It is widely discussed in the descriptive statistics analysis that the long
run, up to 5 years from the listing day, underperformance in Pakistan is in line with
existing literature of emerging and developed markets (see from Table 2.4 to Table
2.7). Generally, these models produce long run risk-adjusted performance of IPOs
using calendar-time methods. In the asset pricing models, Jenssen’s Alpha is observed
the negative valuvs as expected.
Table 4. 44: Regression Analysis of Capital Asset Pricing Models
CAPMY1 CAPMY2 CAPMY3 CAPMY4 CAPMY5
Panel-A: Equally Weighted
Intercept -0.0064 -0.0053 -0.0033 -0.0038 -0.0023
Rm-Rf 0.9573*** 0.8860*** 0.8917*** 0.8862*** 0.9211***
Adj. R2 0.1676 0.1675 0.1635 0.1663 0.1602
F Statistic 156.595*** 313.382*** 456.976*** 622.012*** 743.558***
Obs. 780 1,560 2,340 3,120 3,900
Panel-B: Value Weighted
Intercept 0.0024 -0.0003 0.0002 -0.0011 -0.0002
Rm-Rf 1.1432*** 0.9791*** 0.9611*** 0.9382*** 0.9611***
Adj. R2 0.0618 0.0764 0.0859 0.0969 0.1035
F Statistic 51.287*** 128.802*** 219.776*** 334.647*** 449.936***
Obs. 780 1,560 2,340 3,120 3,900
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
197
Panel-A of Table 4.44 shows the findings of equally-weighted excess returns
after controlling the market excess returns factor. The coefficients of the intercepts for
year 1 to year 5 appear to be negative but statistically insignificant which indicates
that the IPOs underperform over the five year periods successive to formal listings.
The market excess returns (RM-RF) factor appears to be positive and significant for all
time periods at the 1% level. The findings of value-weighted excess returns validate
the long-run underperformance for all time periods except year 1 and year 3. These
results are similar with other emerging economy study (Ediriwickrama & Azeez,
2016) and developed economy studies (Thomadakis, Nounis & Gounopoulos, 2012,
Gompers, and Lerner, 2003; Espenlaub, Gregory & Tonks, 2000).
Table 4. 45: Regression Analysis of Fama-French Three Factor (FF3F) Models
FF3FY1 FF3FY2 FF3FY3 FF3FY4 FF3FY5
Panel-A: Equally Weighted
Intercept -0.0055 -0.0048 -0.0033 -0.0037 -0.0026
Rm-Rf 0.8987*** 0.8543*** 0.8799*** 0.8699*** 0.9231***
SMBFF3F -0.2170 -0.1332 -0.0610 -0.0723 -0.0038
HML 0.1136 0.1273* 0.0950 0.0757 0.0490
Adj. R2 0.1696 0.1693 0.1644 0.1669 0.1604
F Statistic 52.831*** 105.738*** 153.212*** 208.142*** 248.118***
Obs. 780 1,560 2,340 3,120 3,900
Panel-B: Value Weighted
Intercept 0.0055 0.0003 -0.0003 -0.0016 -0.0007
Rm-Rf 1.0742*** 0.9788*** 0.9802*** 0.9526*** 0.9786***
SMBFF3F -0.1615 -0.0011 0.0406 0.0293 0.0252
HML -0.0728 0.0020 0.0147 0.0301 0.0439
Adj. R2 0.0626 0.0764 0.0860 0.0970 0.1037
F Statistic 17.283*** 42.879*** 73.2914*** 111.623*** 150.217***
Obs. 780 1,560 2,340 3,120 3,900
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
Panel-A of Table 4.45 demostrates the results of equally-weighted excess
returns after controlling the market excess returns, size and the value factors. The
coefficients of intercepts for year 1 to year 5 appear to be negative but statistically
insignificant which specifys that the IPOs underperform over the five year periods
successive to formal listings. The market excess returns (RM-RF) factor appears to be
positive and strongly significant for all time periods at the 1% level each. The
findings of value-weighted excess returns also validate the long-run
198
underperformance for all time periods except year 1 and year 3. The monthly market
excess returns factors in both equally- and value-weighted panels show that IPOs are
subject to a much lower level of systematic risk. Table 4.45 reports that there is a
negative association between the IPO portfolio excess returns and the SMBFF3F
factors. This implies that the larger size firms produce lower returns than the smaller
size. The coefficient of HML appears to be positively related to IPO portfolio returns
but statistically insignificant. This implies that high growth firms generate large
returns than small firms. The findings of Fama-French three-factor analysis are
matched with existing literature of emerging economy studies (Ediriwickrama and
Azeez, 2017, 2016; Mumtaz, Smith and Ahmed, 2016; Liu, Uchida and Gao, 2014;
Moshirian, Ng and Wu, 2010; Ahmad‐Zaluki, Campbell & Goodacre,. 2007 ) and
developed economy studies (Boissin and Sentis, 2014; Brau, Couch and Sutton, 2012;
Choi, Lee and Megginson, 2010; Pukthuanthong and Varaiya, 2007; Rosa,
Velayuthen and Walter, 2003; Espenlaub, Gregory and Tonks, 2000).
Table 4.46 presents the regression results of equally- and value-weighted
Fama-French five-factor models for year 1 to year 5. In these models, the depended
variables are equally- and value-weighted monthly excess returns of IPOs portfolio.
The expnalatory determinants are RM-RF, SMBFF3F, HML, RMW (Robust profitability
minus weak profitability) and CMA (conservative investment minus aggressive
investment). The t-statistics have been estimated through the Huber/White standard
errors.
Table 4. 46: Regression Analysis of Fama-French Five Factor (FF5F) Models
FF5FY1 FF5FY2 FF5FY3 FF5FY4 FF5FY5
Panel-A: Equally Weighted
Intercept -0.0075 -0.0056 -0.0035 -0.0033 -0.0022
Rm-Rf 0.9109*** 0.8619*** 0.8873*** 0.8699*** 0.9187***
SMBFF5F -0.1869 -0.1362 -0.0522 -0.0575 0.0026
HML 0.1246 0.1597** 0.1149* 0.0826 0.0541
RMW 0.0789 0.0980 0.0599 -0.0341 -0.0809
CMA 0.1096 0.0927 0.0626 0.1033 0.1111*
Adj. R2 0.1708 0.1708 0.1651 0.1675 0.1612
F Statistic 31.876*** 64.035*** 92.279*** 125.337*** 149.694***
Obs. 780 1,560 2,340 3,120 3,900
Panel-B: Value Weighted
Intercept -0.0010 -0.0014 -0.0006 -0.0012 -0.0004
199
Rm-Rf 1.0601*** 0.9667*** 0.9769*** 0.9502*** 0.9863***
SMBFF5F -0.1894 -0.0371 0.0341 0.0259 0.0531
HML -0.0085 0.0349 0.0499 0.0629 0.0665
RMW 0.2090 0.0495 0.0490 0.0158 -0.0425
CMA 0.4191 0.3774** 0.3214*** 0.3389*** 0.3151***
Adj. R2 0.0681 0.0817 0.0907 0.1020 0.1080
F Statistic 11.305*** 27.644*** 46.538*** 70.738*** 94.311***
Obs. 780 1,560 2,340 3,120 3,900
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
Panel-A of Table 4.46 demonstrates the results of equally-weighted excess
returns after controlling the market excess returns, size, value, profitability and
investment factors. The coefficients of the intercepts for year 1 to year 5 appear to be
negative, but statistically insignificant which indicates that the IPOs underperform
over the five year periods successive to formal listings. The market excess returns
(RM-RF) factor appears to be positive and strongly significant for all time periods at
the 1% level each. The coefficients of market risk indicate a lower level of systematic
risk. Table 4.46 reports a negative association of SMBFF5F factors which means large
firms produce lower returns than the smaller firms. The coefficient of HML appears to
be positive and statistically significant in equally weighted but insignificant in value-
weighted findings. This implies that the large value firms produce greater returns than
smaller firms. The coefficient of RMW appears to be positive but statistically
insignificant. This implies that the large profitable firms produce more excess returns
than the fewer profit firms. The coefficient of CMA appears to be positive and
statistically significant only in the value-weighted analysis. This implies that the firms
put more investment in the capital expenditure generates more returns than the firms
with stagnant investment strategy. The finding of Fama-French five-factor analysis is
consistent with the existing literature of emerging economy (Ediriwickrama and
Azeez, 2017).
200
5. Chapter 5
5. Summary and Conclusion
201
The summary and conclusion chapter is split into five sections. This chapter starts
with, after conversing the key research objectives and methodologies employed for
this study, the summary of comprehensive findings is conversed that begins with the
insights of alternative valuation methods used by underwriters and the power of pre-
IPO valuations to enlighten the cross-sectional variation in the post-IPO prices of each
valuation method, the role of prospectus information on the short-run mispricing and,
longer period underperformance, and to provide comparative analysis of long-run
price performance using asset pricing models as robust measures to address the issues
related long-run returns measurement. This chapter pursued by the policy
implications, results summary, limitations and future directions f or other scholars.
The key objectives of thesis includes; (1) to provide insights of preferred
valuation methods when underwriters valuing IPOs and the important cross-sectional
determinants inview of valuation theories, (2) to compare the bias and accuracy
attached with each valuation method by using the ‘real-world’ estimates disclosed in
the prospectus documents to study that how underwriter accurately price the IPOs, (3)
to provide insights of usefulness on the prospectus information during the initial
valuations, the initial market ‘mispricing’ and longer period underperformance, and
finally, (4) to validate the rubustness of long-run underperformance using event- and
calendar-time methods to address the issues and sensitivity related LRR
measurements.
To meet above-mentioned research objectives several econometric models
were used. In the analysis of the underwriter preference of several valuation methods,
the frequency distribution and binary logit regression model for each alternative
valuation method were used. The SPE and APE were used to calculate the bias and
the accuracy attached to each valuation method respectively. The accounting-based
valuation model was used to examine the effect of fundamental, risk and signaling
factors on the initial valuations and aftermarket price performance. The CAPM,
Fama-French three- and five-factor models were used to inspect and affirm the long-
run underperformance anomaly, and to address the issues related to long-run returns
measurements.
202
5.1 The Analysis of Pre-IPO Valuation Dynamics
An extant literature on IPO valuation is especially thin regarding how underwriter
value the IPOs. In this section, the underwriter’s detailed valuation analysis data is
extracted from the prospectus documents of 88 IPOs which are analyzed for pre-IPO
valuation dynamics wrapping the period of 17 years from 2000 to 2016. This study
aims to address the questions of (i) How do investment bank select each valuation
method when valuing IPOs? (ii) To find bias and accuracy attached to each valuation
method that how underwriters accurately value the IPOs?
In this study lead underwriters repeatedly used DDM, DCF and multiples
valuation methods to value firms they bring public. This study finds that the Pakistani
lead underwriters prefer to select multiples valuation method when valuing mature
firms, the firms having less assets-in-tangibility and during bearish market sentiment
phase. The discounted cash flow method is used when valuing young firms, the firms
which gradually increase their fixed assets by investing more in capital expenditures,
profitable and rapidly growing firms. The dividend discount model is preferred by
underwriters when valuing firms that offer major part of their incomes as a corporate
payout to its shareholders. In the Pakistani primary market, the preference of DDM
should cater the demand of dividends paying equities.
In this study, Wilcoxon Sign Rank test and standard t-statistics on the medians
and mean values of signed prediction errors of each valuation method were used to
estimate the bias associated with each valuation method. The findings document that
most of the valuation methods were associated with positive values of mean and
median valuation prediction errors which were statistically significant different from
zero. The findings reveal that the DDM and DCF methods seem to be unbiased value
estimators because their median valuation prediction errors were only (-1.92%) and (-
0.50%) respectively. This implies that the lead underwriters accurately estimate the
intrinsic value of newly listed firms’ equity. These findings inline with Deloof,
Maeseneire & Inghelbrecht (2009, 2002) and Francis, Olsson & Oswald (2000) but
contrary to Roosenboom (2012); Cassia, Paleari & Vismara, 2004).
This study presents the mean, median and the percentage of valuation errors
are within 15% or less of actual estimates of absolute prediction errors to estimate the
accuracy attached to each valuation method. Our findings reveal that the DCF
203
estimates were significantly smaller (20.28%) and the degree of central tendency of
percentage is within 15% was the highest (45.00%) than other methods. This implies
that the valuation accuracy of DCF was the highest. The results of DCF highest
valuation accuracy are consistent with Deloof et al., (2009, 2002), and Berkman,
Bradbury & Ferguson (2000). On the other side, the P/E ratio estimates were
significantly larger (34.28%) and the degree of central tendency of percentage is
within 15% was lowest (25.00%) than other methods. This implies that the valuation
accuracy of P/E ratio was smallest. The results of P/E lowest valuation accuracy is
consistent with Goh, Rasli, Dziekonski & Khan (2015), Deloof et al., (2009, 2002),
Cassia, Paleari & Vismara (2004), Berkman, Bradbury & Ferguson (2000) and Kim &
Ritter, (1999).
The findings conclude that how underwriters accurately value the IPOs and to
set the preliminary offer prices with the intention to better understand why initial
‘mispricing’ exists. This communicates to the observation of Ritter & Welch (2002)
who conclude that “the solution to the underpricing puzzle has to lie in focusing on
the setting of the offer price” (p. 1803). An important observation was that investment
banks deliberately offer discount to fair value estimates to set strike price for IPOs,
cause the reason of initial underpricing. Our findings reveal that the majority
investment banks don’t completely depend on comparable multiples valuation when
valuing IPOs. These findings contrary with the extant literature that most studies rely
on the multiples methodse (Houston et al., 2006; Purnanandam & Swaminathan,
2004; Kim & Ritter, 1999).
5.2 The Analysis of Post-IPO Price Performance
In this section, 86 IPOs were analyzed for initial valuations analysis and initial market
‘mispricing’ analysis floated during 2000-2016. The sample was reduced to 65 firms
for long-run price performance analysis, those who survive more than 5 years from
the first trading date in the stock exchange, listed during 2000-2012. The average
offer size offered to general public equals PKR 992.40 million during the sample
period. The Network Microfinance Bank Limited offered minimum offer size equals
PKR 40.00 million, while the Habib Bank Limited offered highest offer size equals
PKR 12,161.25 million. The purpose of this thesis to settle research questions of (i)
“Does the prospectus information contributes to value the IPOs during ex-ante pricing
204
decision process? (ii) Does the prospectus information have an explanatory power to
explain the short-run underpricing and long-run underperformance? and Does the
Calendar-time approach validate the IPOs long-run underperformance anomaly?”.
The initial valuations analysis has been carried out on the basis of accounting
based valuation models, both at the offer prices and the 1st trading day market prices.
This model theorizes that the preliminary prices of IPOs are an increasing function of
fundamental and signaling factors, while a decreasing function of risk factors. The
findings reveal that the initial prices mainly depend on the fundamental factors, such
as book value and dividend payouts. The dummy variable of negative earnings was
appear to be positive and significant by linked with initial prices. Only four out of six
risk factors, the financial leverage, capacity risk, firm beta and offer size, are
statistically significant and validate that preliminary offer prices are the diminishing
function of risk factors. From signaling variables, only firm age was statistically
significant but the association was not as expected.
The analysis of IPOs price performance seeks to unfold two anomalies: the
short-run underpricing and long-run underperformance. The level of underpricing was
observed to be 32.85% which was greater than the US, the UK and other developed
countries (Loughran and Ritter, 2003; Ljungqvist, 2009; Lee, et al., 2012). Though,
these initial excess returns were lower than in China, Jordan, India and Sri Lank (Yu
and Tse, 2006; Marmar, 2010; Shelly and Singh, 2008; Peter, 2007). This indicates
that the initial participants buy shares at IPO offer prices and sell them on 1st trading
day earn abnormal returns of 32.85% from their investments. This research extend
underpricing analysis in various aspects such as: (1) the findings show that highest
market capitalization IPOs produce returns of 19.64%, while lowest market
capitalization IPOs produce returns of 39.46%. This implies that there is inverse
relation between level of underpricing and firm size. These results were consistent
with the assumption of ex-ante uncertainty hypothesis, (2) this study argue that the
percentage of initial underpricing of IPOs issued in hot-issue market was significantly
higher than the IPOs issued in cold-issue market. The findings of the hot-cold issues
analysis are consistent with the changing risk composition hypothesis (Ritter, 1984)
and signaling hypothesis (Allen & Faulhaber, 1989), (3) the on average underpricing
(8.46%) of IPOs issued through bookbuilding auction was lower than the IPOs
(41.23%) issued through fixed price auction. This implies that the institutional
205
investors, who are more informed, prefer bookbuilding mechanism because they get
more shares at a better price due to their active participation in the price discovery
process, (4) Even when the firms categorized to the privatization IPOs and non
PIPOs, the results show that the PIPOs underpricing was higher than the underpricing
of non PIPOs, (5) the results show that the underpricing of survivor IPOs (firms that
delist from the PSX during five years since the date of formal listing) was higher than
the non-survivor IPOs, (6) The analysis of year-wise reveal that, from 2006 to 2007,
the highest underpricing was observed because the investment community was
optimistic about economy due to high GDP growth rate, low inflation rate, stable
exchange rate and healthy FDI in that period, and (7) the findings of sector-wise
analysis revealed that the firms listed in Oil & Gas sector and Chemicals sector
produce higher initial excess returns than other sectors. The finding of IER regression
analysis reveals that the earnings disclosed before the IPO, financial leverage,
efficiency risk, firm beta and the underwriter reputation were the key determinants of
prospectus information to explain the variation in level of underpricing.
In the LRR analysis, this study observed that the newly listed firms didn’t
maintain their initial excess returns pattern and produce negative returns over five
year periods from the 1st trading date. The most common approaches such as BHAR
and CAR were employed to measure the long-run abnormal returns, both known as
Event-time technique. The BHARs produce negative returns of -23.52% and -65.22%
in year 3 and year 5 respectively. On the similar pattern, the CARs produce negative
returns of -24.62% and -29.37% in year 3 and year 5 respectively. These findings
were consistent with the existing literature of Pakistan, while this study contributes by
estimating value-weighted BHARs and CARs that validates the long run poor
performance measured by equally-weighted methods. This study extends long-run
performance analysis in various aspects such as: (1) in the long run year-wise
analysis, the large underperformance was observed often in the hot issue phases while
over performance was observed in the cold issues. This implies that the firms
outperform the market that went public after the US internet bubble crisis in 1999-
2001 and the US subprime mortgage crisis in 2008-2009 periods. The results of CARs
and BHARs were more or less similar, (2) the findings of sector-wise analysis depicts
that the firms listed in Automobile & Electrical Goods sector produce worst negative
returns, while the firms listed in Modaraba & Foods produce positive returns over
206
longer periods, (3) the privatization IPOs produce better returns than the non-
privatization IPOs (see Table 4.33) in the long run, (4) the IPOs over-priced at formal
listing by lead underwriters perform worst than IPOs deliberately under-priced at
formal listing in the long run, and (5) the financial services IPOs perform more
negative than the non-financial IPOs over a longer time period after listing. The
finding of LRR regression analysis reveals that the book value of shareholder’s
equity, earnings disclosed before the IPO, capital availability risk, firm beta,
underwriter reputation, the percentage of shares offered and initial excess returns were
significant determinants that explain the variation in the LRR.
The calendar-time approach was used as robust measure to validate long-run
underperformance. The long term performance was estimated using the Jenson’s
alpha (coefficient of intercept term) estimated through the capital asset pricing model
proposed by the Sharpe-Lintner (1964) and Fama-French three and five factor models
proposed by Fama & French (1993, 2015). The coefficients of intercept produce
negative signs for all asset pricing models which represent the negative performance
in long run. The market risk premium was the most significant determinant in all asset
pricing models, while HML-value factor (in equally-weighted FF5F) and CMA-
investment factor (in value-weighted FF5F) were also significant determinants in the
Fama-French five-factor models.
This study concludes the preference and accuracy of each valuation method
with fair-value estimates using ‘real-world’ data disclosed in the offering documents.
The findings of model’s misspecification are also verified in PSX as already talked in
the existing literature of IPO with regards to the selection of methods. The long-run
poor performance was also observed irrespective of any measurement techniques
most common even-time and calendar-time.
5.3 Policy Implications of the Study
After the significant contributions of this thesis to the existing literature on IPO and
behavioral corporate finance field, the findings of this research have significant
implications for positive social financial reforms and draw attention to the foremost
significance of numerous stakeholders engaged in the IPOs implementation process
on the PSX.
207
5.3.1 The Pakistan Stock Exchange
This study identifies a few irregularities which require some serious actions by PSX.
The findings of valuation prediction errors document that each valuation method
produces valuation bias; only DCF estimates produce consistent values. The second
issue, the deliberate discount in the offer prices by investment banks caused, the main
reason of large underpricing in the market. Due to this deliberate discount, the issuing
firms lose so much money as they expected to be from the IPO transactions. Third, a
number of IPOs could not survive more than 5 years from the 1st trading date. Due to
non-survivor IPOs, the investors community loses their major investments in the
shape of a significant drop in prices (capital loss). Fourth, non-privatization IPOs
produce more negative returns in the long run than PIPOs. Last but not least, the pace
of delisting of non-IPO firms is more than the pace of new offerings in the PSX.
A more critical review is required by PSX in order to ensure that the new
offerings are accurately valued and imitate the underlying fundamentals. The
aforementioned irregularities of issuing firms losses in terms of large underpricing
and valuation prediction errors could be irritating private firms to get listed on the
capital market. The PSX need more marketing and road shows to attract more private
firms get listed in order to control the pace of delisting and increase the pace of new
listings.
5.3.2 The Investment Banks/Underwriters
Based on the findings related to valuation dynamics, there is still more space for
further developments in the initial pricing and valuation process of IPOs by lead
underwriters. The results (see Table 4.3) show that the Pakistani underwriters
employed DDM, DCF and comparable multiples methods when valuing IPOs. The
findings also reveal that the choice of valuation methods did not inevitably rely on
firm-specific characteristics and market-related factors. The analysis related to the key
determinants of each valuation method show that the investment banks offered price
discounts without taking any special care to the firm-specific characteristics and the
economic environment of market. Due to this indiscretion, the findings of valuation
prediction errors show that each valuation method produces positive bias. This
analysis suggests that the forthcoming IPOs should be valued and priced based on
their unique firm-specific features and the economic environment of market.
208
The findings show that the IPOs sponsored by prestigious underwriters
produce higher offer prices (see Table 4.15) and less underpricing (see Table 4.24)
during the IPO implementation process, however, the less reputed underwriters
offered more discount in offer prices than the prestigious underwriters due to under-
subscription risk. The IER analysis also shows that the IPOs offered through
bookbuilding produces less underpricing than the IPOs offered through fixed price
without taking care of the level of underwriter reputation (see Table 4.20). The
findings can be contributed and strengthen the valuation process of underwriters in
two ways: (1) infrequent underwriters can launch potential IPOs in the period of Hot-
issue market when participants are intentionally to pay more for riskier assets without
paying more attention to fundamentals than looking for future prospects, and (2) the
IPOs offered through the bookbuilding mechanism can reduce the risk of under-
subscription because well-informed investors actively participate in the primary
market when they think prices are less than their fair value estimates.
The benefits of proposed measures continue in a way that the lead
underwriters could take the advantage of the corporate perception of their professional
services and investors confidence in their reliable valuations.
5.3.3 The Unlisted/Potential IPO-issuing Firms
The findings of this study provide learning opportunity to unlisted firms from the
oversights of previous IPO issuing firms. The oversights of previously issued IPOs
include: ‘window-dressing’ or artificially manipulate their previous financial
statements, overestimations about future prospects and over prediction about future
financial statements. These oversights caused the reason of over/underpricing
irregularities observed in the PSX. The continuation of above-mentioned oversights
by forthcoming IPO firms getting listed on the PSX could further lose the confidence
of investors in the PSX primary market, which leads to under subscription of new
offerings and loss of wealth as they expected to be.
Under the findings of this study, the issuing firm need to pay more attention
when appointing a lead underwriter, legal advisor, auditor, listing auction mechanism
(fixed price vs bookbuilding) and market sentiments before the IPO transaction.
209
5.3.4 The Investment Community
The key aim of this study was to enhancement of investors confidence and increase
the pace of new offerings on the PSX. The findings of this study have shown mixed
results as abnormal returns on the 1st trading date and capital loss in the long run such
as negative returns over three to five years.The individuals and portfolio managers
can devise their investment strategies under the findings of this study in order to earn
good returns and avoid losses.
The institutional investors can play important role to improve the accuracy of
initial valuations and pricing process in Pakistan. The institutional and high-worth
investors could add value by aggressively attending road shows and before
bookbuilding meetings with the lead underwriters to give suggestions about expected
price bands based on their own research reports. In this way, they can transform their
role in the ex ante valuations and pricing process of upcoming IPOs.
5.4 Limitations of the Study
There are a few limitations as: First, the key source of data for conducting analyses
for this study was the prospectus documents. Thus, the completion of this study only
relies on the ability to get the offering documents. Due to a limited track record of
IPOs activity in the Pakistan as compared to the other regional emerging economies,
the scholar only arranged 88 IPOs offering documents listed during 2000-2016. For
an empirical analysis, the number of IPOs taken as a sample was probably small, that
may come into some econometric and/or statistical issues in order to produce good
results when a more in-depth examination is required.
Due to the importance of behavioral finance in the decision of valuations and
investments, the behavioral finance should continue its evolution from broad
description of imperfect rationality and its consequences such as investors/market
sentiments, the analysis of valuation biases, overestimation of payoff and the
underestimation of risk. This study underweight above aspects due to more
concentration required in post-issue performance paradigms.
When estimating the long-run price performance as compared to the existing
literature, the LRR possibly sensitive to the measurement techniques and benchmarks
used. In this study, the majority IPO firms were small in size and small capitalization
210
firm’s benchmark index was needed to produce accurate long term performance,
which is unavailable in PSX. In this study when asset pricing models were engaged as
robust measures to validate the long run underperformance phenomenon, only 225
non-IPO firms instead all firms were used to estimating the market risk premium,
size, value, profitability and investment factors in order to produce more accurate
results due to time and the unavailability of complete data constraints.
5.5 Suggestions for Future Research
In this study, the extensive exertions have been done to examine the choice and
accuracy of valuation methods employed by lead underwriters and issues related to
post-issue price performance of the IPOs floated during 2000-2016. In reality, there is
always space for improvement because the world becomes a global village. Due to the
unavailability of resources and time constraints, a few suggestions for future research
directions have been discussed below.
First, a rigorous scrutiny of due diligence of company information, financial
feasibility/model, oversights of accounting standards and earnings management
related issues in the offering documents are required by regulatory institutions that
form the basis of initial valuations and post-issue price performance. This scrutiny
helps to assure the transparency of financial data reported in prospectus and
establishing the confidence of investors on IPO activity on the PSX. According to
Cervellati et al., (2013), the issues related to transparency intends to provoke a stock
crisis and the wipe out the confidence of investors on the whole financial system.
In future, the scholars can conduct a survey among investment
banks/underwriters along with the valuation information disclosed in the prospectuses
to shed more light on why and when investment banks employ a specific valuation
method. Further, the studies regarding the selection and accuracy of valuation
methods deserve more attention in similar contexts such as mergers & acquisitions
and private equity investments. According to the existing literature, the capital
expenditures used as a proxy for a need for financing, growth and cost of credit is still
debatable. Academic researchers have discussed the importance of macroeconomic
and capital market determinants when issuing firms motivated to raise capital for new
projects. In future, other variables can be added in the pre-IPO and post-IPO analysis
211
such as asset risk, total productivity factor, exchange rate and GDP growth used by
developed countries researchers.
The findings of this study reveal that the prospectus information have impact
on the initial valuations and observed underpricing in various esteems. An extant
literature on IPO also validates that the higher initial returns are taking place due to
higher demand and/or oversubscription of IPOs. So, it could be exciting to investigate
whether the prospectus information has the power to foresee the extent of subscription
of new offerings.
Last but not least, the IPO primary market of the PSX is not widely explored so far
and there are still different issues which can be examined such as impact of
macroeconomic, capital market and firm-specific determinants on in IPOs activity,
issues related to measurement of the reputation of underwriters and legal advisors, to
find the reasons of non-survivor IPOs, comparison between the IPOs issued through
the fixed price and bookbuilding mechanism, and pre- & post-issue operating
efficiencies.
212
References
213
Aamir, M., Nadeem, H. M., Naheed, K., & Khan, A. B. (2018). How Companies
Value Stock Prices After Going Public: Evidence from Emerging Pakistan
economy. Journal of Accounting and Finance in Emerging Economies, 4(1),
29-38.
Abdulai, M. S. (2015). Valuation, Pricing, and Performance of Initial Public
Offerings on the Ghana Stock Exchange (Doctoral dissertation, Walden
University).
Abubakar, N. B., & Uzaki, K. (2012). A test of initial public offerings (IPOs)
underpricing performance in Malaysian stock exchange. In Third Asian
Business and Management Conference, Osaka, Japan.
Adjasi, C. K., Osei, K. A., & Fiawoyife, E. U. (2011). Explaining underpricing of
IPOs in frontier markets: Evidence from the Nigeria Stock
Exchange. Research in International Business and Finance, 25(3), 255-265.
Afza, T., Yousaf, H., & Alam, A. (2013). Information asymmetry, corporate
governance and IPO under-pricing. Science International, 25(4), 989-997.
Agathee, U. S., Sannassee, R. V. & Brooks, C. (2012). The underpricing of IPOs on
the stock exchange of Mauritius. Research in International Business and
Finance, 26(2), 281-303.
Agarwal, S., Liu, C., & Rhee, S. G. (2008). Investor demand for IPOs and
aftermarket performance: Evidence from the Hong Kong stock
market. Journal of International Financial Markets, Institutions and
Money, 18(2), 176-190.
Aggarwal, R., Bhagat, S., & Rangan, S. (2009). The impact of fundamentals on IPO
valuations, Financial Management, 38(2), 253-284.
Aggarwal, R., Leal R., & Hernandez, L. (1993). The aftermarket performance of
initial public offerings in Latin America. Journal of Financial Management,
22(1), 42-53.
214
Aggarwal, R., Krigman, L., & Womack, K. (2002). Strategic IPO underpricing,
information momentum, and lockup expiration selling. Journal of Financial
Economics, 66, 105–137.
Aggarwal, R., & Rivoli, P. (1990). Fads in the initial public offering market?
Financial Management, 19(4), 45-57.
Ahmad-Zaluki, N. A., Campbell, K., & Goodacre, A. (2007). The long run share
price performance of Malaysian Initial Public Offerings (IPOs). Journal of
Business Finance & Accounting, 34 (1), 78-110.
Alford, A.W. (1992). The Effect of the Set of Comparable Firms on the Accuracy of
the Price-Earnings Valuation Method. Journal of Accounting Research, 30(1),
94-108.
Allen, D. E., Morkel-Kingsbury, N. J., & Piboonthanakiat, W. (1999). The long-run
performance of initial public offerings in Thailand. Applied Financial
Economics, 9(3), 215-232.
Allen, F., & Faulhaber, G. (1989). Signalling by Underpricing in the IPO Market.
Journal of Financial Economics, 23(2), 303-323.
Alqahtani , F., A., & More, V. (2012). Underpricing Phenomenon in New Zealand
IPO Market Factors Associated with Average Initial Return. Available at
SRN: https://ssrn.com/abstract=2215402.
Alvarez, S., & Gonz´alez, V. M. (2005). Signalling and the Long-run Performance of
Spanish Initial Public Offerings (IPOs). Journal of Business Finance &
Accounting, 32(2), 325–50.
Amemiya, T. (1981). Qualitative response models: A survey. Journal of economic
literature, 19(4), 1483-1536.
Amihud, Y., Hauser, S., & Kirsh, A. (2003). Allocations, Adverse Selection, and
Cascades in IPOs: Evidence from the Tel Aviv Stock Exchange. Journal of
Financial Economics, 68(1), 137-158.
Amor, S. B., & Kooli, M. (2017). Intended use of proceeds and post-IPO
performance. The Quarterly Review of Economics and Finance, 65, 168-181.
215
Ansotegui, C., & Fabregat, J. (1999). Initial public offerings on the Spanish Stock
Exchange. documento de trabajo, ESADE.
Antón, M, J,, Antonio., Rodríguez, S, &Gema. (2011). IPO methods: A survey among
Spanish investors, African Journal of Business Management, 5(9), 3745-3748.
Arosio, R., Paleari, S., & Giudici, G. (2001). The market performance of Italian IPOs
in the long-run. In Eighth Asia Pacific Finance Association Annual
Conference, Bangkok, Thailand, 22-25.
Asquith, P., M. Mikhail, & Au, A. (2005). Information Content of Equity Analyst
Reports. Journal of Financial Economics, 75, 245–82.
Aussenegg, W., & Jelic, K. (2007). Operating Performance of Newly Privatized Firms
in Central European Transition Economies. European Financial Management,
13(5), 853-879.
Aussenegg, W. (2000). Privatization versus Private Sector Initial Public Offerings in
Poland. Multinational Finance Journal, 4 (1), 69-99.
Bagherzadeh, B. (2010). Are IPOs Really Underpriced or Overpriced: Evidence from
Tehran Stock Exchange. Human Sciences MODARES, 15 (1), 77:107.
Baker, M., & Ruback, R. (1999). Estimating Industry Multiples. Working paper,
Harvard University.
Baker, M., & Wurgler, J. (2004). A Catering Theory of Dividends. Journal of
Finance, 59(3), 1125-1165.
Banerjee, S., Dai, L., & Shrestha, K. (2011). Cross-country IPOs: What explains
differences in underpricing? Journal of Corporate Finance, 17(5), 1289-1305.
Barber, B. M., & Lyon, J. D. (1996). Detecting abnormal operating performance: The
empirical power and specification of test statistics. Journal of financial
Economics, 41(3), 359-399.
Barber, B. M., & Lyon, J. D. (1997). Detecting long-run abnormal stock returns: the
empirical power and specification of test statistics. Journal of Financial
Economics, 43(3), 341-373.
216
Barber, B. M., Lyon, J. D., & Tsai, C. (1999). Improved methods for tests of long-run
abnormal returns. Journal of Finance, 54(1), 165-201.
Baron, D. P. (1982). A model of the demand for investment banking advising and
distribution services for new Issues. Journal of Finance, 37(4), 959-976.
Barry, C. B., Muscarella, C. J., Peavy, J. W. & Vetsuypens, M. R. (1990). The role of
venture capital in the creation of public companies: Evidence from the
goingpublic process. Journal of Financial Economics 27(2), 447-471.
Barth, M. E., & Kallapur, S. (1996 ). The effect of cross-sectional scale differences
on regression results in empirical accounting research. Contemporary
Accounting Research, 13(2), 527-567.
Barth, M. E., Beaver, W. H., Hand, J. R., & Landsman, W. R. (1999). Accruals, cash
flows, and equity values. Review of Accounting Studies, 4(4), 205-229.
Bartov, E., Mohanram, P., & Seethamraju, C. (2002). Valuation of internet stocks -
an IPO perspective. Journal of Accounting Research, 40(2), 321-346.
Beatty, R., & Ritter, J. (1986). Investment banking, reputation and the underpricing
of initial public offerings. Journal of Financial Economics, 15(2), 213-232.
Beatty, R., Riffe, S., & Thompson, R. (2000). IPO pricing with accounting
Information. Southern Methodist University working paper series.
Beaver, W. H., Hand, J. R., & Landsman, W. R. (1999). Accruals, cash flows, and
equity values. Review of Accounting Studies, 4(3-4), 205-229.
Beaver, W., Kettler, P., & Scholes, M. (1970). Accounting based risk vs market
Determined. Accounting Review, 45(4), 654-682.
Belghitar, Y., & Dixon, R. (2012). Do venture capitalists reduce underpricing and
underperformance of IPOs? Applied Financial Economics, 22(1), 33-44.
Beneda, N., & Zhang, Y. (2009). Heterogeneous relationship between IPO return and
risk across idiosyncratic variance characteristics. The Quarterly Review of
Economics and Finance, 49(4), 1298-1316.
217
Berkman, H., Bradbury, M. E., & Ferguson, J. (2000). The accuracy of price-
earnings and is counted cash flow methods of IPO equity valuation. Journal
of International Financial Management and Accounting, 11(2), 71-83.
Bessler, W., & Thies, S. (2007). The long-run performance of initial public offerings
in Germany. Managerial Finance, 33(6), 420-441.
Bhattacharya, (1979). Imperfect Information, Dividend Policy, and the ‘Bird in the
Hand’ Fallacy. The Bell Journal of Economics, 10(1), 259-270.
Bhojraj, S., & Lee, C. (2002). Who is my Peer ? A Valuation-Based Approach to the
election of Comparable Firms. Journal of Accounting Research, 40(2), 407–
439.
Bildik, R. & Yilmaz, M. K. (2006). The Market Performance of Initial Public
Offerings in the Istanbul Stock Exchange. Available at
SSRN: https://ssrn.com/abstract=952526
Boabang, F. (2005). The opening, short, medium and long term performance of
Canadian unit trust initial public offerings (IPOs). Journal of Business
Finance & Accounting, 32(7‐8), 1519-1536.
Boatsman, J. R., & Baskin, E. F. (1981). Asset valuation with Incomplete Markets.
Accounting Review, 56(1), 38-53.
Bodnaruk, A., Kandel, E., Massa, M., & Simonov, A. (2008). Shareholder
Diversification and the Decision to Go Public. Review of Financial Studies,
21(6), 2779-2824.
Boissin, R., & Sentis, P. (2014). Long-run performance of IPOs and the role of
financial analysts: some French evidence. The European Journal of
Finance, 20(2), 125-149.
Booth, J. R., & Chua, L. (1996). Ownership dispersion, costly information, and IPO
underpricing. Journal of Financial Economics, 41(2), 291-310.
Borges, M. R. (2007). Underpricing of Initial Public Offerings: The Case of Portugal.
International Atlantic Economic Society, 13(1), 65-80.
218
Brau, J. C., Couch, R. B., & Sutton, N. K. (2012). The desire to acquire and IPO long
run underperformance. Journal of Financial and Quantitative Analysis. 47,
493-510.
Bradley, D. J., Jordan, B. D., & Ritter, J. R. (2003). The quiet period goes out with a
bang. The Journal of Finance, 58(1), 1-36.
Brav, A. (2000). Inference in long-horizon event studies: a Bayesian approach with
application to initial public offerings. Journal of Finance, 55(5), 1979-2016.
Brav, A., & Gompers, P. A. (1997). Myth or reality? The long-run underperformance
of initial public offerings: evidence from venture and non-venture capital-
backed companies. Journal of Finance, 52(5), 1793-1821.
Brennan, M. J., & Franks, J. (1997). Underpricing, ownership and control in initial
public offerings of equity securities in the UK. Journal of Financial
Economics, 45(3), 391-413.
Brown, E. (1999). Long run performance analysis of new sample of UK IPOs.
University of Edinburgh.
Bruton, G. D., Filatotchev, I., Chahine, S., & Wright, M. (2010). Governance,
ownership structure, and performance of IPO firms: The impact of different
types of private equity investors and institutional environments. Strategic
Management Journal, 31(5), 491-509.
Burkart, M., & Lee, S. (2008). One share-one vote: The theory. Review of
Finance, 12(1), 1-49.
Cai, X., Liu, G. S., & Mase, B. (2008). The long-run performance of initial public
offerings and its determinants: the case of China. Review of Quantitative
Finance and Accounting, 30(4), 419-432.
Campbell, C. J., Rhee, S. G., Du, Y., & Tang, N. (2008). Market sentiment, IPO
underpricing, and valuation. IOWA State University.
Carter, R., & Manaster, S. (1990). Initial public offerings and underwriter reputation.
Journal of Finance, 45(4), 1045-1067.
219
Carter, R., Dark, F., & Singh, A. (1998). Underwriter Reputation, Initial Returns, and
the Long-Run Performance of IPO Stocks. Journal of Finance, 53(1), 285-
311.
Cassia, L., Paleari, S., & Vismara, S. (2004). The valuation of firms listed on the
nuovo mercato: the peer comparables approach. Advance Financial
Economics, 10, 113–129.
Celis, C., & Maturana, G. (1998). Initial public offerings in Chile. Revista Abante,
1(1), 7-31.
Cervellati, E. M., Sandro, A., & Piras, L. (2013). The need for transparency,
responsibility and accountability: The case of Facebook IPO. Corporate
Ownership & Control, 11(1), 316-325.
Chahine, S. (2008). Underpricing versus gross spread: New evidence on the effect of
sold shares at the time of IPOs. Journal of Multinational Financial
Management, 18(2), 180-196.
Chambers, D., & Dimson, E. (2009). IPO underpricing over the very long run.
Journal of Finance, 64(3), 1407-1443.
Chang, K. P., & Tang, Y. M. (2007). Pricing Taiwan's initial public offerings. Asia-
Pacific Journal of Accounting & Economics, 14(1), 69-84.
Chang, K., Wee, J. B., & Yi, H. C. (2012). Does national culture influence the firm’s
choice of debt maturity? Asia‐Pacific Journal of Financial Studies, 41(4), 424-
457.
Chen, G., & Firth, M. (1999). The accuracy of profit forecasts and their roles and
associations with IPO firm valuations. Journal of international financial
management & accounting, 10(3), 202-226.
Chen, Kevin, C. W., Zhihong, Chen, & Wei, K. C. J. (2011). Agency costs of free
cash flows and the effect of shareholder rights on the implied cost of capital.
Journal ofFinancial and Quantitative Analysis, 46(1), 171-207.
220
Cheng, C. S., & McNamura, R. (2000). The Valuation Accuracy of the Price-
Earnings and Price-Book Benchmark Valuation Models. Review of
Quantitative Finance and Accounting, 15(4), 349-370.
Cheng, S. R., & Shiu, C. Y. (2005). Examination the long-run performance of
Taiwanese initial public offerings. International journal of information and
management sciences, 16(2), 17-34.
Chi, J., McWha, M., & Young, M. (2010). The performance and the survivorship of
New Zealand IPOs. International Review of Financial Analysis, 19(3), 172-
180.
Chi, J., Wang, C., & Young, M. (2010). Long-run outperformance of Chinese initial
public offerings. Chinese economy, 43(5), 62-88.
Choi, S. D., Lee, I., & Megginson, W. (2010). Do privatization ipos outperform in the
long run? Financial Management, 39(1), 153-185.
Chong, D., & Druckman, J. N. (2010). Dynamic public opinion: Communication
effects over time. American Political Science Review, 104(4), 663-680.
Chorruk, J., & Worthington, A. C. (2010). New evidence on the pricing and
performance of initial public offerings in Thailand, 1997–2008. Emerging
markets review, 11(3), 285-299.
Colaco, H. M., De Cesari, A., & Hegde, S. P. (2013). Retail investor sentiment and
ipo valuation. University of Connecticut.
Colaco, H. M., De Cesari, A., & Hegde, S. P. (2017). The waiting period of initial
public offerings. The European Journal of Finance, 24(5), 363-390.
Colak, G., & Günay, H. (2011). Strategic waiting in the IPO markets. Journal of
Corporate Finance, 17(3), 555-583.
Cornelli, F., & Goldreich, D.,. (2001) Book building and strategic allocation, Journal
of Finance, 56, 2337-2369.
Cotter, J., Goyen, M., & Hegarty, S. (2005). Offer pricing of Australian industrial
initial public offers. Accounting and Finance, 45(1), 95−125.
221
Damodaran, A. (1994). Damodaran on Valuation: Security Analysis for Investment
and Corporate Financ, Buch. Wiley, New York.
Damodaran, A. (2007). Valuation approaches and metrics: a survey of the theory and
evidence. Foundations and Trends in Finance, 1(8), 693-784.
Danbolt, J., & Rees,W. (2002). The valuation of European financial firms. Review of
Accounting and Finance, 1(1), 5-24.
Datar, V., & Mao, D. Z. (2006). Deep underpricing of China’s IPOs: sources and
Implications. International Journal of Financial Services Management, 1(3),
345-362.
Dawson, S. M. (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.
DeAngelo, L. E. (1990). Equity Valuation and Corporate Control. Accounting
Review, 65(1), 93–112.
Deb, S. G. (2009). Some Insights into IPO underpricing in India. Vilakshan: the
XIMB Journal of Management, 6(2), 1-15.
Deloof, M., Maeseneire, W., & Inghelbrecht, K. (2002). The valuation of IPOs by
investment banks and the stock market: empirical evidence. UFSIA-RUCA,
Department of Applied Economics.
Deloof, M., Maeseneire, W., & Inghelbrecht, K. (2009). How do investment banks
value initial public offerings (IPOs)? Journal of Business Finance &
Accounting, 36(2), 130-160.
Demirakos, E. G., Strong, N. C., & Walker, M. (2010). Does valuation model choice
affect target price accuracy?. European Accounting Review, 19(1), 35-72.
Dolvin, S., & Pyles, M. (2006). Venture capitalist quality and IPO certification.
Venture Capital, 8(4), 353-371.
Drake, P. D., & Vetsuypens, M. R. (1993). IPO underpricing and insurance against
legal liability. Financial Management, 22(1),64-73.
222
Drobetz, W., Kammermann, M., & Wälchli, U. (2005). Long-run performance 174 of
initial public offerings: the evidence for switzerland. Schmalenbach Business
Review (sbr), 57(3), 1-69.
Easton, P., & Harris,T. (1991). Earnings as an explanatory variable for returns.
Journal of Accounting Research, 29(1), 19-36.
Easton, P., & Sommers, G. A. (2003). Scale and scale effect in market-based
accounting research. Journal of Business Finance and Accounting, 30(2), 55-
72.
Ebnezar, J., Nagarathna, R., Bali, Y., & Nagendra, H. R. (2011). Effect of an
integrated approach of yoga therapy on quality of life in osteoarthritis of the
knee joint: A randomized control study. International journal of yoga, 4(2),
55-63.
Eckbo, B. E., & Norli, Ø. (2005). Liquidity risk, leverage and long-run IPO
returns. Journal of Corporate Finance, 11(1-2), 1-35.
Ediriwickrama, T. C., & Azeez, A. A. (2016). Multi factor explanation to IPO long
run underperformance anomaly: Sri Lankan evidence. The International
Journal of Accounting and Business Society, 23(2), 15-27.
Ediriwickramaa, T. C., & Azeeza, A. A. (2017) Illiquidity Based Factor Construction
in Asset Pricing: An Analysis on Long Run Performance of Sri Lankan Initial
Public Offering Stocks. Colombo Business Journal,8(1), 1-22.
Eijgenhuijsen, H., & Rob, V. D. R. (1997). Initial Public Offerings in Mexico and
Argentina 1991-1994. Research Memorandum 9, VU University Amsterdam,
Faculty of Economics, Business Administration and Econometrics.
Espenlaub, S., & Tonks, I. (1998). Post‐IPO directors’ sales and reissuing activity:
An empirical test of IPO signalling models. Journal of Business Finance &
Accounting, 25(10), 1037-1079.
Espenlaub, S., Gregory, A., & Tonks, I. (2000). Re-assessing the long-term
underperformance of UK initial public offerings. European Financial
Management, 6(3), 319-342.
223
Espenlaub, S., Gregory, A., & Tonks, I. (2000). Re‐assessing the long‐term
underperformance of UK Initial Public Offerings. European Financial
Management, 6(3), 319-342.
Fama, E., & French, K. (1992). The Cross-Section of Expected Returns. Journal of
Finance, 47(2), 427-465.
Fama, E., & French, K. R. (1993). Common risk factors in stock and bond returns.
Journal of Finance, 33, 3-56.
Fama, E., & French, K. (1995). Size and Book-to-Market Factors in Earning and
Return. Journal of Finance, 50(1), 131-155.
Fama, E., & French, K. (1997). Industry Costs of Equity. Journal of Financial
Economics, 43(2), 153-194.
Fama, E., & French, K. (1998). Value versus Growth: The International Evidence.
Journal of Finance, 53(6), 1975-1999.
Fama, E. F. (1998). Market Efficiency, Long-Term Returns, and Behavioral Finance.
Journal of Financial Economics, 49(3), 283-306.
Fama, E., & French, K. (2015). A five-factor asset pricing model. Journal of
Financial Economics, 116(1), 1–22.
Feltham, G. A., Hughes, J. S., & Simunic, D. A. (1991). Empirical assessment of the
impact of auditor quality on the valuation of new issues. Journal of
Accounting and Economics, 14(4), 375-399.
Fernandez, P. (2001), ‘Internet Valuations: The Case of Terra-Lycos’, Working Paper
(IESE Business School).
Florio, M., & Manzoni, K. (2004). Abnormal Returns of UK Privatizations: from
Underpricing to Outperformance. Applied Economics, 36(2), 119-136.
Field, L. C., & Sheehan, D. P. (2004). IPO underpricing and outside
blockholdings. Journal of Corporate Finance, 10(2), 263-280.
Firth, M. (1998). IPO Profit forecasts and Their Role in Signalling Firm Value and
Explaining Post-listing Returns. Applied Financial Economics, 8(1), 29-39.
224
Francis, J., Olsson, P., & Oswald, D. (2000). Comparing the accuracy and
explainability of dividend, free cash flow, and abnormal earnings equity value
estimates. Journal of Accounting Research, 38(1), 45-70.
Gilson, S. C., Hotchkiss, E. S., & Ruback, R. S. (2000). Valuation of bankrupt firms.
Review of Financial Studies, 13(1), 43-74.
Glordano. M., Paleari, S., & Vismara, S. (2011). IPO pricing: growth rates implied in
offer prices. Annals of Finance, 7(1), 53-82.
Goergen, M., Khurshed, A., & Mudambi, R. (2007). The long-run performance of
UK IPOs: can it be predicted? Managerial Finance, 33(6), 401-419.
Goh, C. F., Rasli, A., Dziekonski, K., & Khan, S. U. R. (2015). Market-based
Valuation Multiples: Evidence from Agribusiness Sector. Pertanika Journal of
Social Sciences & Humanities, 23(1), 209-222.
Gompers, P. A., & Lerner, J. (2003). The Really Long‐Run Performance of Initial
Public Offerings: the Pre‐Nasdaq Evidence. The Journal of Finance, 58(4),
1355-1392.
Gopalaswamy, A. K., Chaturvedi, K., & Sriram, N. (2008). Long Run Post Issue
Performance of Fixed Price and Book Built IPOs: An Empirical Study on
Indian Markets. Journal of Advances in Management Research, 5 (2), 64-76.
Gordon, M. J., & Jin, J. (1993). Risk, asymmetric payoffs, and the underpricing of
initial public offerings. Research in Finance, 11, 133-165.
Gounopoulos, D., Nounis, C. P., & Stylianides, P. (2007). The short and long term
performance of initial public offerings in the Cyprus Stock Exchange. Journal
of Financial Decision Making, 4(1), 1-28.
Govindasamy, P. (2010). The long run performance of initial public offerings in
South Africa, MBA dissertation, University of Pretoria.
Grinblatt, M., & Hwang, C. Y. (1989). Signalling and the pricing of new issues.
Journal of Finance, 44(2), 393-420.
225
Hanley, K. W., & Hoberg, G. (2010). The Information Content of IPO Prospectuses.
Review of Financial Studies, 23(7), 2821-2864.
Hao, Q. (2007). Laddering in Initial Public Offerings. Journal of Financial
Economics, 85(1), 102-122.
Hasan, T., & Quayes, S. (2008). Under-pricing of initial public offerings in
Bangladesh, Applied Financial Economics, 4(1), 5–8.
Hayn, C. (1995). The information content of losses. Journal of Accounting and
Economics, 20(2), 125-153.
Helwege, J., & Liang, N. (2004). Initial public offerings in hot and cold
markets. Journal of financial and quantitative analysis, 39(3), 541-569.
Herawati, A., Achsani, N. A., Hartoyo, S., & Sembel, R. (2017). IPO company stock
valuation analysis 2000-2014. International Journal of Organizational
Innovation, 9(3), 72-132.
Hearn, B.A. (2010). The effects of banking relationship, private equity and lead
managers on the performance of West African IPO firms. South African
Journal of Economics,79(4), 430-451.
Hirschey, M. (1985). Market structure and market value. Journal of Business, 58(1),
89-98.
Hoque, M., & Musa, M. (2001). The Long Term Performance of IPOs in Bangladesh.
Journal of Business Administration, University of Dhaka, 27(1), 64-78.
Houge, T., Loughran, T., Suchanek, G., & Yan,X. (2001). Divergence of opinion,
uncertainty, and the quality of initial public offerings. Financial Management,
30(4), 5-23.
Houston, J., James, C., & Karceski, J. (2006). What a difference a month makes:
Stock analyst valuations following initial public offerings. Journal of
Financial and Quantitative Analysis, 41(1), 111-137.
How, J., & Yeo, J. (2000). The pricing of underwriting services in the Australian
capital market. Pacific-BasinFinance Journal, 8(3), 347−373.
226
How, J., Jennifer L., & Julian, Y. (2007). The use of the comparable firm approach in
valuing Australian IPOs. International Review of Financial Analysis,16(2),
99-115.
How, J., Lam, J., & Yeo, J. (2007). The use of the comparable firm approach in
valuing Australian IPOs. International Review of Financial Analysis, 16(2),
99-115.
How, J. C., Ngo, K., & Verhoeven, P. (2011). Dividend initiations and long-run IPO
performance. Australian Journal of Management, 36(2), 267-286.
Hughes, P. J., & Thakor, A. V. (1992). Litigation risk, intermediation, and the
underpricing of initial public offerings. The Review of Financial Studies, 5(4),
709-742.
Hwang, C. Y., & Jayaraman, N. (1995). Long-term performance of IPOs and non-
IPOs: Evidence from the firms listed on the Tokyo Stock Exchange. In: Bos,
T., Fetherston, T.A. (Eds.), Advances in Pacific Basin Financial Markets. JAI
Press Inc., Greenwich, CT, 317-337.
Huyghebaert, N., & Hulle, C.V., (2006), «Structuring the IPO: Empirical evidence on
the portions of primary and secondary shares», Journal of Corporate Finance,
12, 296- 320.
Ibbotson, R. (1975). Price performance and common stock issue. Journal of
Financial Economics, 2(3), 235-272.
Ikenberry, D., Lakonishok, J., & Vermaelen, T. (1995). Market under reaction to
open market share repurchases. Journal of financial economics, 39(2), 181-
208.
Ikoku, A. E. (1998). Influence-Seeking and the Pricing of Initial Public Offerings and
Privatizations: Evidence From The Nigerian Equity Market. Available at
SSRN 57179.
Islam, M. A., Ali, R., & Ahmad, Z. (2010). Underpricing of IPOs: The Case of
Bangladesh. Global Economy and Finance Journal, 3(1), 44-61.
227
Islam, S., Lee, K., Fekete, A., & Liu, A. (2012). How a consumer can measure
elasticity for cloud platforms. In Proceedings of the 3rd ACM/SPEC.
International Conference on Performance Engineering, (pp. 85-96). ACM.
Ivashkovskaya, I., & Kuznetsov, I. (2007). An empirical study of country risk
adjustments to market multiples valuation in emerging markets: the case for
Russia. E-Journal Corporate Finance, 3, 26-52.
Jain, A. B., & Kini, O. (1994). The post-issue operating performance of IPO firms.
Journal of Finance, 49(5), 1699-1726.
Jakobsen, J. B., & Sørensen, O. (2001). Decomposing and Testing Long‐term
Returns: an Application on Danish IPOs. European Financial
Management, 7(3), 393-417.
Jakobsen, J. B., & Sørensen, O. V. (2000). Decomposing and testing long-run
returns: with an application to initial public offerings in Denmark.
Copenhagen Business School. European Financial Management, 7(3), 393-
417.
Javid, A. Y., & Malik, H. (2016). Performance and capital structure of IPOs in
Pakistan from 2000 to 2015. Financial Innovation, 2(1), 14-33.
Jegadeesh, N., & Titman, S. (1993). Returns to buying winners and selling losers:
Implications for stock market efficiency. The Journal of Finance. 48(1), 65-
91.
Jegadeesh, N., Weinstein, M., & Welch, I. (1993). An empirical investigation of IPO
returns and subsequent equity offerings. Journal of Financial
Economics, 34(2), 153-175.
Jelic, R., & Briston. R. (2003). Privatization Initial Public Offerings: the
PolishExperience. European Financial Management, 9 (4), 457-484.
Jenkinson, T., & Ljungqvist, A. (2001). Going public: The theory and evidence on
how companies raise equity finance. Oxford University Press
Jenkinson, T., & Jones, H. (2009). IPO pricing and allocation: a survey of the views
of institutional investors, Review of Financial Studies,22, 1477-1504.
228
Jenkinson, T., Jones, H., & Suntheim, F. (2016). Quid Pro Quo? What Factors
Influence IPO Allocations to Investors? The Journal of Finance, 73(5), 2303-
2341.
Jensen, & Meckling. (1976). Theory of the firm: Managerial behavior, agency costs,
and ownership structure, Journal of Financial Economics, 3, 305-360.
Jewartowski, T., & Lizińska, J. (2012). Short-and long-term performance of Polish
IPOs. Emerging Markets Finance and Trade, 48(2), 59-75.
Johnson, J., & Miller,R. (1988). Investment banker prestige and the underpricing of
initial public offerings. Financial Management, 17(2), 19-29.
Kafayat, A. (2014). The Factors Affecting the Money Left on the Table by Pakistani
IPO Issuers. Acta Universitatis Danubius. Œconomica, 10(4). 51-60.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under
risk. Econometrica: Journal of the econometric society, 47(2), 263-292.
Kaneko, T. & Pettway R. H. (2003). Auctions versus book building of Japanese
IPOs. Pacific-Basin finance journal, 11 (4), 439-462.
Kaplan, S. N., & Ruback, R. S.(1995). The valuation of cash flow forecasts: An
empirical analysis. Journal of Finance, 50(4), 1059-1093.
Kayani, S., & Amjad, S. (2011). Investor interest, under-pricing and trading volume
in Pakistan Secondary Market. Business and Economics Journal, 2, 1-15.
Keasey, K., & McGuiness, P. (1992). An empirical investigation of the role of
signalling in the valuation of unseasoned equity issues. Accounting and
Business Research, 22(86), 133-142.
Keloharju, M. (1993). The winners curse, legal-liability, and the long-run
priceperformance of IPOs in Finland. Journal of Financial Economics, 34(2),
251-277.
Kerins, F., Kutsuna, K., & Smith, R. (2007). Why are IPOs underpriced? Evidence
from Japan's hybrid auction-method offerings. Journal of Financial
Economics, 85(3), 637-666.
229
Keun Yoo, Y. (2006). The valuation accuracy of equity valuation using a combination
of multiples. Review of Accounting and Finance, 5(2), 108-123.
Khurshed, A., Mudambi, R., & Goergen, M. (1999). On the long-run performance of
IPOs. Managerial Finance, 33(6), 401-419.
Kim, J. B., Krinsky, I., & Lee,J. (1995). The role of financial variables in the pricing
of Korean initial public offerings. Pacific-Basin Finance Journal, 3(4), 449-
464.
Kim, W., & Weisbach, M. S. (2008). Motivations for public equity offers: An
international perspective. Journal of Financial Economics, 87(2), 281-307.
Kim, M., & Ritter, J. (1999). Valuing IPOs. Journal of Financial Economics, 53(3),
409-437.
Kiymaz, H. (2000). The Initial and Aftermarket Performance of IPOs in an Emerging
Market: Evidence from Istanbul Stock Exchange. Journal of Multinational
Financial Management, 10(2), 213-227.
Klein, A. (1996). Can investors use the prospectus to price IPOs? Journal of
Financial Statement Analysis, 2(1), 23-39.
Koh, F., & Walter, T. (1992). A direct test of rock's model of the pricing of
unseasoned issues. Journal of Financial Economics, 23(2), 251-272.
Komenkul, K., Brzeszczyński, J., & Sherif, M. (2012). Re-Appraising the Long-Run
Performance of Thai IPOs, Working paper.
Kooli, M., & Suret, J. M. (2004). The aftermarket performance of initial public
offerings in Canada. Journal of Multinational Financial Management, 14(1),
47-66.
Kothari, S. P., & Warner, J. B. (1997). Measuring long-horizon security price
performance. Journal of Financial Economics, 43(3), 301-341.
Kothari, S., & Zimmerman, J. (1995). Price and return models. Journal of
Accounting and Economics, 20(2), 155-192.
230
Kryzanowski, L., & Rakita, I. (2000). An empirical analysis of Canadian seasoned
equity offerings. Working paper, Concordia University.
Kucukkocaoglu, G. (2008). Underpricing in Turkey: a comparison of the IPO
methods. International Research Journal of Finance and Economics, 13(1),
162-182.
Kumar, P. C., & Tsesekos, G. P. (1993). Asymmetric information, investment
banking contracts and the certification hypothesis. Journal of Banking and
Finance, 17(!), 117-129.
Lamberto, A. P., & Rath, S. (2010). The survival of initial Public Offerings in
Australia. The International Journal of Business and Finance Research, 4(1),
133-149.
LeClair, M. S. (1990). Valuing the Closely-Held Corporation: The Validity and
Performance of Established Valuation Procedures. Accounting Horizons, 4(3),
31-42.
Lee, P. J., Taylor, S. L., & Walter, T. S. (1996). Australian IPO pricing in the short
and long run. Journal of Banking & Finance, 20(7), 1189-1210.
Lee, P. J., Taylor, S. L., & Walter, T. S. (1999). IPO underpricing explanations:
implications from investor application and allocation schedule. Journal of
Financial & Quantitative Analysis, 34(4), 425-444.
Leland, H., & Pyle, D. (1977). Informational asymmetries, financial structure, and
financial intermediation. J Finance, 32(2), 371–387
Leone, A., Rock, S., & Willenborg, M. (2003). Disclosure of intended use of
proceeds and underpricing in initial public offerings. Working paper series
University of Rochester.
Lev, B. (2001). Intangibles: Management, Measurement, and Reporting, Brookings
Institution Press: Washington D.C.
Levine, R., & Renelt, D. (1992). A sensitivity analysis of cross-country growth
regressions. The American economic review, 82(4), 942-963.
231
Levis, M. (1993). The long-run performance of initial public offerings: the UK
experience 1980-1988. Financial Management, 22(1), 28-41
Lian, Q., & Wang, Q. (2009). Does the Market Incorporate Previous IPO Withdrawals
When Pricing Second‐Time IPOs? Financial Management, 38(2), 357-380.
Lie, E., & Lie, H. (2002). Multiples Used to Estimate Corporate Value. Financial
Analysts journal, 58(2), 44–54.
Lin, H. L., Pukthuanthong, K., & Walker, T. J. (2013). An international look at the
lawsuit avoidance hypothesis of IPO underpricing. Journal of Corporate
Finance, 19, 56-77.
Liu, J., Nissim, D., & Thomas, J. (2002). Equity valuation using multiples. Journal of
Accounting Research, 40(1), 135-172.
Liu, J., Nissim, D., & Thomas, J. (2004). Price multiples based on forecasts and
reported values of earnings, dividends, sales, and cash flows: An international
analysis. Working Paper, Yale University.
Liu, X., & Ritter, J. R. (2010). The Economic Consequences of IPO Spinning.
Review of Financial Studies, 23(5), 2024-2059.
Liu, J., Uchida, K., & Gao, R. (2012). Political connections and the long-term stock
performance of Chinese IPOs. Journal of International Financial Markets,
Institutions and Money, 22(4), 814-833.
Ljungqvist, A. P. (1996). Can Firms Outwit the Market?. Timing Ability and the
Long Run Performance of IPOs. Empirical issues in raising equity capital.
Amsterdam, North Holland
Ljungqvist, A.P, & Wilhelm, W. J. (2002). IPO allocations: Discriminatory or
discretionary? Journal of Financial Economics, 65(2), 167-201.
Ljungqvist, A., T. Jenkinson, P., & Wilhelm, Jr., W. J.,. (2003). Global integration in
primary equity markets: The role of U.S. banks and U.S. investors, Review of
Financial Studies, 16, 63-99.
232
Ljungqvist, A., Marston, F., & Wilhelm, W. J. (2006). Competing for Securities
Underwriting Mandates: Banking Relationships and Analyst
Recommendations. The Journal of Finance, 61(1), 301-340.
Lo, K., & Lys, T. (2000). The Ohlson model: contribution to valuation theory,
limitations, and empirical applications. Journal of Accounting, Auditing &
Finance, 15(3), 337-367.
Logue, D. E. (1973). On the pricing of unseasoned equity offerings: 1965-1969.
Journal of Financial and Quantitative Analysis, 8(1), 91-103.
Logue, E., Rogalski, R., Seward, J., & Foster-Johnson, L. (2002). What is special
about the roles of underwriter reputation and market activities in initial public
offerings? Journal of Business, 75(2), 213-243.
Loughran, T. (1993). NYSE Vs NASDAQ Returns: Market Microstructure or the
Poor Performance of IPOs? Journal of Financial Economics, 33(2), 241-260.
Loughran, R., Ritter, J., & Rydqvist, K. (1994). Initial Public Offerings: International
insights. Pacific-Basin Finance Journal, 2(2), 165−199.
Loughran, T., & Ritter, J. (1995). The New Issues Puzzle. Journal of Finance, 50(1),
23–51.
Loughran, T., & Ritter, J. R. (2004). Why Has IPO Underpricing Changed Over
Time? Financial Management, 5-37.
Loughran, & Ritter. (2002) Why Don't The Issuers Get Upset About Leaving Money
On The Table in IPOs? Review of Financial Studies, 15, 413-443.
Loughran, T., & McDonald, B. (2013). IPO first-day returns, offer price revisions,
volatility, and form S-1 language. Journal of Financial Economics, 109(2),
307-326.
Low, S. W., & Yong, O. (2011). Explaining over-subscription in fixed-price IPOs—
Evidence from the Malaysian stock market. Emerging Markets Review, 12(3),
205-216.
233
Lowry, M., & Shu, S. (2002). Litigation Risk and IPO Underpricing. Journal of
FinancialEconomics, 65(3), 309-335.
Lowry, M. (2003). Why Does IPO Volume Fluctuate So Much? Journal of Financial
Economics, 67(1), 3-40.
Lowry, M., Officer, M. S., & Schwert, G. W. (2010). The variability of IPO initial
returns. The Journal of Finance, 65(2), 425-465.
Lyn, E. O., & Zychowicz, E. J. (2003). The performance of new equity offerings in
Hungary and Poland. Global Finance Journal, 14(2), 181-195.
Lyon, D. J., Barber, B. M., and Tsai, C. (1999). Improved methods of tests of long-
run abnormal stock returns. Journal of Finance, 54(1),165-201.
Lyandres, E., Sun, L., & Zhang, L. (2007). The new issues puzzle: Testing the
investment-based explanation. The Review of Financial Studies, 21(6), 2825-
2855.
McCarthy, E. (1999). Pricing IPOs: science or science fiction. Journal of
Accountancy, 5(3), 1-58.
McDonald, J. G., & Fisher, A. K. (1972). New‐Issue Stock Price Behavior. The
Journal of Finance, 27(1), 97-102.
Megginson, W. L., & Weiss, K. A. (1991). Venture capitalist certification in initial
public offerings. The Journal of Finance, 46(3), 879-903.
Merikas, A., Gounopoulos, D., & Nounis, C. (2009). Global shipping IPOs
performance. Maritime Policy & Management, 36(6), 481-505.
Michaely, R., & Shaw, W. (1994). The pricing of initial public offerings - test of
adverse selection and signalling theories. Review of Financial Studies, 7(2),
279-319.
Michel, A., Oded, J., & Shaked, I. (2014). Ownership structure and performance:
Evidence from the public float in IPOs. Journal of Banking & Finance, 40, 54-
61.
234
Mikkelson, W. M., Partch, & Shah, K. (1997). Ownership and Operating
Performance of Companies that Go Public. Journal of Financial Economics,
44(3), 281-307.
Miller, E. (1977). Risk, uncertainty, and divergence of opinion. The Journal of
finance,32(4), 1151–1168.
Miller, M. H., & Modigliani, F. (1966). Some estimates of the cost of capital to the
electric utility industry, 1954-57. The American Economic Review, 56(3),
333-391.
Miller, R. E., & Reilly, F. K. (1987). An examination of mispricing, returns and
uncertainty for initial public offerings. Financial Management, 16(2), 33-38.
Mills, R.W., 2005. Assessing growth estimates in IPO valuations—a case study.
Journal of Applied Corporate Finance 17, 73–78.
Moorman, E. (2010). Are IPO's more underpriced in a hot or cold market?.
Moshirian, F., Ng, D., & Wu, E. (2010). Model specification and IPO performance:
New insights from Asia. Research in International Business and
Finance, 24(1), 62-74.
Mudrik, A., & Imam, G. (2002). Analysis of factors influencing the level of under-
pricing in Jakarta Stock Exchange. Journal of Business and Accounting, 4(1),
74-88.
Mumtaz, M. Z., & Ahmed, A. M. (2014). Determinants of Underpricing of Initial
Public Offerings: Evidence from Pakistan. Journal of Business &
Economics, 6(1), 47-80.
Mumtaz, M. Z., & Smith, Z. A. (2015). Is Growth Rate Implicit In Ipo
Prices?. Journal for Economic Forecasting, 18(4), 70-89.
Mumtaz, M. Z., Smith, Z. A., & Ahmed, A. M. (2016). The Aftermarket
Performance of Initial Public Offerings in Pakistan. The Lahore Journal of
Economics, 21(1), 23-68.
235
Muscarella, C. J., & Vetsuypens, M. R. (1989). A simple test of Baron's model of
IPO underpericing. Journal of Financial Economics, 24(1), 125-135.
Nanda, V., & Yun, Y. (1997). Reputation and financial intermediation: An empirical
investigation of the impact of IPO mispricing on underwriter market
value. Journal of Financial Intermediation, 6(1), 39-63.
Nel, W. S., Bruwer, B. W., & Le Roux, N. J. (2013). The impact of industry
classification on the valuation accuracy of multiples in the South African
capital market. The International Business & Economics Research Journal
(Online), 12(1), 79.
Neuberger, B. M., & Hammond, C. T. (1974). A study of underwriters' experience
with unseasoned new issues. Journal of Financial and Quantitative Analysis,
9(2), 165-180.
Nwude, C. E. (2010). Valuation and pricing of equity securities in an emerging stock
market: Evidence from Nigerian banking sector. Available at SSRN#1649439
Ohlson, J. (1995). Earnings, book values and dividends in equity valuation.
Contemporary Accounting Research, 11(2), 661-687.
Omran, M. (2005). Under-pricing and long-run performance of share issue
privatization in the Egyptian stock market. Journal of Financial Research,
28(2), 215-234.
Reyneke, M. J. (1997). The timing and subsequent performance of initial public
offerings (IPOs) on the Johannesburg Stock Exchange. Journal of Business
Finance & Accounting, 24(10), 1401-1420.
Pagano, M., Panetta, F., & Zingales, L. (1998). Why do companies go public? An
empirical analysis. The journal of finance, 53(1), 27-64.
Paleari, S., Signori, A., & Vismara, S. (2014). How do underwriters select peers when
valuing IPOs?. Financial Management, 43(4), 731-755.
Park, Y. S., & Lee, J. J. (2003). An empirical study on the relevance of applying
relative valuation models to investment strategies in the Japanese stock
market. Japan and the World Economy, 15, 331-339.
236
Pástor, Ľ., & Stambaugh, R. F. (2003). Liquidity risk and expected stock
returns. Journal of Political economy, 111(3), 642-685.
Paudyal, K., Saadouni, B., & Briston, R. (1998). Privatization Initial Public Offering
in Malaysia: Initial Premium and Long-Term Performance. Pacific-Basin
Finance Journal, 6, 427-451.
Penman, S. (2001). Financial statement analysis and security valuation. McGraw-
Gill International Edition.
Penman, S. (2007), Financial Statement Analysis and Security Valuation (3rd ed.,
New York, NY: McGraw-Hill).
Pereiro, L. E. (2006). The practice of investment valuation in emerging markets:
Evidence from Argentina. Journal of Multinational Financial Management,
16(2), 160-183.
Perera, W., & Kulendran, N. (2012). New evidence of short-run underpricing in
Australian IPOs. Financial markets & corporate governance conference
Peter, S. (2007). Performance of initial public offerings and privatized offers
Evidence from a developing country. Department of Industrial Management,
Sri Lanka Managerial Finance, 33(10), 798-809.
Peterle, P., & Berk, A. S. (2016). IPO Cycles in Central and Eastern Europe: What
Factors Drive These Cycles?. Czech Journal of Economics and
Finance, 66(2).
Prabhala, N. R., & Puri, M. (1999). What Type of IPOs Do Underwriters Support and
Why? The Role of Price Support in the IPO Process. Graduate School of
Business, Stanford University.
Pratt, S., R. Reilly and R. Schweihs (2000) Valuing a Business (4th ed., New York,
NY: McGraw- Hill).
Purnanandam, A. K., & Swaminathan, B. (2004). Are IPOs really underpriced?
Review of Financial Studies, 17(3), 811-848.
237
Rajan, R., Servaes, H., Loughran, C. T., Petersen, M., Ritter, J., Stulz, R., & Vishny,
R. (2003). The effect of market conditions on initial public offerings. In J.
McCahery and,
Rasheed, A., Khalid Sohail, M., Din, S. U., & Ijaz, M. (2018). How Do Investment
Banks Price Initial Public Offerings? An Empirical Analysis of Emerging
Market. International Journal of Financial Studies, 6(3), 77-93.
Reber, B., & Vencappa, D. (2016). Deliberate premarket underpricing and aftermarket
mispricing: New insights on IPO pricing. International Review of Financial
Analysis, 44, 18-33.
Rees, W. P. (1997). The arrival rate of initial public offers in the UK. European
Financial Management, 3(1), 45-62.
Rees, W. P. (1999). Influences on the value relevance of equity and net income in the
UK. Managerial Finance, 25(2), 58-65.
Reilly, F. K., & Hatfield, K.(1969). Investor experience with new stock issues.
Financial Analyst Journal, 25(5), 73-80.
Richardson, G., & Tinaikar, S. (2004). Accounting based valuation models: what
have we learned? Accounting & Finance, 44(2), 223-255.
Ritter, J. R. (1984). The ‘Hot’ Issue Market of 1980. Journal of Business, 57(2), 215-
240.
Ritter, J. (1991). The Long-Run Performance of Initial Public Offering. Journal of
Finance, 46(1), 3–27.
Ritter, J. (1999). Initial Public Offerings. In D. Logue, & J. Seward (Eds.). Warren
Gorham and Lamont Handbook of Modern Finance, reprinted with
modifications in Contemporary Finance Digest, 2(1). 5−30.
Ritter, J. R., & Welch, I. (2002). A review of IPO activity, pricing and allocations.
Journal of Finance, 57(4), 1795-1828.
Ritter, J. R., & Zhang, D. (2007). Affiliated mutual funds and the allocation of initial
public offerings. Journal of Financial Economics, 86(2), 337-368.
238
Ritter, J., 2011. Equilibrium in the IPO market. Annual Review of Financial
Economics, 3, 347 – 374.
Rizwan, M. F., & Khan, S. U. (2007). Long-run performance of public vs. private
sector initial public offerings in Pakistan. The Pakistan Development Review,
46(4), 421-433.
Rock, K. (1986).Why New Issues are Underpriced. Journal of Financial Economics,
15(2), 187-212.
Roosenboom, P., van der Goot, T., & Mertens, G. (2003). Earnings management and
initial public offerings: Evidence from the Netherlands. The International
Journal of Accounting, 38(3), 243-266.
Roosenboom, P. (2007). How Do Underwriters Value Initial Public Offerings? An
EmpiricalAnalysis of the French IPO Market.Contemporary Accounting
Research, 24(4), 1217–1243.
Roosenboom, P. (2012). Valuing and pricing IPOs. Journal of Banking & Finance,
36(6), 1653-64.
Rosa, R., Velayuthen, G., & Walter, T. (2003). The sharemarket performance of
Australian venture capital-backed and non-venture capital-backed
IPOs. Pacific-Basin Finance Journal, 11(2), 197-218.
Ruud, J. S. (1993). Underwriter price support and the IPO underpricing puzzle.
Journal of Financial Economics, 34(2), 135-151.
Sahoo, S., & Rajib, P. (2010). After market pricing performance of initial public
offerings (IPOs): Indian IPO market 2002–2006. Vikalpa, 35(4), 27-44.
Sahoo, S., & Rajib, P. (2013). Comparable firm’s P/E multiple and IPO valuation: an
empirical investigation for Indian IPOs. Decision, 40(1-2), 27-46.
Samarakoon, L. P. (2010). The short-run underpricing of initial public offerings in
the Sri Lankan stock market. Journal of Multinational Financial
Management, 20(4), 197-213.
239
Schultz, P. H., & Zaman, M. A. (1994). After market support and underpricing of
initial public offerings. Journal of Financial Economics, 35(2), 199-219.
Schuster, J. A. (1996). Underpricing and Crises—IPO Underperformance in
Germany. Discussion Paper, London School of Economics.
Schreiner, A., & Spremann, K. (2007). Multiples and their valuation accuracy in
European equity markets. Available at SSRN#957352
Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under
conditions of risk. The journal of finance, 19(3), 425-442.
Sherman, A. E., (2005). Global trends in IPO methods: Book building versus auctions
with endogenous entry. Journal of Financial Economics. 78, 615–649
Shelly, & Singh, B. (2008). Oversubscription and IPO Underpricing: Evidence from
India. ICFAI Journal of Applied Finance, 14(12), 65-73.
Shleifer, A. & Vishny, R. (1997). A survey of corporate governance. Journal of
Finance, 52(2), 737-782.
Sohail, M., Bilal, H., Rukh, L., & Fatima, S. (2018). Determinants of Under Pricing of
Initial Public Offerings and its Long Run Performance in Pakistan. Pakistan
Journal of Social Sciences,38(2), 747-755
Sohail, A. (2015). IPO Performance in Short Run and Long Run: A Case of Karachi
Stock Exchange Listed Firms in Pakistan. International Journal of Economics
and Empirical Research (IJEER), 3(7), 320-326.
Sohail, M. K., & Nasr, M. (2007). Performance of initial public offerings in
Pakistan. International Review of Business Research Papers, 3(2), 420-441.
Sohail, M. K., & Raheman, A. (2009). Determinants of under-pricing of IPOs
regarding financial & non-financial firms in Pakistan. European Journal of
Economics, Finance and Administrative Sciences, 15, 62-73.
Sohail, M. K., & Raheman, A. (2010). Examining the Short-run IPOs Performance in
State of Economy: Normal, Boom & Recession. International Research
Journal of Finance and Economics, 35, 173-186
240
Stehle, R., Ehrhardt,O., & Przyborowsky, R. (2000). Long-run stock performance
ofGerman initial public offerings and seasoned equity issues. European
Financial Management, 6(2),173-196.
Su, C., Bangassa, K., & Brookfield, D. (2011). Long‐Run Performance of Chinese
Initial Public Offerings: Further Evidence. Asia‐Pacific Journal of Financial
Studies, 40(2), 285-316.
Su, D., & Fleisher, B. M. (1999). An empirical investigation of underpricing in
Chinese IPOs. Pacific-Basin Finance Journal, 7(2), 173-202.
Sullivan, M. J., & Angelo, A. U. (2001). The Underpricing of Initial Public Offerings
inthe Philippines from 1987 to 1997. CBE Working paper series.
Sumway & Tyler (1997). The delisting bias in CRSP data. Journal of finance, 50,
327-340
Thomadakis. S., Nounis C., & Gounopoulos D. (2012). Long-term Performance of
Greek IPOs. European Financial Management, 18 (1), 117-141.
Ting, Y. U., & Tse, Y. K. (2006). An empirical examination of IPO underpricing in
the Chinese A-share market. China economic review, 17(4), 363-382.
Tinic, S. (1988). Anatomy of initial public offerings of common stock. Journal of
Finance, 43(4),789-822.
Welch, I. (1989). Seasoned offerings, imitation costs, and the underpricing of initial
public offerings. Journal of Finance, 44(2), 421-429.
Westerholm, P. J. (2006). Industry clustering in Nordic initial public offering
markets. International Review of Finance, 6(2), 25-41.
White, H. L., (1980). A heteroscedastic–consistent covariance matrix estimator and a
direct test of heteroscedasticity. Econometrica, 48, 817–838
Wiser, R., Pickle, S., & Goldman, C. (1998). Renewable energy policy and electricity
restructuring: a California case study. Energy Policy, 26(6), 465-475.
241
Xia, N., Junyan, S., & Pei, G. (2013). Empirical study on initial public offering (IPO)
underpricing and long-run performance: Evidence from Chinas A-share
market. African Journal of Business Management, 7(11), 852-861.
Yar, M. M. S., & Javid, A. Y. (2014). Liquidity Benefits from Underpricing:
Evidence from Initial Public Offerings Listed at Karachi Stock Exchange (No.
2014: 101). Pakistan Institute of Development Economics.
Yoon, J. (2015). Multiples Valuation and Abnormal Returns. Honors Dissertation,
Duke University, Durham, North Carolina
Yu, & Tse. (2006). An Empirical Examination of IPO Underpricing in the Chinese A-
Share Market. China Economic Review, 17(4),363–82.
Yuhong, F. (2010). Pre-operating performances and IPOs after market returns, Allied
Academies International Conference, Academy of Accounting and Financial
Studies Proceedings, 15(1),1–28.
Zarafat, H., & Vejzagic, M. (2014). The Long-Term Performance of Initial Public
Offerings: Evidence from Bursa Malaysia. Journal of applied economics,
4(1), 42-51.
242
Appendix
243
Table A. 1: Characteristics of IPO Firms used in Sample
Sr.
No.
Listing
Year Symbol IPO Company Name Sector Auction Type
Offer
Price
1 2016 AWWAL Awwal Modaraba Modaraba Fixed Price 10.00
2 2016 HTL Hi-Tech Lubricants Ltd Oil & Gas “Bookbuilding 62.50
3 2016 TPLPL TPL Properties Ltd Property & Investment Bookbuilding 12.50
4 2016 LOADS Loads Limited Automobile & Electrical Bookbuilding 34.00
5 2015 SINDM Sindh Modaraba Modaraba Bookbuilding 10.00
6 2015 SYS Systems Limited Technology & Comm. Bookbuilding 40.00
7 2015 SPEL Synthetic Products Enterprises Ltd Engineering & Allied Bookbuilding 30.00
8 2015 MUGHAL Mughal Iron & Steel Industries Ltd Engineering & Allied Bookbuilding 34.00
9 2015 DCR Dolmen City "REIT" Property & Investment Bookbuilding 11.00
10 2015 ASC Al Shaheer Corporation Ltd Foods & Allied Bookbuilding 95.00
11 2015 ASTL Amerli Steels Ltd Engineering & Allied Bookbuilding 51.00
12 2014 EFERT Engro Fertilizer Ltd Fertilizer Bookbuilding 28.25
13 2014 AVN Avanceon Limited Technology & Comm. Bookbuilding 14.00
14 2014 HASCOL Hascol Petroleum Ltd Oil & Gas Bookbuilding” 56.50
15 2014 EPQL Engro Powergen Qadirpur Ltd Power Gen. & Dis. Fixed Price 30.02
16 2014 SPWL Saif Power Ltd Power Gen. & Dis. Bookbuilding 30.00
17 2013 LPL Lalpir Power Ltd Power Gen. & Dis. Bookbuilding 22.00
18 2012 NEXT Next Capital Ltd Investment Sec/Banks Bookbuilding 10.00
19 2012 TPL TPL Trakkar Ltd Automobile & Electrical Bookbuilding 10.00
20 2012 ASL Aisha Steel Mills Ltd Engineering & Allied Fixed Price 10.00
21 2011 ISL “International Steels Ltd Engineering & Allied Bookbuilding 14.06
22 2011 PKGP Pakgen Power Ltd Power Gen. & Dis. Fixed Price 19.00
23 2011 EFOODS Engro Foods Ltd Foods & Allied Fixed Price 25.00
24 2011 TDIL TPL Direct Insurance Ltd Insurance & Leasing Bookbuilding 10.00
25 2010 GGL Ghani Gases Ltd Chemicals Bookbuilding 14.00
26 2010 FATIMA Fatima Fertilizer Company Ltd Fertilizer Bookbuilding 13.50
27 2010 SMCPL Safe Mix Concrete Products Ltd Cement & Allied Fixed Price 12.50
28 2010 AGL Agritech Limited Chemicals Fixed Price 30.00
29 2010 AMTEX Amtex Limited Textile Bookbuilding 13.00
30 2010 WTCL Wateen Telecom Ltd Technology & Comm. “Fixed Price 10.00
31 2009 MDTL Media Times Ltd Technology & Comm. Fixed Price 10.00
32 2009 NPL Nishat Power Ltd Power Gen. & Dis. Fixed Price 10.00
33 2009 NCPL Nishat Chunian Power Ltd Power Gen. & Dis. Fixed Price 10.00
34 2008 AHBL Arif Habib Bank Ltd Commercial Banks Fixed Price 21.00
35 2008 IFSL Invest and Finance Securities Ltd” Investment Sec/Banks Fixed Price 10.00
36 2008 THCCL Thatta Cement Company Ltd Cement & Allied Fixed Price 22.50
37 2008 DEL Dawood Equities Ltd Investment Sec/Banks Fixed Price 17.50
38 2008 EPCL Engro Polymer & Chemicals Ltd Chemicals Fixed Price 18.00
39 2008 KASBSL KASB Securities Ltd Investment Sec/Banks Fixed Price 67.50
40 2008 FCIBL First Credit and Investment Bank Ltd Investment Sec/Banks Fixed Price 10.00
41 2008 AHIM Arif Habib Investment Management Ltd Investment Sec/Banks Fixed Price 125.00
42 2008 DOL Descon Oxychem Ltd Chemicals Fixed Price 10.00
43 2007 ARM Allied Rental Modaraba Modaraba Fixed Price 10.00
244
44 2007 AHL Arif Habib Limited Investment Sec/Banks Fixed Price 100.00
45 2007 HIRAT Hira Textile Mills Ltd Textile Fixed Price 12.50
46 2007 PACE PACE Pakistan Ltd Property & Investment Fixed Price 14.00
47 2007 JSIL JS Abamco Limited Investment Sec/Banks Fixed Price 65.00
48 2007 FLYNG Flying Cement Company Ltd Cement & Allied Fixed Price 14.00
49 2007 PASL Pervez Ahmed Securities Ltd Investment Sec/Banks Fixed Price 10.00
50 2007 SPL Sitara Peroxide Ltd Chemicals Fixed Price 10.00
51 2007 HBL Habib Bank Limited Commercial Banks Fixed Price 235.00
52 2007 DSL Dost Steels Limited Engineering & Allied Fixed Price 10.00
53 2006 BOK The Bank of Khyber Ltd Commercial Banks Fixed Price 15.00
54 2006 BIPL BankIslami Pakistan ltd Commercial Banks Fixed Price 10.00
55 2005 AMBL Network Microfinance Bank Ltd Commercial Banks Fixed Price 10.00
56 2005 IHFL International Housing Finance Ltd Commercial Banks Fixed Price 12.50
57 2005 JSCL Jahangir Siddiqui Capital Markets Ltd Investment Sec/Banks Fixed Price 52.50
58 2005 APL Attock Petroleum Limited Oil & Gas Fixed Price 57.75
59 2005 KAPCO Kot Addu Power Company Ltd Power Gen. & Dis. Fixed Price 30.00
60 2005 DFSML Dewan Farooque Spinning Mills Ltd Textile Fixed Price 10.00
61 2005 UBL United Bank Limited Commercial Banks Fixed Price 50.00
62 2005 ETNL Eye Television Network Ltd Technology & Comm. Fixed Price 10.00
63 2005 ZTL Zephyr Textiles Limited Textile Fixed Price 10.00
64 2005 CHENAB Chenab Limited Textile Fixed Price 18.00
65 2005 NETSOL Netsol Technologies Ltd Technology & Comm. Fixed Price 25.00
66 2005 WTL WorldCall Telecom Ltd Technology & Comm. Fixed Price 10.00
67 2005 DSIL D.S Industries Limited Textile Fixed Price 10.00
68 2005 STPL Siddiqsons Tin Plate Ltd Engineering & Allied Fixed Price 35.00
69 2004 OGDC Oil & Gas Development Company Ltd Oil & Gas Fixed Price 32.00
70 2004 WCBL WorldCall BroadBand Ltd Technology & Comm. Fixed Price 10.00
71 2004 MACFL MACPAC Films Ltd Automobile & Electrical Fixed Price 15.00
72 2004 CTTL Callmate Telips Telecom Ltd Technology & Comm. Fixed Price 10.00
73 2004 SNL Southern Networks Limited Technology & Comm. Fixed Price 10.00
74 2004 BAFL Bank Alfalah Limited Commercial Banks Fixed Price 30.00
75 2004 PPL Pakistan Petroleum Limited Oil & Gas Fixed Price 55.00
76 2004 FNEL First National Equities Limited Investment Sec/Banks Fixed Price 10.00
77 2003 ICL Ittehad Chemicals Limited Chemicals Fixed Price 10.00
78 2003 TRG TRG Pakistan Limited Technology & Comm. Fixed Price 10.00
79 2003 PICT Pakistan International Container Ltd Transporation & Comm. Fixed Price 10.00
80 2002 WCML WorldCall Multimedia Limited Technology & Comm. Fixed Price 10.00
81 2002 NBP National Bank of Pakistan Commercial Banks Fixed Price 10.00
82 2002 ACPL Attock Cement Pakistan Limited Cement & Allied Fixed Price 10.00
83 2002 BYCO Bosicor Pakistan Limited Oil & Gas Fixed Price 10.00
84 2001 BWCL Bestway Cement Limited Cement & Allied Fixed Price 10.00
85 2001 AHSL Arif Habib Securities Ltd Investment Sec/Banks Fixed Price 80.00
86 2000 WCPL WorldCall Payphone Limited Transporation & Comm. Fixed Price 15.00
87 2000 DFML Dewan Farooque Motors Ltd Automobile & Electrical Fixed Price 10.00
88 2000 MEBL Al Meezan Investment Bank Ltd Commercial Banks Fixed Price” 11.50
245
Table A. 2: The List of Non-Survivor IPO Firms During Sample Period
Sr.
No.
Listing
Year Symbol IPO Company Name Sector Action
1 2010 WTCL Wateen Telecom Ltd Technology & Comm. Delisted
2 2006 BIPL BankIslami Pakistan ltd Commercial Banks Merged
3 2005 CHENAB Chenab Limited Textile Delisted
4 2004 WCBL WorldCall BroadBand Ltd Technology & Comm. Merged
5 2004 CTTL Callmate Telips Telecom Ltd Technology & Comm. Delisted
6 2004 SNL Southern Networks Limited Technology & Comm. Delisted
7 2002 WCML WorldCall Multimedia Ltd Technology & Comm. Merged
8 2000 WCPL WorldCall Payphone Ltd Transporation & Comm. Merged
Table A. 3: The List of State Owned Enterprize (SOE) IPO Firms
Sr.
No.
Listing
Year Symbol IPO Company Name Sector
1 2015 SINDM Sindh Modaraba Modaraba
2 2007 HBL Habib Bank Limited “Commercial Banks
3 2006 BOK The Bank of Khyber Ltd Commercial Banks
4 2005 KAPCO Kot Addu Power Company Ltd Power Gen. & Dist.
5 2005 UBL United Bank Limited Commercial Banks”
6 2004 OGDC Oil & Gas Development Company Ltd Oil & Gas
7 2004 PPL Pakistan Petroleum Limited Oil & Gas
8 2002 NBP National Bank of Pakistan Commercial Banks
246
Table A. 4: Firms Post-IPO Cash and Stock Dividends (%) Announcements History
Sr.
No
Yea
r Company Name
Year t+1 Year t+2 Year t+3 Year t+4 Year t+5
Cas
h%
Stoc
k%
Cas
h%
Stoc
k%
Cas
h%
Stoc
k%
Cas
h%
Stoc
k%
Cas
h%
Stoc
k%
1 2016 Awwal Modaraba 2.00 0.00 Na na na na na na Na na
2 2016 Hi-Tech Lubricants Ltd 27.0 0.00 Na na na na na na Na Na
3 2016 TPL Properties Ltd 0.00 0.00 Na na na na na na Na Na
4 2016 Loads Limited 10.0 10.0 Na na na na na na Na Na
5 2015 Sindh Modaraba 1.60 0.00 3.50 0.00 na na na na Na Na
6 2015 Systems Limited 12.5 0.00 18.6 0.00 na na na na Na Na
7 2015 Synthetic Products Ent. . 10.0 0.00 15.0 0.00 na na na na Na Na
8 2015 Mughal Iron & Steel Ind. 5.00 15.0 30.0 0.00 na na na na Na Na
9 2015 Dolmen City "REIT" Ltd 0.80 0.00 10.4 0.00 na na na na Na Na
10 2015 Al Shaheer Corporation 0.00 0.00 0.00 50.0 na na na na Na Na
11 2015 Amerli Steels Ltd 0.00 0.00 20.0 0.00 na na na na Na Na
12 2014 Engro Fertilizer Ltd 30.0 0.00 60.0 0.00 70.0 0.00 na na Na Na
13 2014 Avanceon Limited 22.5 0.00 20.0 0.00 10.0 25.0 na na Na Na
14 2014 Hascol Petroleum Ltd 32.0 0.00 50.0 11.0 35.0 20.0 na na Na Na
15 2014 Engro Powergen Qadirpur 15.0 0.00 35.0 0.00 30.0 0.00 na na Na Na
16 2014 Saif Power Ltd 0.00 0.00 15.0 0.00 37.5 0.00 36.5 0.00 Na Na
17 2013 Lalpir Power Ltd 25.0 0.00 10.0 0.00 20.0 0.00 20.0 0.00 Na Na
18 2012 Next Capital Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
19 2012 TPL Trakkar Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.50 0.00
20 2012 Aisha Steel Mills Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
21 2011 International Steels Ltd 0.00 0.00 0.00 0.00 10.0 0.00 0.00 0.00 0.00 0.00
22 2011 Pakgen Power Ltd 65.0 0.00 30.0 0.00 25.0 0.00 10.0 0.00 20.0 0.00
23 2011 Engro Foods Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100 0.00
24 2011 TPL Direct Insurance Ltd 5.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
25 2010 Ghani Gases Ltd 0.00 0.00 0.00 0.00 0.00 0.00 5.00 25.0 0.00 0.00
26 2010 Fatima Fertilizer Co. Ltd 0.00 0.00 15.0 0.00 20.0 0.00 25.0 0.00 27.5 0.00
27 2010 Safe Mix Concrete Prod. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
28 2010 Agritech Limited 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
29 2010 Amtex Limited 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 30.0
30 2010 Wateen Telecom Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
31 2009 Media Times Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
32 2009 Nishat Power Ltd 0.00 0.00 0.00 0.00 0.00 0.00 20.0 0.00 30.0 0.00
33 2009 Nishat Chunian Power Ltd 0.00 0.00 20.0 0.00 35.0 0.00 60.0 0.00 65.0 0.00
34 2008 Arif Habib Bank Ltd 0.00 11.1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
35 2008 Invest and Finance Sec. 0.00 0.00 0.00 0.00 11.5 0.00 0.00 0.00 0.00 10.0
36 2008 Thatta Cement Company 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
37 2008 Dawood Equities Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
38 2008 Engro Polymer & Chem. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
247
39 2008 KASB Securities Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5.00 0.00
40 2008 First Credit and Inv. Bank 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
41 2008 Arif Habib Inv. Manag’t 25.0 80.0 0.00 0.00 0.00 20.0 15.0 0.00 22.5 0.00
42 2008 Descon Oxychem Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
43 2007 Allied Rental Modaraba 10.0 0.00 20.0 0.00 15.0 0.00 22.0 0.00 23.0 25.0
44 2007 Arif Habib Limited 100 10.0 25.0 25.0 15.0 25.0 0.00 20.0 0.00 0.00
45 2007 Hira Textile Mills Ltd 0.00 0.00 0.00 0.00 0.00 0.00 10.0 0.00 10.0 0.00
46 2007 PACE Pakistan Ltd 0.00 17.5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
47 2007 JS Abamco Limited 0.00 0.00 0.00 25.0 0.00 0.00 0.00 0.00 0.00 0.00
48 2007 Flying Cement Company 0.00 10.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
49 2007 Pervez Ahmed Securities 0.00 0.00 20.0 27.5 0.00 0.00 0.00 0.00 0.00 0.00
50 2007 Sitara Peroxide Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
51 2007 Habib Bank Limited 40.0 10.0 55.0 20.0 60.0 10.0 65.0 10.0 70.0 10.0
52 2007 Dost Steels Limited 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
53 2006 The Bank of Khyber Ltd 0.00 0.00 0.00 0.00 0.00 25.0 0.00 0.00 0.00 0.00
54 2006 BankIslami Pakistan Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
55 2005 Network Microfinance Bnk 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
56 2005 Interl Housing Finance Ltd 12.5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
57 2005 JS. Capital Markets Ltd 25.0 0.00 25.0 0.00 25.0 100 0.00 159.
7
0.00 243.7
8 58 2005 Attock Petroleum Limited 50.0 33.0 120 0.00 140 20.0 200 20.0 250 0.00
59 2005 Kot Addu Power Company 80.0 0.00 81.0 0.00 60.0 0.00 54.5 0.00 64.5 0.00
60 2005 Dewan Farooque Spinning. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
61 2005 United Bank Limited 25.0 25.0 30.0 25.0 30.0 25.0 25.0 10.0 25.0 10.0
62 2005 Eye Television Network 0.00 0.00 0.00 0.00 0.00 0.00 53.0 0.00 16.0 0.00
63 2005 Zephyr Textiles Limited 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
64 2005 Chenab Limited 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
65 2005 Netsol Technologies Ltd 0.00 0.00 0.00 0.00 0.00 37.0 10.0 40.0 0.00 0.00
66 2005 WorldCall Telecom Ltd 0.00 15.0 0.00 15.0 0.00 0.00 0.00 0.00 0.00 0.00
67 2005 D.S Industries Limited 0.00 0.00 0.00 0.00 10.0 0.00 0.00 100 0.00 0.00
68 2005 Siddiqsons Tin Plate Ltd 0.00 0.00 10.0 10.0 15.0 0.00 15.0 0.00 10.0 0.00
69 2004 Oil&Gas Development Co. 40.0 0.00 75.0 0.00 90.0 0.00 90.0 0.00 95.0 0.00
70 2004 WorldCall BroadBand Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
71 2004 MACPAC Films Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
72 2004 Callmate Telips Telecom 0.00 0.00 10.0 10.0 30.0 47.5 0.00 0.00 0.00 0.00
73 2004 Southern Networks Ltd 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
74 2004 Bank Alfalah Limited 0.00 25.0 12.0 33.0 0.00 30.0 15.0 23.0 0.00 12.5
75 2004 Pakistan Petroleum Ltd 55.0 0.00 90.0 0.00 110 10.0 155 10.0 130 20.0
76 2004 First National Equities Ltd 25.0 0.00 60.0 0.00 15.0 1500 0.00 0.00 0.00 0.00
77 2003 Ittehad Chemicals Limited 15.0 0.00 0.00 0.00 0.00 20.0 0.00 20.0 15.0 0.00
78 2003 TRG Pakistan Limited 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
79 2003 Pakistan Int’l Container 0.00 0.00 0.00 0.00 0.00 0.00 20.0 0.00 0.00 30.0
80 2002 WorldCall Multimedia Ltd 0.00 0.00 0.00 0.00 0.00 0.00 10.0 0.00 0.00 0.00
248
81 2002 National Bank of Pakistan 12.5 10.0 12.5 20.0 15.0 20.0 25.0 20.0 40.0 15.0
82 2002 Attock Cement Pakistan 15.0 10.0 10.0 0.00 12.5 0.00 12.5 0.00 50.0 0.00
83 2002 Bosicor Pakistan Limited 0.00 0.00 0.00 0.00 0.00 0.00 7.50 0.00 0.00 0.00
84 2001 Bestway Cement Limited 5.00 0.00 7.50 0.00 7.50 0.00 10.0 10.0 10.0 10.0
85 2001 Arif Habib Securities Ltd 50.0 0.00 50.0 20.0 100 33.3 150 150 100 50.0
86 2000 WorldCall Payphone Ltd 15.0 0.00 0.00 20.0 0.00 25.0 0.00 0.00 0.00 15.0
87 2000 Dewan Farooque Motors 0.00 0.00 0.00 0.00 10.0 0.00 10.0 5.00 10.0 0.00
88 2000 Al Meezan Inv. Bank Ltd 17.5 0.00 5.00 10.0 5.00 10.0 0.00 15.0 0.00 16.0
Note: ‘na’ - corporate payout announcements are pending. The value of Cash Dividends is w.r.t IPO
firm par value and Bonus Dividends are w.r.t total number of outstanding shares in the announcement
year.
249
Table A. 5: Cross-sectional Regressions of Bias of Valuation Methods
Model Bias DDM Bias DCF Bias Multiples Bias Fair_Value
Dep. Variable (1) (2) (3) (4)
Intercept 41.8111* -137.216 60.1619** 39.8287***
(2.134) (0.648) (4.404) (5.230)
Size -26.7512*** 14.405 -0.9427 -3.4499
(-4.321) (0.622) (-0.125) (-1.119)
Firm Age -23.8963 -2.6768
(-1.491) (-0.390)
Property, plant & Equip. -1.7184** -0.2489** 0.1184 -0.0147
(-2.428) (2.278) (0.484) (-0.169)
Operating Profitability 0.4778** 0.102 0.0427
(2.616) (0.839) (0.515)
Sales Growth 0.0249 -0.1162* -0.0376
(0.124) (-1.863) (-0.804)
Dividend Payout -0.2483 -0.2697*
(-1.171) (-1.819)
Technology 6.0638 -10.7649
(0.435) (-1.303)
Market Returns 1.7355** 0.7473** 0.3225
(2.963)
(2.772) (1.449)
Ex ante 13.1131 22.8946*
(0.624) (1.731)
Underwriter Reputation -33.6536** -5.9192
(-2.709) (-0.764)
Dilution Factor 0.87 0.1381
(1.502) (0.609)
Adj. R-Square 0.7301 0.3520 0.14104 0.13928
F-Statistic 6.3123 3.3093 1.94 1.089
Prob(F-Statistic) 0.0211 0.044 0.5073 0.3893
N 15 20 65 88
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
250
Table A. 6: Value Relevance Regressions (at IPO Offer prices)
Indep. Variable Parameter Adj. R2
(%) N Wald test
Intercept Slop
Dividend Discount Model 0.1410 1.0147** 63.60 15 0.1072***
(0.303) (3.062)
Discounted Cash Flow 0.2096 0.8871*** 47.82 20 3.0879*
(1.300) (7.479)
Multiples Valuation 0.0690 1.0223*** 84.12 65 14.8495***
(0.929) (18.64)
P/E Ratio 0.1477 0.9737*** 80.93 36 10.5429**
(1.218) (12.379)
P/B Ratio 0.1397 0.9532*** 86.31 51 10.7959**
(2.165) (23.216)
Fair Value Estimate 0.0866 1.0032*** 82.29 88 15.8657***
(1.234) (18.684)
251
Table A. 7: Cross-sectional Regressions of Accuracy of Valuation Methods (IPO Offer
Prices)
Model Accuracy
DDM
Accuracy
DCF
Accuracy
Multiples
Accuracy
Fair_Value
Dep. Variable (1) (2) (3) (4)
Intercept 41.8111* -137.216 65.4363*** -91.5432
(2.1951) (-0.6485) (4.771) (-1.306)
Total Assets -26.7511** 14.405 -6.4498* 14.0793**
(-4.3211) (0.622) (-1.696) (2.119)
Firm Age -14.9159 -11.4821
(-0.956) (-0.932)
Property, plant & Equip. -1.7184** -0.2489** 0.0108 -0.0333
(-2.4289) (-2.278) (0.056) (-0.179)
Operating Profitability 0.4779** 0.0452 0.0114
(2.616) (0.342) (0.154)
Sales Growth 0.0249 -0.1519** -0.0116
(0.124) (-2.564) (-0.295)
Dividend Payout -0.1117 -0.2331
(-0.530) (-1.622)
Technology 10.6343 -17.8955*
(0.752) (-1.898)
Market Returns 1.7355** 0.5742** 0.1409
(2.963)
(2.288) (0.837)
Ex ante 4.0154 17.5754
(0.366) (1.839)*
Underwriter Reputation -16.6048 -4.2789
(-1.225) (-0.459)
Dilution Factor 0.2100 0.3422
(0.485) (0.675)
Adj. R-Square 0.7301 0.3521 0.25390 0.13275
F-Statistic 6.3123** 3.309** 1.578 1.029
Prob(F-Statistic) 0.002 0.044 0.1339 0.4296
N 15 20 65 88
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
252
Table A. 8: Empirical findings of basic valuation models
Panel-A: Full IPO Sample
Variable Model 1 (OP/BV) Model 2 (FDCP/BV)
Intercept 1.2438*** 1.6468***
9.9956 10.5149
Earnings 0.8296 1.8377**
1.5347 2.1922
D 0.2872** 0.5315**
2.3081 2.1810
Dividends 2.6308*** 4.9231***
3.1365 3.3355
Und_Contract -0.0088 -0.4231**
-0.0541 -2.1796
Adj. R2 0.128073 0.267279
F-Statistic 2.937702** 7.295537***
Wald-test 152.5427*** 158.7339
N 86 86
Panel-B: Non-Privatization IPO Sample
Variable Model 1 (OP/BV) Model 2 (FDCP/BV)
Intercept 1.2482*** 1.6349***
10.0407 9.9907
Earnings 0.6247 1.2762
1.1639 1.4377
D 0.2131*** 0.3228*
2.7446 1.8140
Dividends 2.5270** 4.6549**
2.3069 2.3390
Und_Contract 0.0160 -0.2924
0.0892 -1.3803
Adj. R2 0.11484 0.205175
F-Statistic 2.30286** 4.58195***
Wald-test 149.9320*** 130.8001***
N 78 78
253
Table A. 9: Cross-sectional Analysis of Full Sample Valuation Models
Variable Model 1 (FDCP/BV) Model 2 (FDCP/BV)
Fundamental Factors
Intercept (BV/BV) -3.6541* -3.4947
(-2.7298) (-1.6487)
Earnings (EPS/BV) -0.7769* -0.7972*
(-2.8134) (-2.8796)
D -0.2852 -0.2936
(-0.9180) (-0.9498)
Dividends (DPS/BV) 3.8685** 3.8313**
(2.2135) (2.1882)
Risk Factors
Financial Leverage 0.0012 0.0009
(0.2467) (0.1873)
Capital Availability Risk 0.0016 0.0022
(0.2969) (0.4042)
Efficiency Risk 0.0074** 0.0077**
(2.1028) (2.1959)
Capacity Risk 0.0000 0.0002
(-0.0051) (0.0530)
Firm Beta 0.0449** 0.0453**
(2.6713) (2.6886)
Offer Size 0.5114** 0.4822**
(2.1420) (2.0114)
Signal Factors
Underwriter Reputation 0.0026 0.0103
(0.0229) (0.0877)
Firm Age -0.0009 -0.0012
(-0.1215) (-0.1574)
Percentage of Shares
Offered 0.0147 0.0151
(1.1870) (1.2174)
Privatization - 0.2506
(0.6185)
Adj. R-square 0.39576 0.39814
F-Statistic 3.87531*** 3.56207***
N 86 86
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
254
Table A. 10: Equally and Value-weighted Monthly Returns using BHAR & CAR
Equally-Weighted Value-Weighted
Month BHARt CARt BHARt CARt
1 -0.0216 -0.0551 -0.0381 -0.0412
2 -0.0259 -0.0731 -0.0772 -0.0878
3 -0.0115 -0.0399 -0.0606 -0.0569
4 0.0077 -0.0226 0.0039 0.0229
5 -0.0022 -0.0411 -0.0862 -0.0422
6 -0.0281 -0.0534 -0.0907 0.0199
7 -0.0793 -0.0982 -0.1683 -0.0458
8 -0.1095 -0.1177 -0.1475 -0.0346
9 -0.1019 -0.1101 -0.1493 -0.0456
10 -0.0788 -0.1226 -0.0937 -0.0382
11 -0.0909 -0.1134 -0.0890 -0.0422
12 -0.0771 -0.1305 -0.1028 -0.0622
13 -0.0892 -0.1261 -0.1089 -0.0648
14 -0.1130 -0.1505 -0.1600 -0.1288
15 -0.1336 -0.1395 -0.0873 -0.0775
16 -0.1063 -0.1262 -0.0146 0.0550
17 -0.1148 -0.1515 -0.0039 0.0700
18 -0.1353 -0.1503 -0.0750 0.0349
19 -0.1383 -0.1515 -0.1096 -0.0013
20 -0.1444 -0.1479 -0.1021 -0.0128
21 -0.1363 -0.1656 -0.1038 -0.0157
22 -0.1503 -0.1798 -0.1249 -0.0104
23 -0.0916 -0.1551 -0.1297 0.0024
24 -0.1721 -0.1833 -0.1745 -0.0154
25 -0.2048 -0.1902 -0.1992 -0.0418
26 -0.1759 -0.1739 -0.2160 -0.0485
27 -0.1826 -0.2022 -0.2258 -0.0704
28 -0.1741 -0.2012 -0.2351 -0.0769
29 -0.1154 -0.1949 -0.2941 -0.1133
30 -0.0557 -0.2275 -0.3074 -0.1468
31 -0.1550 -0.2439 -0.3541 -0.1939
32 -0.1910 -0.2481 -0.3892 -0.1806
33 -0.2335 -0.2611 -0.4279 -0.2061
34 -0.1733 -0.2438 -0.4140 -0.2234
35 -0.1427 -0.2642 -0.4427 -0.2508
36 -0.2352 -0.2462 -0.4338 -0.2583
37 -0.2575 -0.2274 -0.4677 -0.2657
38 -0.2476 -0.2352 -0.4634 -0.2515
39 -0.3016 -0.2504 -0.5189 -0.2808
40 -0.2860 -0.2744 -0.4930 -0.2812
41 -0.3639 -0.3117 -0.4700 -0.3196
42 -0.3722 -0.3158 -0.4869 -0.3185
255
43 -0.3595 -0.2869 -0.4713 -0.2893
44 -0.4349 -0.2870 -0.5203 -0.2931
45 -0.3357 -0.2676 -0.4397 -0.2784
46 -0.3912 -0.2774 -0.4197 -0.2477
47 -0.4130 -0.3038 -0.3811 -0.2499
48 -0.3388 -0.3188 -0.3557 -0.2501
49 -0.4190 -0.3281 -0.4055 -0.2553
50 -0.4532 -0.3073 -0.4112 -0.2535
51 -0.5178 -0.3319 -0.4527 -0.2905
52 -0.5395 -0.3430 -0.3903 -0.2091
53 -0.6070 -0.3628 -0.3951 -0.2398
54 -0.6716 -0.3899 -0.4177 -0.2571
55 -0.5888 -0.3551 -0.4443 -0.2986
56 -0.5646 -0.3817 -0.4504 -0.3045
57 -0.5286 -0.3392 -0.4642 -0.2867
58 -0.6101 -0.3340 -0.5027 -0.3063
59 -0.5965 -0.2878 -0.4995 -0.2825
60 -0.6522 -0.2937 -0.5843 -0.3206
Figure A. 1: Equal- and Value-weighted Monthly returns using BHAR and CAR
-0.8000
-0.7000
-0.6000
-0.5000
-0.4000
-0.3000
-0.2000
-0.1000
0.0000
0.1000
0.2000
1 3 5 7 9 11131517192123252729313335373941434547495153555759
EWBHAR
VWBHAR
EWCAR
VWCAR
256
Figure A. 2: Year-wise Long run returns for year 1 using BHAR and CAR
Figure A. 3: Year-wise Long run returns for year 2 using BHAR and CAR
Figure A. 4: Year-wise Long run returns for year 3 using BHAR and CAR
-1.0000
-0.5000
0.0000
0.5000
1.0000
1.5000
BHAR1
CAR1
-2.0000
-1.5000
-1.0000
-0.5000
0.0000
0.5000
1.0000
BHAR2
CAR2
-2.0000
-1.5000
-1.0000
-0.5000
0.0000
0.5000
1.0000
1.5000
2.0000
2.5000
3.0000
BHAR3
CAR3
257
Figure A. 5: Year-wise Long run returns for year 4 using BHAR and CAR
Figure A. 6: Year-wise Long run returns for year 5 using BHAR and CAR
-3.0000
-2.0000
-1.0000
0.0000
1.0000
2.0000
3.0000
4.0000
5.0000
6.0000
7.0000
8.0000
BHAR4
CAR4
-4.0000
-3.0000
-2.0000
-1.0000
0.0000
1.0000
2.0000
3.0000
4.0000
5.0000
6.0000
7.0000
BHAR5
CAR5
258
Figure A. 7: Sector-wise Long run returns for year 1, 3 & 5 using CAR
Figure A. 8: Sector-wise Long run returns for year 1, 3 & 5 using BHAR
-3.0000
-2.0000
-1.0000
0.0000
1.0000
2.0000
3.0000
4.0000
5.0000
BHAR1
BHAR3
BHAR5
-2.0000
-1.5000
-1.0000
-0.5000
0.0000
0.5000
1.0000
1.5000
2.0000
CAR1
CAR3
CAR5
259
Table A. 11: Correlation Matrix of Variables used in LRR Models
Variable CAR1Y CAR3Y
CAR
5Y
BV/
OP
EPS/
OP Fin Lev Captl Rsk Eff Rsk Cpcty Rsk Firm Beta
Offer
Size Und Rep Firm Age POS Priv IERs
CAR3Y 0.563
5.41***
CAR5Y 0.489 0.815
4.45*** 11.17***
BV/ OP 0.044 0.240 0.239
0.35 1.96** 1.95**
EPS/ OP 0.340 0.493 0.403 0.455
2.87*** 4.51*** 3.5*** 4.06**
Fin Lev 0.033 0.005 0.064 0.180 -0.101
0.26 0.04 0.51 1.45 -0.81
Captl
Rsk -0.199 -0.219 -0.198 -0.042 -0.364 0.130
-1.61 -1.78* -1.60 -0.33 -3.11** 1.04
Eff Rsk -0.106 -0.168 -0.114 -0.178 -0.510 0.142 0.249
-0.84 -1.35 -0.91 -1.43 -4.71** 1.14 2.04**
Cpcty
Rsk 0.080 -0.027 -0.078 -0.186 -0.277 -0.084 0.246 0.113
0.63 -0.21 -0.62 -1.50 -2.29** -0.67 2.01** 0.90
Firm
Bbta 0.354 0.107 0.090 -0.266 0.236 0.043 -0.281 -0.181 -0.148
3.01*** 0.86 0.71 -2.1** *1.93 0.34 -2.32** -1.46 -1.19
Offer
Size 0.055 -0.132 -0.131 -0.213 0.114 0.184 -0.141 -0.169 -0.296 0.395
0.43 -1.05 -1.05 -1.73* 0.91 1.48 -1.13 -1.36 -2.46** 3.41***
Und
Rep -0.205 -0.076 -0.023 -0.142 -0.117 -0.150 0.008 0.106 0.201 -0.057 -0.161
-1.66 -0.61 -0.18 -1.14 -0.93 -1.20 0.07 0.85 1.62 -0.45 -1.29
260
Firm
Age 0.033 -0.034 -0.066 -0.068 0.105 0.101 -0.261 -0.116 -0.588 0.329 0.488 -0.208
0.266 -0.276 -0.525 -0.541 0.842 0.807 -2.14** -0.93 -5.77*** 2.76*** 4.44*** -1.69*
POS -0.061 -0.034 -0.034 -0.082 -0.066 -0.179 0.134 0.152 0.383 -0.055 -0.413 0.213 -0.390
-0.48 -0.27 -0.27 -0.65 -0.52 -1.44 1.07 1.22 3.29*** -0.44 -3.60*** 1.73* -3.36***
Priv 0.039 0.077 0.094 0.175 0.302 0.219 -0.318 -0.257 -0.403 0.346 0.577 -0.324 0.520 -0.319
0.31 0.61 0.75 1.41 2.51** 1.78* -2.67*** -2.11** -3.50*** 2.93*** 5.62*** -2.71*** 4.83*** -2.67**
IERs 0.127 -0.060 -0.024 -0.024 0.057 0.165 0.032 0.172 0.016 0.354 0.016 -0.185 -0.089 0.146 0.157
1.023 -0.478 -0.195 -0.193 0.453 1.333 0.256 1.393 0.128 3.008*** 0.134 -1.500 -0.711 1.172 1.270
Resi
-0.037 -0.060 -0.005 -0.278 -0.189 -0.073 0.035 0.003 0.048 -0.014 -0.031 -0.010 0.002 -0.024
-
0.151
-
0.112
-0.295 -0.479 -0.046 -2.3** -1.52 -0.58 0.282 0.023 0.381 -0.115 -0.249 -0.083 0.018 -0.198 -1.21 -0.89
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
261
Table A. 12: Regression Analysis of LRR Models using Full Sample
Variable CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
Fundamental Factors
Intercept 0.2316 1.9072 4.4199 2.9799 2.5394
(0.116) (0.533) (1.375) (0.945) (0.414)
Book Value (BV/OP) -0.3108*** -0.5179** -0.3160** -0.2668** -0.0820
(-2.99) (-3.213) (-2.472) (-2.411) (-0.515)
Earnings (EPS/OP) 1.5572** 4.1493*** 4.0465*** 3.9712*** 2.3118***
(2.464) (3.873) (2.914) (3.403) 2.331)
Risk Factors
Financial Leverage 0.0035** -0.0033 -0.0022 0.0013 -0.0064
(2.309) (-0.947) (-0.654) (0.191) (-1.126)
Capital Availability Risk 0.0006 0.0184** 0.0141** 0.0154*** 0.0276**
(0.147) (2.169) (2.373) (2.708) (2.211)
Efficiency Risk 0.0023 0.0014 -0.0007 -0.0055 -0.0006
(0.584) (0.222) (-0.182) (-1.070) (-0.072)
Capacity Risk 0.0053* 0.0080* 0.0023 0.0032 0.0041
(1.711) (1.781) (0.406) (0.648) (0.474)
Firm Beta 0.0162** -0.0107 -0.0174* -0.0246*** -0.0411***
(2.085) (-1.072) (-1.875) (-3.46) (-3.13)
Offer Size 0.0114 -0.3410 -0.6375* -0.4888 -0.5896
(0.050) (-0.801) (-1.707) (-1.291) (-0.800)
Signal Factors
Underwriter Reputation -0.6878** -0.0911 -0.4758 -0.1957 0.0897
(-2.772) (-0.241) (-1.136) (-0.574) (0.319)
Firm Age -0.0054 0.0134 0.0101 0.0104 0.0105
(-0.682) (1.361) (0.665) (0.738) (0.390)
Percentage of Shares Offered -0.0092 -0.0468** -0.0079 -0.0114 -0.0430**
(-0.805) (-2.065) (-0.951) (-0.918) (-2.015)
Initial Excess Returns -0.0041** -0.0018 -0.0037 0.0000 -0.0034
(-2.521) (-0.849) (-1.100) (0.001) (-0.636)
Residuals -0.2115** -0.4873** -0.1355 -0.0006 -0.0032
(-2.032) (-2.875) (-0.920) (-0.004) (-0.014)
Adj. R-square 0.36543 0.41774 0.37015 0.43021 0.32478
F-Statistic 2.171** 2.704*** 2.215** 2.845*** 1.813**
N 65 65 65 65 65
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
262
Table A. 13: Regression Analysis of LRR Models using Non-PIPO Sample
Variable CAR1Y CAR2Y CAR3Y CAR4Y CAR5Y
Fundamental Factors
Intercept 2.0607 4.3372 8.8667 8.5533 10.4274*
(0.938) (0.953) (1.323) (1.268) (1.695)
Book Value (BV/OP) -0.6195** -0.7017* -0.4252 -0.6007 -0.0897
(-2.485) (-1.784) (-0.605) (-0.898) (-0.357)
Earnings (EPS/OP) 1.7561** 4.4310*** 6.2795*** 6.7107*** 3.6182***
(2.409) (3.188) (3.206) (3.766) (4.136)
Risk Factors
Financial Leverage 0.0035 -0.0054 0.0011 0.0113 0.0059
(1.556) (-0.727) (0.113) (0.748) (0.800)
Capital Availability Risk 0.0015 0.0282** 0.0427*** 0.0470*** 0.0356**
(0.315) (2.621) (3.839) (3.873) (2.634)
Efficiency Risk -0.0010 -0.0091 -0.0129* -0.0144* -0.0179**
(-0.333) (-1.497) (-1.781) (-1.801) (-2.553)
Capacity Risk 0.0052 0.0136** 0.0165* 0.0196** 0.0102
(1.540) (2.653) (1.728) (2.269) (1.280)
Firm Beta 0.0143 -0.0029 -0.0212 -0.0389** -0.0433***
(1.351) (-0.204) (-0.999) (-2.165) (-2.81)
Offer Size -0.1806 -0.6703 -1.4913* -1.5802* -1.4772**
(-0.714) (-1.274) (-1.93) (-1.993) (-2.071)
Signal Factors
Underwriter Reputation -0.8780*** -0.0559 -0.6773 -0.1404 0.2827
(-3.28) (-0.105) (-0.979) (-0.211) (1.366)
Firm Age -0.0009 0.0218 0.0289 0.0149 -0.0227
(-0.092) (1.350) (0.814) (0.513) (-1.046)
Percentage of Shares Offered -0.0064 -0.0534** -0.0182 -0.0195* -0.0476**
(-0.547) (-2.111) (-1.327) (-1.682) (-2.315)
Initial Excess Returns -0.0049*** -0.0014 -0.0086* -0.0050 -0.0030
(-3.10) (-0.591) (-1.769) (-0.834) (-0.602)
Residuals -0.2405** -0.4971** -0.2568 0.0997 0.2657
(-2.53) (-2.484) (-1.026) (0.446) (1.408)
Adj. R-square 0.46106 0.42292 0.45913 0.45985 0.52469
F-Statistic 2.763*** 2.367** 2.742*** 2.751*** 3.566***
N 58 58 58 58 58
t-statistics using white(1980) heteroscedastic standard errors are within parentheses. ***significant at
the 1% level, **significant at the 5% level and *significant at the 10% level.
263
Table A. 14: Descriptive Statistics of Variables Used in Value-weighted Analysis
Variable Mean Min Percentiles
Max SD Obs. 25th 50th 75th
VWIPO_Rf 0.0067 -0.8236 -0.0818 -0.0105 0.0686 9.3853 0.2340 3,900
Rm_Rf 0.0071 -0.3853 -0.0180 0.0067 0.0516 0.3074 0.0780 3,900
SMBFF3F 0.0071 -1.5213 -0.0223 0.0086 0.0400 0.8702 0.0713 3,900
SMBFF5F 0.0071 -1.5213 -0.0222 0.0083 0.0390 0.8702 0.0679 3,900
HML 0.0062 -0.7036 -0.0235 0.0003 0.0304 0.8916 0.0619 3,900
RMW 0.0035 -0.3553 -0.0221 0.0044 0.0322 0.4820 0.0489 3,900
CMA -0.0016 -0.3231 -0.0265 -0.0013 0.0228 0.6573 0.0499 3,900
264
Table A. 15: Correlation matrix of variable used in value-weighted analysis
IPO_RF RM_RF SMBFF3F SMBFF5F HML CMA RMW
IPO_RF 1
RM_RF 0.3216*** 1
21.211
SMBFF3F -0.151*** -0.4907*** 1
-9.536 -35.160
SMBFF5F -0.1472*** -0.5006*** 0.9717*** 1
-9.295 -36.105 257.102
HML -0.0061 -0.0571*** 0.1159*** 0.1743*** 1
-0.382 -3.574 7.291 11.052
RMW -0.0555*** -0.1892*** 0.0862*** 0.0573*** -0.2714*** 1
-3.473 -12.032 5.407 3.584 -17.605
CMA 0.0618*** 0.0001 0.0153 -0.0097 -0.1491*** 0.2893*** 1
3.871 0.007 0.956 -0.608 -9.417 18.871
***significant at the 1% level, **significant at the 5% level and *significant at the 10% level.
265
Table A. 16: Year-wise New-Listings and De-Listings in PSX
Year De-Listings New Listings PSX Listed Firms
1997 6 4 782
1998 11 1 779
1999 8 0 769
2000 6 3 762
2001 19 4 759
2002 40 4 725
2003 16 6 706
2004 57 17 666
2005 18 19 659
2006 19 9 658
2007 11 14 656
2008 11 10 656
2009 6 4 651
2010 13 6 651
2011 10 4 639
2012 69 4 573
2013 10 4 560
2014 9 6 557
2015 11 8 554
2016 0 4 558
Total 350 131 Source: PSX DataStream
Figure A. 9: The cumulative effect of New-Listings & De-Listings in PSX
0
50
100
150
200
250
300
350
400
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
De-Listings
New Listings