Valuation and Underpricing of Initial Public Offerings: Evidence from Germany in Different Market Cycles [1 ed.] 9783896445834, 9783896735836

Citation preview

Valuation and Underpricing of Initial Public Offerings

Studienreihe der Stiftung Kreditwirtschaft an der Universität Hohenheim Herausgeber: Prof. Dr. Joh. Heinr. v. Stein

Band 48

Susanna Holzschneider

Valuation and Underpricing of Initial Public Offerings Evidence from Germany in Different Market Cycles

Verlag Wissenschaft & Praxis

Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar.

ISBN 978-3-89673-583-6 © Verlag Wissenschaft & Praxis Dr. Brauner GmbH 2011 D-75447 Sternenfels, Nußbaumweg 6 Tel. +49 7045 930093 Fax +49 7045 930094 [email protected] www.verlagwp.de

Alle Rechte vorbehalten Das Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar. Das gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronischen Systemen. Printed in Germany

VORWORT

5

Vorwort Die Dissertation „Valuation and Underpricing of Initial Public Offerings – Evidence from Germany in Different Market Cycles“ ist im Rahmen meiner Tätigkeit am Lehrstuhl für Bankwirtschaft und Finanzdienstleistung der Universität Hohenheim und in den zwei Semestern als Visiting PhD an der Stern School of Business der New York University entstanden. Die Dissertation wurde von der Fakultät für Wirtschafts- und Sozialwissenschaften der Universität Hohenheim im November 2010 angenommen. Ich möchte mich bei Herrn Prof. Dr. Burghof und den Kolleginnen und Kollegen des Lehrstuhls für Bankwirtschaft für die Betreuung und Unterstützung sowie für die Aufnahme der Dissertation in die Studienreihe der Stiftung Kreditwirtschaft bedanken. Besonderer Dank geht an meine Eltern, die mich immer unterstützt haben und mir jederzeit mit Rat und Tat zur Seite standen. München, 22. Dezember 2010 Susanna Holzschneider

TABLE OF CONTENT

7

Table of Contents Table of Contents ...................................................................................................... 7  Tables ..................................................................................................................... 11 Figures ..................................................................................................................... 13 Abbreviations .......................................................................................................... 14  Introduction ............................................................................................................. 17  Chapter I: Theory and Evidence of IPO Underpricing ........................................... 21  I 

Introduction ................................................................................................... 21 

II  Theory and Evidence Based on Asymmetric Information Distribution ....... 22  II.1   Information Asymmetries between Issuer and Investor ....................... 22  II.1.1 Signalling Theory............................................................................ 22  II.1.2 Certification of Quality ................................................................... 25  II.2   Information Asymmetries between Investors....................................... 29  II.2.1 The Winner’s Curse ........................................................................ 29  II.2.2 Information Revelation ................................................................... 32  II.3   Information Asymmetries between Underwriter and Issuer ................ 37  III   Theory and Evidence Based on Symmetric nformation Distribution .......... 38  III.1  Underwriter Price Support .................................................................... 38  III.2  Litigation Risk ...................................................................................... 39  III.3  Company’s Ownership Structure ......................................................... 41  III.4  Behavioral Finance ............................................................................... 43  III.5  Information Momentum ....................................................................... 45  IV   Discussion ..................................................................................................... 46  V  Conclusion .................................................................................................... 49  Chapter II: How Do Pre-IPO Shareholders Determine Underpricing? .................. 51  I 

Introduction ................................................................................................... 51 

II  Related Literature and Development of Hypothesis .................................... 52  III  Research Design ........................................................................................... 56  III.1  Sample Selection and Data Sources ..................................................... 56 

TABLE OF CONTENT

8

III.2  Definition of Variables ......................................................................... 57  III.3  Definition of Hot and Cold Periods ...................................................... 63  IV  Empirical Results .......................................................................................... 63  IV.1  Firm and Transaction Characteristics ................................................... 63  IV.2  Pre-IPO Ownership Characteristics ...................................................... 69  IV.3  Regression Analysis ............................................................................. 74  IV.3.1 Variables Explaining IPO Underpricing ....................................... 74  IV.3.2 Determinants of IPO Underpricing in 1997-2001 and 2002-2007 ..................................................................................... 79  IV.3.3 Pre-IPO Ownership and Underpricing in Different Market Phases ........................................................................................... 83  IV.3.4 Pre-IPO Ownership and Underpricing with Positive Investor’s Information ................................................................................... 88  V  Conclusion .................................................................................................... 96  Chapter III: Do “Herding” Effects on Firm Multiples Determine IPO Valuation? ..................................................................................................... 97  I 

Introduction ................................................................................................... 97 

II  Related Literature and Development of Hypotheses .................................... 98  III  Research Design ......................................................................................... 102  III.1  Sample Selection and Data Sources ................................................... 102  III.2  Methodology and Definition of Variables.......................................... 103  IV  Empirical Results ........................................................................................ 108  IV.1  Descriptive Statistics of Firm and Market Characteristics ................. 108  IV.2  IPO’s and Comparable Firm Multiples .............................................. 114  IV.3  IPO Valuation ..................................................................................... 124  IV.3.1 Regression Estimates on IPO Valuation ..................................... 124  IV.3.2 IPO Valuation in Hot and Cold Markets ..................................... 127  IV.3.3 Information Asymmetries and IPO Valuation ............................ 132  V  Conclusion .................................................................................................. 136  Conclusion ............................................................................................................ 139  References ............................................................................................................. 141  Appendix: Chapter I .............................................................................................. 153 

TABLE OF CONTENT

9

AI  Information Asymmetries between Issuer and Investor ........................... 153  A I.1  Signalling Theory ............................................................................... 153  A I.2  Certification of Quality ....................................................................... 156  AII   Information Asymmetries between Investors ........................................... 160  A II.1 The Winner’s Curse ............................................................................ 160  A II.2 Information Revelation ....................................................................... 164  AIII  Information Asymmetries between Issuer and Investor ........................... 166  AIV  Symmetric Information ............................................................................. 166  Appendix: Chapter II ............................................................................................ 173  Appendix: Chapter III ........................................................................................... 175

TABLES

11

Tables Chapter II: Table I

Defintion of Variables ...................................................... 62

Chapter II: Table II

Firm and Offer Characteristics ......................................... 67

Chapter II: Table III

Firm Characteristics in Hot and Cold Markets ................ 68

Chapter II: Table IV

Ownership Structure ......................................................... 71

Chapter II: Table V

Ownership Structure in Hot and Cold Markets ................ 73

Chapter II: Table VI

Regression Models on Underpricing (1) .......................... 76

Chapter II: Table VII

Regression Models on Underpricing (2) .......................... 78

Chapter II: Table VIII Regression Models on Underpricing: 1997-2001, 2002-2007......................................................................... 81 Chapter II: Table IX

Regression Models on Underpricing: Hot Time Period .. 83

Chapter II: Table X

Regression Models on Underpricing: Hot IPO Underpricing..................................................................... 85

Chapter II: Table XI

Regression Models on Underpricing: Hot IPO Volume .. 87

Chapter II: Table XII

Regression Models on Underpricing: Price Level ........... 91

Chapter II: Table XIII Regression Models on Underpricing: IPO Underpricing and Price Level ................................................................. 94 Chapter II: Table XIV Regression Models on Underpricing: IPO Volume and Price Level........................................................................ 95 Chapter III: Table I

Definition of Variables ................................................... 108

Chapter III: Table II

Firm Characteristics ....................................................... 110

Chapter III: Table III

Offer Characteristics ...................................................... 111

Chapter III: Table IV

Macroeconomic Conditions ........................................... 113

Chapter III: Table V

Regression Models on Market Value ............................. 125

Chapter III: Table VI

Regression Models on Market Value: Cold Market ...... 128

Chapter III: Table VII Regression Models on Market Value: Hot Market ........ 131 Chapter III: Table VIII Regression Models on Market Value: Controlling for Asymmetric Information ................................................ 135

12

TABLES

Appendix II: Table I

Underwriter Activity ....................................................... 173

Appendix III: Table I

Regression Models on Market Value (Hot/Cold) ........... 177

FIGURES

13

Figures Chapter II: Figure I

IPOs per Month between 1997-2007 ............................... 64

Chapter II: Figure II

Inital Returns per Month between 1997-2007 ................. 65

Chapter III: Figure I

MB Ratios of IPOs ......................................................... 115

Chapter III: Figure II

MB Ratios of Technology IPOs ..................................... 115

Chapter III: Figure III MB Ratios of Industry IPOs ........................................... 116 Chapter III: Figure IV PE Ratios of IPOs ........................................................... 117 Chapter III: Figure V

PE Ratios of Technology IPOs ...................................... 118

Chapter III: Figure VI PE Ratios of Industry IPOs ............................................ 118 Chapter III: Figure VII IPOs per Year and Industry ............................................ 119 Chapter III: Figure VIII Cross-Correlation of MB Ratios of IPOs and Market ... 120 Chapter III: Figure IX Cross-Correlation of PE Ratios of IPOs and Market ..... 121 Chapter III: Figure X

Autocorrelation of MB Ratios of IPOs .......................... 122

Chapter III: Figure XI Autocorrelation of PE Ratios of IPOs ............................ 123 Appendix III: Figure I Auto/ Cross-Correlation of MB/PE Ratios of all IPOs .. 175 Appendix III: Figure II Auto/ Cross-Correlation of MB/PE Ratios of Technology IPOs ............................................................ 176 Appendix III: Figure III Auto/ Cross-Correlation of MB/PE Ratios of Industry IPOs ................................................................................ 176

ABBREVIATIONS

14

Abbreviations BVK:

Bundesverband Deutscher Kapitalbeteiligungsgesellschaften German Private Equity and Venture Capital Association e.V.

CAPEX:

capital expenditure

CDAX:

Composite DAX (Deutscher Aktienindex)

CEO:

chief executives officer

CRSP:

Center for Research in Security Prices

DCF:

discounted cash flow

DDM:

dividend discount model

EBITDA:

earnings before interest, taxes, depreciation and amortization

e.g.:

for example, abbreviation of Latin “exempli gratia”

EPS:

earnings per share

et. seq.:

and the following, abbreviation for Latin “et sequens”

et. seqq.:

and those that follow, abbreviation for Latin “et sequentia”

etc.:

and so forth, abbreviation for Latin “et cetera”

EVCA:

European Venture Capital and Private Equity Association

FAZ:

Frankfurter Allgemeine Zeitung

ICB:

Industry Classification Benchmark

IPO:

initial public offering

M&A:

merger and acquisition

MB:

market-to-book

Mio:

million

MLP:

master limited partnership

NASDAQ: National Association of Securities Dealers Automated Quotations NEMAX:

performance index of “Neuer Markt”

NVCA:

National Venture Capital Association

OLS:

ordinary least square

PE:

price-earnings

ABBREVIATIONS

pp.:

Pages

R&D:

research and development

ROE:

return on equity

SEO:

seasoned equity offering

S&P:

Standard and Poor

USA:

United States of America

VC:

venture capital

VCs:

venture capitalists

15

INTRODUCTION

17

Introduction The valuation of firms going public is a very complex topic and one often discussed in corporate finance literature. When firms decide to complete an initial public offering (IPO), to raise new equity by issuing shares on a public stock market, their shares have to be priced to allow potential investors to be found. The valuation of a private firm and its shares is quite difficult, because the stock market value, normally the best indicator of expected firm’s growth and profitability in the eyes of investors, is not available. In almost every stock market across several countries the phenomenon of “underpricing” can be seen in the process of going public.1 Share prices in the secondary market on or after the first day of trading are higher than the initial offering price. These shares are allocated to investors at considerably lower prices than could normally be attained. This dissertation investigates the puzzle of IPO valuation and underpricing, focusing on the German stock exchange and newly-listed firms, in three chapters. Some explanations of initial returns after the first trading day of stocks have been theoretically explored and related empirical investigations have confirmed some theories and contradicted others. Previous research papers have mainly considered the USA stock market and have investigated several explanations based on the given market environment. The German stock exchange, however, can be described as less developed and less liquid than the USA’s. The number of completed IPOs is considerably lower, because German firms rely less on the market for public equity and bank financing is generally very important. There has been a distinct historical development of the institutional environment in both countries. Although interesting differences in stock market and financing systems exist, recent literature about IPO underpricing does not include any direct comparison between offering prices in the USA and in Germany. The impact of institutional factors and the role of investment banks on the phenomenon of IPO underpricing are especially interesting to consider. Therefore, chapter I (Theory and Evidence of IPO Underpricing) of this dissertation focuses on the existing literature, to highlight differences or similarities in IPO underpricing. Overall, the existing theories and empirical findings do not allow consideration of one of these explanations in isolation, and therefore the effects of institutional environments are difficult to determine. Over time, some changes in the explanatory factors have been recognised. However, the empirical results for the German IPO market mainly consider the period of the Neuer Markt, which was the most important stock segment for small1

Underpricing is measured as the difference between the shares’ closing price after the first trading day and IPO offer price divided by the offer price.

18

INTRODUCTION

and medium-sized companies during the dot-com phase of 1997 and 2001. These years were characterised by enormous stock price increases, especially in the technology and internet sectors. IPO volume increased exceptionally during these years and provided a large sample for empirical investigation into this research topic. Some literature suggests that those very high offer prices were even higher than the “true value” of the firms concerned, although prices of newly issued shares increased rapidly after the first day of trading. Germany saw average initial returns of 31% from 1997 to 2001. The market phases of high IPO volume as well as high underpricing, or rather initial returns, are described as “hot” market periods. A “cold” period with oppositional characteristics often follows, with saturated demand for new stocks. Also, in Germany, hardly any firm decided to complete an initial public offering after the bursting of the technology bubble. After 2004, the market started to recover and IPO volume increased again, although the enormous average levels of initial returns seen previously have not since been reached. These market cycles make drawing a conclusion about the determinants of IPO underpricing even more difficult. The extreme differences in the market characteristics of the German stock exchanges have to be taken into account in investigating the decision to go public and the market value of a firm. Therefore, the empirical studies in this dissertation focus on both market phases, with a sample of IPOs from 1997-2007. This is the longest time period investigated with German data and offers the largest sample of IPOs, compared to previously finished papers related to the Neuer Markt and this research topic. Furthermore, chapter II (How Do Pre-IPO Shareholders Determine Underpricing?) investigates the pre-IPO ownership structure of firms in more detail. This requires the collection, by hand, of information from IPO prospectuses, something which has not previously been completed for an 11-year time period. The firm’s shareholder structure prior to the initial public offering is interesting to analyse, as several theories consider information asymmetries, monitoring requirements and agency conflicts between owners, as well as investment banks and potential investors, to influence the optimal offer price and the amount of “money left on the table” due to underpriced shares. German firms in particular are often held by only few shareholder groups prior to the IPO, compared to the USA, which enables reassessment of previous research results into different firm’s conditions. Furthermore, the analysis includes the implications of ownership structures in different market cycles and their effects on IPO underpricing, which further contributes to the existing research literature. The results confirm that, in the German market, the determinants of initial returns have changed over the time period investigated. Indications of different pre-IPO shareholders’ bargaining interests in terms of offering prices and underpricing are also confirmed, for the divergent market environments.

INTRODUCTION

19

Although the empirical findings suggest a degree of willingness to leave money on the table, the offering prices of newly issued shares should reflect the firm’s value. Investment banks, acting as underwriters in the IPO process, have to value the firm according to their characteristics, accounting information and expected profitability, and so determine approximate share prices. Chapter III, entitled “Do “Herding” Effects on Firm Multiples Determine IPO Valuation?” discusses this issue in more detail. The system of using comparable multiples (e.g. priceearnings ratios or market-to-book ratios) of firms already publicly traded in order to determine the expected market value of the IPO is investigated in particular detail. The German stock market during the period of the Neuer Markt showed severe variability of valuation, especially in particular industry segments. Here, the question of whether changes in IPO valuation can be explained by the corresponding development of public industry-related firms is analyzed. Herding effects on specific information and the neglect of firm characteristics are considered as possible explanations. However, the results indicate that the valuation of industryrelated firms or their multiple ratios are minor explanatory factors. Overall stock market performance seems to be a more reasonable value driver for IPOs, especially in hot markets. Overall, the findings of this dissertation make major contributions to the research of IPO underpricing and valuation. Its focus on the German market and the environment of different market cycles in particular give further insights into this topic.

THEORY AND EVIDENCE OF IPO UNDERPRICING

21

Chapter I: Theory and Evidence of IPO Underpricing I

Introduction

The phenomenon of underpricing of initial public offerings is seen at more than 15% on average, across almost every country. This was first noted by Logue (1973), Reilly (1973), and Ibbotson (1975), who also suggested some explanations for why issuers “leave so much money on the table” 2 when going public. Others theoretically explored these suggestions in later works and related empirical investigations partly confirmed some theories and contradicted others. Over time some changes in the determinants of underpricing have been recognised and have influenced further research about this topic. This chapter focuses on the existing literature to highlight the development and explanations of this widely discussed topic in corporate finance. Some literature overviews also give insights into existing research work, for example Jenkinson/Ljungqvist (2002) or Ritter/Welch (2002), which mainly focus on ownership and share allocation as possible determining factors. Also Ritter (2003) refers to several theories of the mechanisms of pricing and valuation in the context of the stock market characteristics of USA and Europe. The subject of this chapter, however, is the direct comparison between offering prices and underpricing in the USA and Germany. Several differences in the IPO process and institutional environment exist in these countries and the implications for offering prices of IPOs are especially interesting to consider. The broad overview of theory and evidence from USA and Germany contributes to the existing literature and shows some aspects which have to be discussed in further detail. Section II of this first chapter discusses previous research papers based on the assumption of asymmetric information distribution between the participants in the IPO process. It distinguishes between the issuer, underwriter and investors. One party has an informational advantage over the others. Section III is based on theories that are not related to the agency theory, and are sometimes described as “adhoc” explanations. In both sections, the early theoretical models and their implications are first described briefly. Secondly, some extended theories and developments are considered, and then related empirical evidence and results are discussed. In the case of no empirical proofs being available for the German market, reference is often made to studies about the stock market in Europe. Furthermore, 2

This describes the difference between the offer price and the first closing market price multiplied by the number of shares sold at the IPO.

THEORY AND EVIDENCE OF IPO UNDERPRICING

22

it should be noted that many studies of Germany investigate a much smaller sample size (maximum 353 firms) than studies in the USA, with samples including about 3000 IPOs and more.3 In section IV some aspects of the various explanations are discussed and give ideas for further research. Section V concludes, summarising the important differences between the various explanations of underpricing in both countries.

II Theory and Evidence Based on Asymmetric Information Distribution II.1

Information Asymmetries between Issuer and Investor

II.1.1 Signalling Theory One strand of literature considers the asymmetry in information available to the issuer of an IPO and to potential investors. In these models, the issuer is well informed about the firm’s value and risk of the future cash flows, whereas the investors are, by comparison, uninformed and are not able to determine the quality of the firm.4 However, the issuing firm is able to signal the true value or quality to potential investors through underpricing. Allen/Faulhaber (1989: 307 et. seqq.) argue that only good firms underprice their issue, because they expect to recoup losses resulting from lower proceeds when performance is later realised by the market. High-quality firms are able to trade off a lower IPO price against a more favourable interpretation of future high dividends. There is a separating equilibrium when good firms signal and bad firms do not, because of the costliness of foregone returns. The suggestions made by Allen/Faulhaber (1989: 317) are consistent with “hot issue” periods: in an industry with favourable market conditions and positive net present value projects, the new issue market may involve separation and underpricing. This contrasts with a fairly competitive market with low profits and low dividends, in which a pooling equilibrium without underpricing is likely to occur, because signalling becomes too costly for high-quality as well as for low-quality firms. In a related study, Michaely/Shaw (1994: 305) find no evidence linking underpricing, high future earnings and subsequent dividend policy, rather the opposite is found: firms with higher earnings and dividends demonstrate less underpricing when going public. Additionally, the market does not react differently to dividend initiations whether announced by underpriced firms or not.

3 4

An overview of the discussed literature with sample size, date and empirical results can be found in the appendix of chapter I. These models refer to the theories of, e.g., Akerlof (1970), Spence (1974).

THEORY AND EVIDENCE OF IPO UNDERPRICING

23

A similar argument for underpricing is derived from Welch’s model (1989: 436, 441). The issuer signals the firm’s quality to investors by intentionally setting a lower IPO price, because higher compensation due to higher prices in seasoned equity offerings (SEO) is expected. The underpriced firms sell a higher proportion of their shares in the IPO relative to the seasoned offering, when the value of highquality firms or the cost of imitation for low-quality firms increases. Welch (1989: 443 et. seqq.) finds empirical evidence that the timing of SEOs is indeed related to the IPO of a firm. A similar observation is made by Chemmanur (1993: 286 et. seqq.). In his model underpricing generates publicity about the firm. High-quality firms have an incentive to maximize information production, so that the information will be reflected in the secondary market and the price for seasoned offerings will be closer to the true value of the firm. Firms planning to approach the secondary market will have a lower offering price, since the benefits of underpricing and information production are only obtained in SEOs. Additionally, several empirical studies have analysed the implications of these models. For example Jegadeesh/Weinstein/Welch (1993: 173) find that the market reacts favourably to SEO announcements by firms with high IPO returns, in contrast with announcements by firms with relatively low underpricing. However, they conclude that the results for the signalling hypotheses are rather weak. Aftermarket returns are more useful than the level of underpricing in predicting the likelihood of SEOs (Jegadeesh/ Weinstein/Welch (1993: 164)). The results found by Michaely/Shaw (1994: 319) are consistent: Successful firms decide to reissue and unsuccessful ones do not, but there is no association between greater underpricing and the likelihood of a SEO. Also, a study by Garfinkle (1993: 82) finds that the effects of signalling on both the likelihood of reissuing and on the abnormal return to the announcement of a seasoned offering are only incremental. Contradictory results were noted by James (1992: 1875), who considered public securities offerings (common stock, straight debt, convertible debt, and common stock with warrants) in the context of the level of underpricing and firm-specific information costs. The results show that the mean IPO percentage underpricing is higher for firms that do not make a subsequent offer. Bessler/Thies’ (2002: 16 et. seq.) findings for German IPOs before the hot issue period of the “Neuer Markt” are not that explicit. The firms with the highest initial returns or run-up over the following months make SEOs more often, to raise additional capital. However, these firms have the worst performance after one year. These results indicate that underpricing does not necessarily signal the firm’s quality to investors; rather managers have the opportunity to time the SEO successfully. Another signalling hypothesis is based on the inferences made by Leland/Pyle (1977: 376). They consider the entrepreneur's willingness to invest in his own pro-

24

THEORY AND EVIDENCE OF IPO UNDERPRICING

ject as a signal of quality, because the owner refrains from wealth diversification, which is more costly for high-risk firms. The entrepreneur is motivated to choose a greater fraction of ownership in more favourable projects. In Grinblatt/Hwang’s (1989: 395, 407 et. seqq.) related model, the entrepreneur is able to signal the firm’s quality and risk by retaining a fraction of the new issue and by choosing the offering price. In this case ownership is not sufficient to signal the expected value of the project. Moreover, a low offering price is needed to signal the related risk. The results suggest that underpricing is an increasing function of the variance of cash-flow returns, whereas higher variance issues indicate higher expected value.5 Therefore underpricing is positively related to the issuer’s fractional holdings. Opposite implications are seen in Bachmann (2003: 30): To signal the quality of investment opportunities, underpricing is inversely related to the proportion of the original shareholders’ shares sold. Michaely/Shaw’s (1994: 314) empirical test finds that insider ownership has no significant power to explain underpricing and the two-year return, which proxies for the value of the firm. For the German IPO market, the association between ownership and underpricing is investigated by Wasserfall/Wittleder (1994:1511 et.seq.) and Ljungqvist (1997: 1315), who find positive correlations between insider retention rate and IPO underpricing. Similar to Bachmann (2003), in these studies ownership is hypothesised to be a substitute for underpricing, not an additional signal: Higher insider ownership is associated with willingness to carry the risk of the firm after the IPO, so a negative effect on underpricing is expected. Another argument is proven by Barry (1989: 1101 et. seq.), highlighting the ownership of pre-IPO shareholders from another perspective. The wealth loss due to underpricing is greatest for participating owners who sell all of their shares in the IPO, and least for shareholders who sell none. If they issue a small amount of shares in relation to their original holding, they will be less concerned about wealth losses due to underpricing. When the offering is purely secondary, there is no dilution effect on retained shares. The wealth transfer from initial owners to new shareholders due to initial returns is especially costly in large issues with high participation of the previous owners. Habib/Ljungqvist (2001: 449) find significant evidence of reduced underpricing with higher participation of owners, defined as the number of secondary shares sold in relation to overall pre-IPO shares. Besides this, Ljungqvist (1999: 16) also confirms that issuers are more concerned about underpricing where dilution is high, which is measured by the amount of primary shares divided by total original shares prior to the IPO. Also, during the dot-com bubble in the USA the amount of underpricing has been inversely related to the dilution factor and owner participation ratio (Ljungqvist/Wilhelm (2003: 742). For IPOs in Europe’s New Markets during 5

See also chapter I section II.2.1 for similar results for the relation of underpricing and risk (Beatty/Ritter (1986)).

THEORY AND EVIDENCE OF IPO UNDERPRICING

25

1995-2002, Giudici/Roosenboom (2002: 22) confirm that underpricing has been less severe in issues with lower dilution and participation of pre-IPO shareholders. For Germany during the same sample period Franzke (2002: 20) cannot find any significant effects of these variables on the level of underpricing. II.1.2 Certification of Quality Further models considering the asymmetric information problems between issuer and investors involve third parties, which signal the quality of a firm. The issuer can choose intermediaries with a certification-of-quality function instead of signalling the firm’s value by underpricing. Therefore, these models share an assumed negative relation between the alternative signalling action and initial return. According to Titmann/Truemann’s (1989: 166 et. seqq.) theoretical explanation, the choice of a more prestigious auditor will result in a higher offering price for the new issue. High-quality auditors derive additional, and more precise, information from the financial statements, which allow investors to estimate the firm’s value more accurately, so the offering prices for the good quality issues will be higher. The choice of auditor will not be mimicked by an entrepreneur with less favorable information about the firm’s value, because reduced information asymmetries will be less beneficial for him. Beatty (1989: 696, 704 et. seqq.) measured auditor quality by estimating the compensation premium paid by the issuer: the higher the auditor fees, the higher the exhibited reputation capital. The results provide evidence of the ex-ante signalling function, because compensation premiums are higher for those auditors commonly associated with high reputation (currently only the “Big Four” auditors) and these are negatively related to initial returns. Hogan’s (1997: 69, 80 et. seq.) study is more complex, due to the assumption that issuers choose the auditor by comparing the costs and benefits. He suggests that issuers minimize the sum of auditor fees and the amount of underpricing, whereas the auditor choice is considered as an endogenous variable. The results however confirm the assumptions, because IPO firms select auditors to minimize total costs. Besides the revelation of entrepreneurial private information, auditor choice can provide a signal of insurance coverage for investors, which is hypothesised in Willenborg’s (1999: 227 et. seq.) paper.6 To disentangle the effects of these two signals, the sample of IPOs is divided into start-up and established companies. Start-ups are more likely to demand insurance against litigation or risk of failure, rather than additional signalling of the firm value by choosing high quality auditors, whereas for established firms both certification functions are important. The results support the argument. As IPO proceeds increase, the auditor is compen6

See chapter I section III.2 for further evidence of the argument that underpricing reduces litigation risk of the issuer.

26

THEORY AND EVIDENCE OF IPO UNDERPRICING

sated for insurance coverage and the additional quality certification he is providing (Willenborg (1999: 235). Regarding accurate information revelation, Leone/Rock/Willenborg (2006: 2, 15) examine details of the use of proceeds in the IPO prospectus, and Schrand/Verrecchia (2005:16, 18) information revelation such as press releases, finding an inverse relation to IPO underpricing. Hunger (2001:191) compares the revelation requirements of the different stock segments in Germany during the period of the “Neuer Markt”. There is no evidence that stricter firm announcements for admission result in lower initial returns, in fact the opposite is found. Hopp/Dreher (2007: 10, 23) compare countries’ legal and institutional settings and infer that greater corporate transparency and financial disclosure determine the differences in countries’ average underpricing. The findings support the argument that countries’ accounting ratings have a significant negative impact on the level of IPO underpricing. Boulton/Smart/Zutter’s (2007: 13) analysis concentrates on differences in countries’ earnings quality and concludes that less reliable earnings and financial accounting information lead to higher underpricing. In comparison, the evidence of the venture capitalists’ (VCs) certification function is not that consistent. In general VCs are associated with long term monitoring and managerial advice for young firms. The first approach is proven by Megginson/Weiss (1991:880et. seq.), who consider a VC as a partial substitute for, and a complement to, the certification provided by prestigious auditors and underwriters. Third parties only have an incentive to certify the true value of an issue when they have reputational capital at stake which would be forfeit should they be found to be signalling falsely. Megginson/Weiss’ (1991: 890 et. seqq.) results show that VCs are likely to encourage the choice of more prestigious auditors and underwriters. This effect reduces the informational asymmetries between issuers and investors, and provides financial expertise to the firm. The costs of going public are lower for VC-backed IPOs, because lower levels of underpricing and lower underwriter compensation can be achieved. Contradictory evidence is seen in a subsequent study by Bradley/Jordan (2002: 613) considering the hot issue periods of underpricing. After allowing for the industry segment as an explanatory variable for the high initial returns, they find no significant difference between VCbacked and non-VC-backed issues. The empirical investigations for German IPOs are also inconsistent. All studies include the hot issue periods of the “Neuer Markt”, which has been considered as the market segment for young growth companies, the main target for venture capitalists. The IPO exit opportunities during this period were therefore advantages, although still not as important as those in

THEORY AND EVIDENCE OF IPO UNDERPRICING

27

the USA.7 Georgen/Khurshed/Renneboog (2006: 9, 14 et. seq.) show that about 47% of IPOs during 1996-2000 were venture backed. The level of underpricing in this sample averaged 50.77% for VC financed firms compared to 54.75% for firms without VC; the difference is not statistically significant. This is also noted in the study of Kraus (2002: 20) after controlling for further determinants of underpricing. In Franzke’s (2003: 20) and Bessler/Kurth’s (2007: 39) studies, average underpricing is also abnormally high. Surprisingly they find a positive relation between VC support and the level of underpricing. One possible explanation is that VCs are exit-driven and therefore willing to accept lower issue prices. Young VCs are especially subject to “grandstanding”, which means that these intermediaries prefer early exits through an IPO to build up a reputation. Often their portfolio firms fulfil the requirements for public listing but are still associated with higher uncertainty and less monitoring over time.8 A further development of the VCs’ certification function takes into account the insider’s decision to sell shares in the IPO. VCs try to balance the costs of continued involvement and ownership against the adverse market reaction to insider selling. Lin/Smith (1998: 247, 259 et. seqq.) argue that unfavorable interpretation by potential investors can be reduced by developing a reputation for not selling overpriced shares. The results suggest that VCs with an established reputation refrain from selling at the IPO, unless the issue is significantly underpriced. For less reputable VCs the decision to sell at the IPO does not lead to higher initial returns. Brav/Gompers’ (2003: 5, 9) study improves the hypothesis by considering the role of lockup periods for owners and VCs in this context. The restriction on prepublic investors and owners selling stock in the aftermarket for a fixed period of time is also a commitment device to alleviate moral hazard problems. Firms associated with high risk and higher probability of insiders taking advantage of shareholders should impose a longer lockup period to induce investors to participate. Issues with reputable third party certifications are unlikely to exploit outsiders and therefore are associated with shorter lockups. The investigation supports the hypotheses: firms with high quality underwriters and VCs have, on average, shorter lockup periods, whereas high risk and high growth companies utilize longer lockup length. However, underpricing is higher in IPOs with high reputation underwriters and VC financing. In the USA the lockup commitments are voluntary 7

8

In Germany in 2006 only 42 (6.3%) of VC-backed companies went public with a total offer amount of € 437.3 mio. compared to 118 IPO (13.6%) with a volume of € 156.7 mio. in 2000 (BVK (2007a: 40 et.seq.)). While in 2006 in the USA 57 IPOs were concluded with a total offer amount of $5,117 mio. and the highest total offering size of $ 25,419 mio. in 263 IPOs were achieved in 2000 (BVK (2007b: 16)). See also Gompers/Lerner (1996:134 et. seqq.) for further explanations of grandstanding of young VCs.

28

THEORY AND EVIDENCE OF IPO UNDERPRICING

agreements between the issuer and underwriter; typically about 180 days following the IPO. In Germany during the “Neuer Markt” period, the lockup commitments were regulated to be at least 6 months and could be voluntarily extended by the issuer and/or underwriter. In Bessler/Kurth (2007: 43) an extended lockup period is related to underpricing of 22.1%, which is half the size of the observed initial returns of 45.7% in the issues with minimum lockup periods. This is similar to the findings of Brav/Gompers (2003). Considering only IPOs in Germany’s “Neuer Markt”, Georgen/Khurshed/Renneboog (2006: 10, 19) find no significant relationship between underpricing and lockup length or the percentage of lockedup shares in the offering. Another external certification function which can reduce informational asymmetries is the choice of a reputable underwriter. Booth/Smith (1986: 262 et.seqq) developed a model in which an investment bank can certify the price of equity and debt. The issuers’ bank relationship prior to the IPO is also valuable in signalling a firm’s quality. Banks are able to monitor and screen their clients at low cost: Therefore a good bank relationship ameliorates informational asymmetry problems between issuer and potential investors, resulting in reduced underpricing. Slovin/Young (1990: 730, 736 et. seq.) hypothesize that the positive effect is mainly associated with relatively small and unknown firms. In this sample an issuing firm’s agreements on credit lines and debt reduce initial returns by approximately 30% and 32% respectively. Schenone (2004: 2910) suggests the relationship between a lending bank and the underwriter as a relevant signal for market participants. The choice of lending bank, as well as switching to another bank for underwriting the issue, reduces information asymmetries. The change to another underwriter permits the assumption that the relationship bank has unfavorable private information, and this determines a lower firm value at the IPO. The informational gap between investors and issuers exists only when pre-offering banks are not able to act as an underwriter. The regression results confirm these assumptions (Schenone (2004: 2915, 2922 et. seq.): Firms that could have gone public with their own bank, but instead switched banks, are on average of lower value than those that could not have gone public with their bank. These firms experience the highest returns after the first trading day, due to the expected uncertainty about the firm value. In the German banking sector close “Hausbank” relationships are more common, so Klein/Zoeller (2003:5, 10) differentiate between universal banks (commercial bank and investment bank) and specialist banks as underwriters. On one hand, universal banks can offer informational advantage to certify a firm’s quality; on the other hand these banks are able to transfer their loan risk to uninformed investors. Universal banks’ issues are associated with higher than average underpricing, which confirms the presumptions of conflicts of interests between

THEORY AND EVIDENCE OF IPO UNDERPRICING

29

banks and investors. High leveraged firms particularly choose specialised banks, as the level of debt is an indicator of potential conflicts between underwriter and investors. II.2

Information Asymmetries between Investors

II.2.1 The Winner’s Curse A large part of the IPO underpricing theory refers to Rock’s (1986: 190) “winner’s curse” model, which is based on the assumption of informational asymmetries between different groups of investors participating in the IPO. The “informed” investors have full information about the value of a new issue, compared to all other investors, including the issuer and underwriter, who are called the “uninformed”. This argument implies that the market in aggregate knows more than any individual player, and more than the issuer and its agents. Rock also (1986: 192 et seqq.) assumes that the informed only bid for offerings which are attractively priced, while uninformed investors bid indiscriminately. In the case of excess demand leading to rationing of an issue, uninformed investors are more likely to receive overpriced shares than shares of an underpriced issue. Thus the average return is weighted towards the overpriced offerings, which is also known as the “winner’s curse”. As the demand of the informed is insufficient to take up all shares offered, the uninformed must be kept in the market. Issuers must then price shares at a discount, as a compensation to the uninformed for receiving a disproportionate number of overpriced stocks. This does not reduce the bias of rationing of good issues relative to bad issues, but the lower equilibrium price mitigates average losses for the uninformed group of investors. The first empirical test of the model is analyzed by Tinic (1988: 795 et. seq.) finding no evidence for the proposition that institutional investors, considered as the “informed”, receive disproportionately large amounts of underpriced IPOs. Hanley/Wilhelm (1995: 246 et.seq.) found results contradictory to this argument. However, they do not accept Rock’s hypothesis, because institutions also receive a large proportion of the overpriced shares. In a more recent study, Aggarwal/Prabhala/Puri (2002: 1427, 1429) provide some evidence that institutions earn greater profits than, and at the expense of, retail investors because of favourable share allocation. The mean profit per issue for institutions is $14.79 mio., while the retail investors’ profit is on average $5.28 mio. From another perspective, Rock’s implication is accepted in Michaely/Shaw’s (1991: 288 et. seqq.) paper. There is no need for underpricing in issues with an a priori knowledge of information homogeneity among investors. For this purpose, they analyse a sample of IPOs of master limited partnerships (MLP), which are largely avoided by institutional investors due to several differences in tax requirements. The absence of these is expected to reduce the potential

30

THEORY AND EVIDENCE OF IPO UNDERPRICING

bias in share allocation against individual investors. The results show that the average initial return is insignificantly different from zero for the MLP sample. However, the results are questioned in Leite’s model (2004: 369 et. seqq.). He argues that the calculation of average underpricing is underestimated, when demand for allocation is ignored. The initial returns have to be weighted by the extent to which the issue size is increased relative to a pre-stated minimum, because the issue size is positively related to the demand for share allocation. A broadly accepted implication in the underpricing literature, based on Rock’s model, is noted by Beatty/Ritter (1986: 216 et. seqq.). They find evidence that there is an equilibrium relationship between expected underpricing of an IPO and the ex-ante uncertainty of the firm's value. In this framework informed investors have to incur some cost to determine the true value of an issue. The greater the uncertainty about the fundamental value, the greater is the required compensation to become informed and to be willing to submit a purchase order for shares. In the empirical investigation the information about use of proceeds in the prospectus and the inverse of the gross proceeds indicate the ex-ante uncertainty about the firm value. With the same proxy for uncertainty Ljungqvist (1997: 1314 et. seq.) finds a significant negative relation between offer size and the level of underpricing in Germany. However, an interesting note by Habib/Ljungqvist (1998: 383) proves that the inverse of proceeds is an inappropriate proxy for ex-ante uncertainty. Underpricing will decrease in IPO proceeds even in the absence of a change in uncertainty. Nevertheless, generally significant results are found for the expected positive correlations between initial returns and several other proxies for ex-ante uncertainty, e.g., company age, annual sales, offer price and volume as well as standard deviation in daily aftermarket returns.9 Ritter (1984: 222, 228) however finds no evidence that changes in the risk composition of IPOs, measured with the proxies, annual sales and standard deviation of daily returns in the aftermarket, are the only determinants for the abnormally high underpricing in hot issue periods. However, from an issuer’s perspective, the main reason for underpricing is the compensation to investors for risks taken.10 Another extension of Rock’s model is analysed in Carter/Manaster’s (1990: 1046 et. seqq.) article, taking into account the certification function of prestigious investment banks. Here, the informed investors specialize in acquiring information for the most uncertain investments. As their capital migrates to the uncertain is9 10

For detailed listing of several proxies for ex-ante uncertainty in different studies see Jenkinson/Ljungqvist (2001: 70 et.seq.). See Brau/Fawcett (2006: 415). They analyzed several theoretical explanations for going public and underpricing form the perspective of chief financial officers (CFOs).

THEORY AND EVIDENCE OF IPO UNDERPRICING

31

sues, the offer prices must decrease to guarantee the participation of the uninformed for full subscription of the issue. Low dispersion firms do not attract informed investors and do not require protection for the uninformed participants. Therefore they are able to obtain higher offering prices and benefit when making that fact available to the market. To communicate this, the firms are able to contract an underwriter with a reputation of marketing IPOs of low dispersion. Verification of the theoretical results is conducted by Carter/Dark/Singh (1998: 289 et. seq.). In this paper several measures of investment banks’ prestige are considered: Firstly, the four-tier underwriter reputation measure, by Johnson/Miller (1988), secondly the relative market share, by Megginson/Weiss (1991), and thirdly the ten-tier ranking determined by underwriters’ tombstone announcements, by Carter/Manaster (1990). All three methods in the study also have a significant explanatory power regarding the level of underpricing in initial offerings. Also, for the IPOs in Europe, Giudici/Roosenboom (2002:24) determine, with Megginson/Weiss’ (1991) measure, significantly lower underpricing for issues from underwriters with a large market share. Interestingly, Franzke (2003: 14, 22) finds no significant effect of underwriter prestige on offering prices in Germany. The reputation measure takes into account the relative shares of lead managements of the investment bank, and the relative volume of proceeds of issues launched on the “Neuer Markt”. A possible explanation could be that underwriters, especially in Germany, also have further intermediary functions in the IPO firm, e.g., they are related to VCs or they have a lending relationship. Investors therefore could assume greater conflicts of interest and agency problems with a reputable and broadly positioned investment bank than the certification of low firm-specific risk. Hopp/Dreher (2007: 22) argue from a different perspective and find significant results: Restricting banks’ business activities could limit price discovery in the process of going public instead of reducing informational asymmetries between banks and investors. Bank restriction therefore positively influences the level of underpricing. Furthermore, in this context it is interesting to consider the determinants of change in underwriters’ market share. Dunbar (2000: 15 et. seqq.) shows that banks associated with overpriced issues and with the maximum level of abnormally high underpricing lose market share, while the minimum level of abnormal initial return has a positive effect. Investment banks then, can enhance their market share and reputation as neither issuers nor investors are harmed by pricing IPOs. Balvers/McDonald/Miller (1988:610, 618) bring up an interesting relationship in the choice of underwriters and auditors. As both reduce ex-ante uncertainty and can be interpreted as a signal to uninformed investors, they are expected to reduce the winner’s curse problem. It is in the underwriter’s interest to hire a reputable

32

THEORY AND EVIDENCE OF IPO UNDERPRICING

auditor, since better information about earnings makes it easier to price the issue correctly and thereby maintain reputational capital. More prestigious investment banks are associated with auditors of higher reputation; both are negatively correlated to underpricing. But as one reputation variable increases, the effect on initial returns of the other diminishes. The effect of a third expert influencing the informational asymmetries is discussed in Beatty/Welch (1996: 584,588 et. seq.), which includes the reputation of law firms accompanying the process of going public. The study shows that using reputable auditors lowers underpricing by about 3%, while the presence of high-quality lawyers is not an important determinant. For the sample period during 1982-1984, they find that by far the most important determinant for low initial returns is high underwriter quality, while the results for 1992-1994 show that high quality underwriters are associated with more underpricing. The same conclusion is drawn in Cliff/Denis (2004: 2888). They suggest that over time factors other than reputation have become important in the choice of underwriters, for example, having a prestigious analyst covering the IPO firm’s industry. Habib/Ljungqvist (2001: 449) view all possible certification opportunities for uniformed investors as promotion, which is seen as a substitute for underpricing. The results show that ex-ante uncertainty provides no explanation for the issuers’ spending on promotional activities. However, every dollar of promotion reduces wealth loss, in terms of underpricing, by $0.98. The marginal costs are equal to marginal benefits; issuers seem to choose optimal promotion spending. II.2.2 Information Revelation Based on the assumption of investors’ informational advantages, Benveniste/Spindt (1989: 347 et. seq.) developed a theory in which underwriters have more than a certification function in reducing the winner’s curse. Investment banks can induce investors to reveal their information by relating share allocation to investors’ indication of interest. In a procedure similar to that of bookbuilding, the issue is premarketed to regular investors with private information and then opened to occasional investors. When regular investors give a positive signal about their information status, they would normally expect the offer price to rise. As an incentive to be truthful and show strong demand, they must be rewarded. Therefore, issues have to be underpriced and these investors receive a disproportionate allocation of shares. Furthermore, Benveniste/Spindt (1989: 357 et. seq.) differentiate between types of contracts for underwriters’ intermediary function. In a firm-commitment offer the underwriter has to purchase all shares not presold at the offer price. This motivates pre-selling of the whole issue, thereby allotting higher allocations to low-interest investors with negative private information. With a best-effort contract the underwriter is not bound to purchase any unsold

THEORY AND EVIDENCE OF IPO UNDERPRICING

33

shares, but with an early offer price setting this entails greater proceeds uncertainty than a firm-commitment contract. Thus firms facing the most uncertainty may select a best-effort underwriting contract, while those firms whose owners are risk averse are more likely to choose firm-commitment contracts with a guaranteed level of proceeds. As shown in Benveniste/Wilhelm (1990: 194), limitations for underwriters in the marketing of IPOs, such as constraints in price and allocation discrimination between different investors, increase the costs of revealing private information. With the requirements of evenhanded allocation of oversubscribed issues and uniform prices for regular and occasional investors, underpricing is higher than necessary and this reduces expected proceeds. Shermann/Titmann (2002: 5, 15 et. seqq.) also extend the model by assuming that information is costly for investors to obtain. When investment banks want to collect more information for their pricing decision, more investors have to participate and higher underpricing is required to attract them. This is contradictory to Benveniste/Spindt (1989), who state that larger investor groups lower initial returns. In cases of less accurate pricing by underwriters, only few investors will be invited to participate and the level of underpricing will offset the investors’ information costs. When more accurate pricing is required, the size of the potential investor pool, as well as expected underpricing, must increase and informed investors are able to earn increased returns. Also, when there are participation limits, in terms of minimum numbers of investors or a maximum share allocation per investor, both the size of the investor pool and expected underpricing will increase. However, the bookbuilding procedure still leads to underpricing because of uncertainty about the firm value, which is analyzed in Draho (2006: 15 et. seqq.). He concludes that, as long as issuers and investors are risk averse, underpricing due to uncertainty will be necessary if there is insufficient information production independent of other motives for lower offering prices. Comparing a best-effort contract and a firmcommitment contract, the first is likely to be more underpriced because of riskier firms and because of an offer price set in advance of the offer date and before investors have submitted orders. However, compared to a fixed-price mechanism, bookbuilding allows more and better information to be generated, which lowers uncertainty about the market price and reduces the underpricing required. Welch (1992: 695 et. seq., 723) models the information revelation in fixed-price offerings, when investors make their subscription decision sequentially. He suggests that due to investment banks’ limited distribution channels, it takes some time to approach interested investors. Later investors are able to observe how successfully an offering has been sold to date. Those participants base their investment decision solely on the previous investors’ actions and ignore their private information. If early investors believe that the issue is overpriced, they can swamp

34

THEORY AND EVIDENCE OF IPO UNDERPRICING

the information held by all other investors and doom the offering to failure. On the other hand they can create unlimited demand when they consider an issue to be underpriced. This is called the “cascade” effect, and can cause underpriced issues to fail, while overpriced ones can succeed. Early investors are able to demand higher initial returns to start a positive cascade, whereas underwriters try to find investors who are unlikely to communicate, so that the issuer is at less informational disadvantage when pricing the offering. This reduces the explanation of the winner's curse when offerings are sold over a period of time. The implications of these theories are analyzed by Cornelli/Goldreich (2001: 2353 et. seqq.). They look more closely at investors’ bids and the allocation of shares in bookbuilt offerings, finding that insurance companies and pension funds are favored in allocation. Larger bid size and price bids, which provide more private information for underwriters, are also preferred. However, frequent investors do not receive a higher return on their bidding strategy; this means that favorable allocation is necessary to compensate them for also taking shares of poor issues. The impact of different institutional constraints in terms of information available when the offering price is set, and whether there is discrimination in the allocation of shares, is investigated in a country comparison by Loughran/Ritter/Rydqvist (1994: 173 et. seq.). They find that initial returns are lower in firm-commitment contracts, when information about the demand is acquired prior to the offer price, compared to best-effort contracts. In the USA more issues are offered with a firmcommitment contract, while in Europe best-effort offerings are common. However, the effects of the stage at which an offering price is set are offset by different firm characteristics, which additionally influence the amount of first-day returns. No clear pattern emerges, when the influence of discretionary allocation of shares is analyzed. Furthermore, Hopp/Dreher (2007: 24) find different pricing mechanisms, such as fixed-price or bookbuilding offerings, have no impact on variations in underpricing in several countries. If the assumptions regarding information revelation from investors in bookbuilding are accepted, it remains to be considered how underwriters include participants’ demand functions when setting an issue’s offer price. Hanley (1993: 233 et. seq.) hypothesizes: When pre-selling demand for allocation is higher than expected, the final offer price will be set higher than the offer price range disclosed in the preliminary prospectus. If demand is relatively low, the offer price will be below the anticipated offer price, due to unfavorable information revealed in the pre-issue period. The revision of offer prices is accompanied by changes in the number of shares being issued. Hanley’s (1993: 241 et. seqq.) analysis shows that the risk associated with an issue, overall stock market conditions, and the pre-sale

THEORY AND EVIDENCE OF IPO UNDERPRICING

35

activities of the underwriter affect the final offer price. Additionally, there is a significant positive correlation between revisions of the offer price and changes in the number of shares. Issues with positive information revealed also have substantially greater underpricing than offerings with prices below or within the preliminary offer range. This means that underpricing is used in conjunction with increases in share numbers to compensate investors for indicating their positive private information. Bradley/Jordon (2002: 599, 604) break down the total adjustment of the offer price into pre-offer adjustment and final offer adjustment, because adjustments of file ranges previous to the final offer price signal a revision of a firm’s valuation. In the sample, it is apparent that firms are more underpriced if they amend their price range upwards, regardless of the final offer price. However, the initial returns are greatest for firms with an offer price above the final price range; even if the initial price range has been adjusted downwards. Also, when the information revelation process is considered for a syndicate of underwriters, the empirical results are similar. Corwin/Schultz (2005: 467) find that the likelihood and size of an offer price revision increase with the size of the underwriter syndicate and when shares are broadly distributed across underwriters. This shows that ability to reveal private information increases with the number of underwriters and with the number of potential investors. Additionally, market-wide return considered as publicly available information leads to offer price revision, although it is unaffected by syndicate structure. In Liu/Shermann/Zhang (2008: 11 et. seq.) media coverage prior to the offering is regarded as an indicator of investors’ private information, and is an additional proxy for information to underwriters. Depending on positive price revision, more media attention is associated with higher underpricing; an additional item of media attention results in a 2% increase on initial returns. When the news components of publicly available information and market returns are tested, no significant relationship is found. This suggests that underpricing is not used to compensate for the production of public information. Löffler/Panther/Theissen (2005: 468, 472 et. seqq.) analyzed a particularly interesting aspect of this in the German market, where an offer price range revision or a price above the initial offer range is less frequent than in the USA. Downward revisions and lower offer prices than previously expected, however, do occur. They investigate the pre-IPO (grey) market, which is an active market where shares in the process of being issued are already traded in terms of forward trades. Pre-IPO quotes are a good indicator of prices on the first trading day, and are more informative than the offer range and offer price. Pricing accuracy in the pre-IPO trades steadily increases towards the offering date. This indicates that information quality rises both because of the arrival of public information and because of gradual incorporation of private information. The results show that grey market quotes con-

36

THEORY AND EVIDENCE OF IPO UNDERPRICING

tain information related to the value of the IPO firm, which exceeds the information that can be inferred from publicly available control variables such as risk, underwriter reputation, price revision, etc., and this is therefore an explanatory variable for underpricing. Another aspect of the European IPOs is that the mechanism of “pilot fishing” is becoming more established. In the USA this behavior is a signal of a weak issue, and is seen as related to unsuccessful IPO firms. There is no recent literature about this pricing behavior, where underwriters approach 10-15 regular institutional investors before the actual bookbuilding process begins. These investors are expected to offer specialized knowledge about industry perspectives, and to give an estimation of the probabilities of success by making clear their demand for share allocation. Underwriters aim to reveal preliminary views and private information with this mechanism, which can be conveyed in the offer price range in the bookbuilding procedure and can be used to refine thoughts in the transaction structure. Due to limitations on price settings and discretionary allocation or best-effort offerings, European underwriters are not able to include all private signals of demand received during bookbuilding. Underwriters thereby, as Benveniste/Wilhelm (1990) explaine, try to avoid excessive underpricing and reduce the costs of going public. Another simple reason could be that investors’ demand covers a large percentage of the offering right at the beginning of the bookbuilding process, ensuring a more favorable allocation of shares within a successful IPO. Another mechanism for reducing IPO costs and ensuring compensation of investors is suggested by Benveniste/Busaba/Wilhelm (1996: 224 et. seqq.). They discuss the previous models’ neglect of the fact that underwriters have an incentive to overstate the investor’s interest, which can lead to higher offer prices and thereby to higher fees. The incentive problem can be reduced by a commitment to providing post-offer price stabilization, which requires the underwriter to buy-back shares in the aftermarket. Price stabilization is used in complement with penalty bid systems, which avoids (retail) investors directly reselling the shares during the distribution period. This is only offered to informed participants who reveal their positive information, where a repeated interaction between underwriters and investors is assumed. Stabilization in such a case acts as a bonding mechanism to avoid overpricing and compensation for investors, which substitutes the discretionary allocation of underpriced shares as suggested in Benveniste/Spindt (1989); therefore underpricing is less than it would be in the absence of a commitment to price stabilization. This mechanism is considered from another perspective with empirical evidence, in section III.1, where underpricing is a result of underwriter price support.

THEORY AND EVIDENCE OF IPO UNDERPRICING

II.3

37

Information Asymmetries between Underwriter and Issuer

Another less extended and proven theory is developed on the assumption that underpricing performs the function of compensation for underwriters instead of for investors. The first model was described in Baron (1982: 957 et. seqq.) with bank’s superior information about the capital market compared to the issuers, in the context of a negotiated fixed price offering. The issuer can delegate the offer price decision and the distribution of securities to the bank, and utilize the informational advantage. The greater the uncertainty of the issuer in terms of the market’s perception, the greater the demand for the investment bank’s advice. Because of the underwriter’s ability to influence potential customers and certify the issue, they receive compensation as an incentive to report private information truthfully and make high distribution efforts. The compensation offered distorts the offer price and effort decisions compared to the first best solution; therefore the underwriters’ superior information and the issuer’s uncertainty reduce the offer price below the first best outcome. A simple test by Muscarella/Vetsuypens (1989: 128,130) rejects the hypothesis of Baron’s model. They examine the IPOs of 38 investment banks during 1978-1987. When the IPO is self-marketed by the bank the underwriter and the issuer are the same, so there should be no information asymmetry and hence less underpricing. However Habib/Ljungqvist (2001: 449, 456), who consider promotion spending as a substitute for underpricing, state that issuers tend to care more about the amount of “money left on the table” when their participation at the offering is higher (Barry (1982)). They suggest that it is at least conceivable that banks sell fewer shares when going public, leading to lower incentives to promote the issue and to decrease underpricing, which is neglected in Muscrella/Vetsuypens (1989). Overall, the intermediaries are presumed to have further and more complex functions in terms of mitigating moral hazard problems, as discussed earlier. On the other hand, some empirical results in the upcoming argumentations regarding underpricing in Section III confirm that underwriters do indeed profit from lower offering prices, e.g., in terms of higher secondary market liquidity or price stabilization commitment in an offering. However, these explanations differ in the important assumption of the existence of symmetric information distribution between IPO participants.

38

THEORY AND EVIDENCE OF IPO UNDERPRICING

III Theory and Evidence Based on Symmetric Information Distribution III.1 Underwriter Price Support Ruud’s (1993: 138) explanation for underpricing disregards adverse selection and moral hazard problems, as well as the assumption of a deliberate underpricing decision in an IPO. The theory refers to the frequent market practice of underwriter price support and stabilization, which prevents a decline in market prices in order to facilitate the distribution of shares. These mechanisms are expected to reduce the number of negative initial returns from what would otherwise be observed in free market trading. Ruud (1993: 144 et. seqq.) analyses the distribution, rather than the mean, of initial returns. IPO return distribution is found to depart sharply from symmetrically distributed ordinary daily stock returns, with a large disparity between mean and median that decreases as the time interval of the measurement increases. The weak or unobservable left tail of the distribution shows that negative returns are likely to be suppressed by price support, which produces a positive average initial return, although offering prices are set at the true market value. Hanley/Kumar/Seguin’s (1993: 181, 188, 193) results, analysing bid-ask spreads and prices of issues to determine the effects of stabilization in the aftermarket, are consistent with these findings. They also find evidence of those activities during the first 10 trading days, because of the lower width of bid-ask spreads, especially for those issues beginning trading below their offer price. This is assumed to guarantee liquidity service for other market makers of the issue, and lowers the downside risk. However, the withdrawal of price support negatively affects after-market prices. Schultz/Zaman (1993: 204 et. seq., 211, 215) also refer to the investor’s option to sell stock back to the underwriter in the aftermarket for the offer price, and distinguish between the underwriter’s overallotment option and price stabilization by repurchases of shares to accumulate inventory where significant price declines are expected. Overallotment gives some flexibility in the number of shares that are actually sold to the public. Underwriters are enabled to allocate up to 115% of the issue at the offering. When prices go up, the option to cover the short position is exercised, and the additional 15% remains in the aftermarket, while a “cold” issue requires the underwriter to cover their short position by buying back shares in the open market, leaving the overallotment option unexercised. Both mechanisms imply that underwriters will quote in the aftermarket at the inside bid more often than at the inside ask, especially with cold IPOs. The results confirm the assumptions: Underwriters are on the inside bid much more frequently than other market mak-

THEORY AND EVIDENCE OF IPO UNDERPRICING

39

ers not encumbered with the underwriter function of the issue. The repurchase of shares averages 21%, while overallotment options are exercised by a 90.4% chance in hot issues, compared to a 26.1% chance of exercise if the price is lower than the initial offering price. Compared to the previous studies, price support in terms of repurchases is here considered to be a permanent mechanism to avoid price decline. In this context, Schultz/Zaman (1993: 202) also regard underpricing as a compensation for underwriters, as it lowers the implicit exercise price and minimizes the value of the put written to investors; no significant evidence for this argument is found. Ellis/Michaely/O’Hara’s (2000: 1053, 1060) article also investigates the pricing decision, as well as aftermarket trading and stabilization activities in relation to each other. Interestingly, in their sample every lead underwriter becomes a market maker, also handles the lion’s share of the trading volume, and accumulates inventories averaging 7.8% of the offerings during the first 20 trading days. A simultaneous analysis of the overallotment option and the accumulated inventory shows that large positions (average inventory of 15.3%) are concentrated on those issues where the overallotment option is not exercised, compared to an inventory position of 1.37% for IPOs with more than 100% shares sold. The underwriters’ price stabilization put option is critical in reducing the risk of price support activities. In contrast to previous studies, Ellis/Michaely/O’Hara (2000: 1062 et.seqq.) conclude that market making is not costly for underwriters: Profits from market making are positive over the first three months. Trading profits increase with the level of underpricing and are in general more profitable for underwriters of successful IPOs. A similar argument is made in Franzke/Schlag (2003: 20 et. seqq.) when considering the effects of overallotment option in Germany’s “Neuer Markt”. They find no evidence that this support mechanism reduces the amount of underpricing or lowers the total costs of going public for issuers. Furthermore, the results do not support the conclusion that an underwriter with the function of a market maker pays off for investors by generating a protection against price drops in the aftermarket. Also, the performance of a supported IPO does not seem to be significantly better during the first month of trading. As there are no remarkable benefits for issuers or investors, the results lead to the provocative hypothesis that overallotments are just another mechanism used by investment banks to allocate attractive shares to favorable clients. III.2 Litigation Risk Another explanation for underpricing is assumed in the function of reducing potential legal liabilities of the firm after going public. In the USA the Securities Act of 1933 implies that the purchasers of an issue can sue every person who has signed the offering’s registration statement, if there has been false or misleading

40

THEORY AND EVIDENCE OF IPO UNDERPRICING

information about the firm in the prospectus. The maximum damage that is recoverable for the claimant is limited by the offering price to the public. Lawsuits are very costly in terms of legal fees, management time and losses in reputation capital for issuers and underwriters. A formal analysis by Hughes/Thakor (1992: 711, et. seq.) assumes that firms and intermediaries intentionally underprice to reduce legal liability and related costs. The underwriters’ a priori reputation, and the longterm performance of an issue especially, influences investors’ beliefs about whether an issue is fairly priced or not. Intentional overpricing increases the likelihood of legal actions. The main implications of the model are as follows: Lower underwriters’ compensation schedules increase the level of underpricing, because compensation for the underwriter and the probability of litigation increase in the issue price. Secondly, higher uncertainty about future cash flows increases underpricing, while more prestigious underwriter reputation reduces the initial returns. In Tinic’s (1988: 805 et. seqq.) empirical investigation of the insurance mechanism of underpricing, similar propositions are tested with two samples; one before and one after the Securities Act of 1933. Evidence is found that underwriters with superior expertise and high reputation capital avoid highly speculative IPOs and underprice less, especially in hot issue periods. Also, underpricing is higher after 1933 with an average of 11.6% compared to 5.17% before the Securities Act, but the number of court decisions against issuers and their agents also increased steadily over the years. A more recent study of Lowry/Shu (2002: 323, 331) finds strong support for the insurance effect. Firms with higher ex-ante litigation risk set lower offering prices. Furthermore, firms which underprice more reduce their litigation risk. In this sample, the occurrence of litigation processes after the IPO are not examined, but more the ex-ante probability of litigation as an endogenous variable and its implications. Drake/Vetsuypen (1993: 71) note contradictory results: greater first day returns do not reduce legal liabilities. The relatively low magnitude and probability of ex-post liability figures also suggest that underpricing would be an expensive form of insurance. The decision to take legal action seems to be driven by declining prices in the aftermarket rather than by initial overpricing of the issue. However, in this study only the frequency of ex-post lawsuits are taken into account, not the ex-ante risk of litigation as in Lowry/Shu (2002). The lawsuit avoidance and insurance function seems to be less constitutive in other countries than in the USA, as concluded in Loughran/Ritter/Rydqvist (1994:174). Hopp/Dreher (2007:21, 25) also compare the legal environments of different countries and their effects on underpricing. One determinant is the availability of class action lawsuits, which are more common in the USA, represent more plaintiffs and result in higher costs in monetary terms and reputation losses. Other measures are introduced for procedural difficulties in recovering investors’ losses due to misleading statements in the offering prospectus through civil liabil-

THEORY AND EVIDENCE OF IPO UNDERPRICING

41

ity cases, as well as for criminal sanctions for the firm and certifying intermediaries, such as underwriters and auditors. However, none of these variables show a significant correlation with the level of underpricing. Additionally, no significant differences in civil and common-law countries can be found, in terms of initial returns, to reduce the litigation risk. III.3 Company’s Ownership Structure In a model proposed by Stoughton/Zechner (1998: 49 et. seqq.) the way that different IPO mechanisms affect share ownership is analysed, when underpricing and rationing are assumed to be rational responses to regulatory constrains of allocation and pricing of issues. In this context, investment bankers are able to differentiate between two investor groups: small and large investors. Only large or institutional investors can apply costly monitoring activities to a firm, which increase with the ownership share and raise the expected firm value. Small shareholders are expected to free-ride on the benefits of monitoring. Due to the limitations of price discrimination between different investors, the underwriter extracts the demand schedule of a potential large investor, and in response sets the optimal offer price to ensure effective monitoring. Finally, shares are allocated to small investors and rationed when their demand is too high due to low offering prices. Underpricing gives an additional incentive to large investors and is positively related to the abilities of strategic rationing. However, with higher monitoring costs, the investment banks value the “risk-sharing” function of small investors more, and they will ration less. Underpricing therefore is negatively related to monitoring cost, although monitoring increases the value of the firm and thereby offsets lower initial offering prices to some degree (Stoughton/Zechner (1998: 58 et. seqq.)). Brennan/Franks (1997: 393, 403 et. seqq.) come to a contradictory conclusion. They presuppose that the existing management wants to retain control of the firm after the IPO and aims to protect private benefits. A low offering price leads to oversubscription of the issue, discrimination against large applications and, in turn, to dispersed ownership patterns. This induces lower monitoring of outside investors and also prevents hostile takeovers of the firm. Additionally, the smaller the initial owners’ holdings, the more vulnerable they are to the loss of private benefits and therefore the greater their incentive to underprice. The empirical results confirm the hypotheses: oversubscription is an increasing function of underpricing. Furthermore, rationing and discrimination against large applicants are significant determinants in the allocation of shares. Evidence for the motivation to retain control and prevent takeovers is found in Boulton/Smart/Zutter (2006: 13, 17): Active markets for corporate control, measured with merger and acquisition (M&A) activity, are associated with greater underpricing. Here, the results also

42

THEORY AND EVIDENCE OF IPO UNDERPRICING

show that high initial returns lead to dispersed ownership, whereas larger ownership of institutional investors after an IPO increases the probability of a takeover. Weaker evidence is found by Field/Sheehan (2004: 270 et. seqq.). Already, about 83% of firms have large outside blockholders before going public. Additionally, there are no significant statistical differences, in terms of changes in ownership structure before and after the IPO, between firms which underprice and those that do not. For German IPOs, Petersen (2007: 32 et. seq.) shows that the control of outside pre-IPO shareholders has no influence on offering prices. Only the previous control of inside shareholders and their marginal costs of underpricing determine the level of initial returns. For this reason, owners seem to deliberately underprice where benefits are weighted to related costs. Fischer (2000: 22) shows that residual or outside blockholders sell proportionally more at the IPO, and significant rationing leads to an average allotment of 3.8%. Thus shares within an initial offering are widely dispersed and remaining shareholders gain control. Zingales’ (1995: 426 et. seqq.) theory also shows that firms prefer dispersed ownership within the IPO, but only because they can extract a higher surplus from negotiation with potential buyers afterwards. In contrast to Brennan/Franks’ (1997) theory, the process of going public is the result of the value maximizing decision made by initial owners, who want to sell the firm eventually. Booth/Chua (1996: 294 et. seqq.) argue that issuers demand a dispersed ownership structure and a broad investor base after the IPO to ensure high secondary market liquidity. A similarity to Brennan/Franks (1997) is that underpricing attracts potential investors, except that in this model they are compensated for incurring pre-bid information costs, which leads to oversubscription and dispersed allocation of shares. The insurance of liquidity is more precisely modelled in Boehmer/Fishe (2000: 3). Issuers want to allocate shares to investors who choose to hold the shares for longer periods and thereby reduce liquidity, as well as to “flippers”, who intend to sell the shares after a 1-2% increase in price and thereby lower aftermarket performance. By allocation to both investor groups, liquidity problems can be solved. Flipping investors value the issue less than long term investors, who assume a high firm value, but are restricted in share allocation, and therefore these groups have an incentive to trade. To induce investors with lower valuation to accept the shares, the issues have to be underpriced, on average. The underwriters can profit from an active trade in the aftermarket, when becoming a market maker. The empirical results from Boehmer/Fishe (2000: 31 et. seqq.) confirm that underwriters have an incentive to underprice in order to allocate shares to different investors and increase their aftermarket trading revenues. Furthermore, Ellul/Pagano (2006: 382 et.seq., 413) examine the effects of expected liquidity and related uncertainty about liquidity on the level of underpricing. They suggest that flippers demand

THEORY AND EVIDENCE OF IPO UNDERPRICING

43

high initial returns as a compensation for trading cost and as a premium for the liquidity risk they face when participating in an offering. The relationship between underpricing and aftermarket liquidity should be stronger in markets with several short-term investors. The sample includes IPOs from the UK, where aftermarket price support by underwriters is less pervasive than in the USA, rendering Boehmer/Fisher’s (2000) arguments negligible. The empirical results confirm their model: high expected liquidity and related liquidity risk in the secondary market are associated with high levels of underpricing. Brau/Fawcett’s (2006: 416) survey considers the determinants from a chief executive officer’s (CEO) perspective and confirms that a wide base of ownership and a desire to increase ex-post trading volume are the main reasons for underpricing in initial offerings. III.4 Behavioral Finance Another interesting explanation is found in Loughran/Ritter (2002: 423 et. seqq.). They introduce aspects of behavioral finance as potential determinants for the issuer’s decision to underprice. Similarly to the argumentation of the prospect theory proposed by Kahneman/Tversky (1979), they assume that issuers care more about the change in wealth than the level of wealth. Those owners will sum the wealth loss due to underpricing with the larger wealth gain on the retained shares from a price jump at the first trading day, so they benefit from a net increase in wealth. Particularly when the offering price is revised upwards, they receive a higher share value than expected and accept higher underpricing. With a downward revision, issuers are more demanding and restrictive in the amount of money left on the table.11 Additionally, underpricing is indirect compensation for underwriters, as it is easier to find investors and those who will engage in rent-seeking behavior to improve their priorities of allocation for future favorable offerings. This is valuable for underwriters, because issuers seem to treat the opportunity cost of underpricing differently from direct fees for investment banks when going public. Ljungqvist/Wilhelm (2005: 1764 et. seq.) empirically support this argument by investigating the issuer’s degree of satisfaction with the underwriter’s performance. They consider the level of wealth loss due to underpricing and wealth gain due to revaluation of retained shares relative to the expected reference value. The issuer’s satisfaction is measured by the decision to choose the underwriter for subsequent securities market transactions. The results (Ljungqvist/ Wilhelm (2005: 1786 et. seqq.)) confirm that underpricing alone is not the major determinant for switching to another underwriter. Besides underwriter reputation and analyst coverage, CEOs are less likely to switch their underwriting bank when 11

For further evidence of the relation of price revision and underpricing see chapter I section II.2.2 and Hanley (1993).

44

THEORY AND EVIDENCE OF IPO UNDERPRICING

they perceive a significant wealth gain. Furthermore, investment banks can profit from higher underwriter spreads in SEOs when the issuers are satisfied with their initial public offering. Another argument is found in the prediction that investors just evaluate the same information differently, so that estimates of firm value do not depend on investors being either informed or uninformed. In this context Gouldey (2006: 40 et. seq.) states that a bookbuilding process only yields an estimated probability distribution of the investors’ intrinsic value of a firm, but cannot reveal a priori the complete aggregate demand function and the stock’s equilibrium secondary market price. Especially when an issue is oversubscribed with a constrained number of shares, underpricing is unintentional, due to the underwriter’s and issuer’s risk aversion of a declining secondary market price when the offer price is set too high. Greater divergence of investors’ valuation, and higher uncertainty about the number of investors, makes it more difficult to predict aggregate demand, therefore underpricing is likely to increase. Another model put forward by Ljungqvist/ Nanda/ Singh (2006:1689 et. seqq.) states that so-called sentiment investors are irrationally exuberant about the prospectus of IPOs resulting in incorrect beliefs about the firm’s value. Issuers try to allocate shares to regular, more rational investors. They can gradually sell these shares to sentiment investors who arrive in the market over time. Underpricing therefore compensates regular investors for losses from the sentiment’s demand decline, because they expect that overvaluation may come to an early end. An increasing difference in the opinions of the two investor groups requires higher underpricing, for compensation. The greater the bargaining power of an issuer compared to the underwriter, the higher the offer price and the lower the first day return, whereas more prestigious underwriters are associated with higher underpricing. Issuers and underwriters take advantage of the market’s misperception, so that the offer price can exceed the fundamental value by the amount of the issuer's surplus subtraction of sentiment investors. With increasing optimism of sentiment investors and higher issuers’ proceeds, more companies have an incentive to go public, which results in hot market periods. Over time, regular investors unload their excessive inventory, especially from those issues which are expected to perform poorly in the future. The empirical implications are partly tested by Cornelli/Goldreich/Ljungqvist (2006: 1197 et.seqq.) with Europe’s pre-IPO markets, because the participants are considered as small, retail investors. When these small investors are mainly driven by irrational value expectation, large, rational investors can take advantage of the grey market implications and sell their inventory to sentiment investors for higher prices in the aftermarket. They find evidence that aftermarket prices are positively related to grey market prices, especially when pre-IPO prices are high. Higher initial offering prices are

THEORY AND EVIDENCE OF IPO UNDERPRICING

45

also found when small investors are considered as overoptimistic: Grey market prices are higher than the midpoint of the filing range. Furthermore, when investors fear that they will not be able to sell allocated shares in the aftermarket due to a larger issue size, the offering prices are set more conservatively. For Germany, evidence from the pre-IPO market is found by Dorn (2003: 18, 22, 25). The results show that pre-IPO purchase volume is higher in large issues with a strong media presence and following months with higher returns by the performance index of Neuer Markt (NEMAX). These relations are consistent with retail investors buying stocks that grab their attention. The determinants for demand in the aftermarket are similar: High media presence and high average underpricing prior to the IPO are associated with high retail purchases. The volume of retail purchases is also higher within the most underpriced issues, while their flipping activity is moderate. As other studies have found that flipping normally increases with underpricing, the result confirms that especially hot issues are passed from large rational investors to retail investors in the aftermarket. III.5 Information Momentum In a model complementary to the prospect theory, Aggarwal/Krigman/Womack (2002:109, 111) show that substantial underpricing attracts analyst coverage and media attention, which in turn attract investors and shift out the demand curve for stocks. As potential investors rely on this outside information, issuers can take advantage of the information momentum effect and strategically choose a lower offer price. The optimal offering price enables the owner to sell the shares for higher prices after the lockup period, instead of maximizing proceeds or long-run firm value. In the empirical investigation, Aggarwal/Krigman/Womack (2002: 117, 123 et seqq.) use the timing and quantity of research recommendation and comments on the IPO, the number of broker comments, and the number of the firm’s mentions in First Calls as proxies for information momentum. They also find evidence for the hypothesis that managers who retain more shares also underprice more, which generates more information momentum. Increased research coverage leads to higher stock prices at the expiration of the lockup period. Furthermore, the number of shares sold at this point increases in research coverage. For Germany, Bessler/Kurth (2007: 48) consider the analyst recommendations of banks, which are also in the position of pre-IPO shareholders in the issue. The firms which receive strong buy recommendations of the invested banks also perform significantly better during the first 6-month than other IPOs. However, after the first year of trading the difference diminishes. It can be assumed that banks can support stock prices and exit after the lockup period at very favorable prices. They are able

46

THEORY AND EVIDENCE OF IPO UNDERPRICING

to take advantage of private information at the cost of other investors. The issuer, on the other hand, is only interested in the bank’s analyst ranking. Loughran/Ritter (2004: 9, 25) find that, especially in the bubble period between 1999-2000, underpricing is 21.2% higher with a top-tier lead underwriter providing analyst coverage. In general during the 1990s, with an increasing importance of “analyst lust”, issuers were more willing to accept underpricing, while in the 1980s the relationship was negative. In the post-bubble period the underwriter’s analyst ability and influences are not an explanatory variable for underpricing. Also in Cliff/Denis (2004: 2880, 2883, 2889) the results show that underpricing is positively related to the analyst coverage of an issue. In this sample (1993-2000), initial returns are 9% higher in IPOs with a lead underwriter who has an all-star analyst covering the issuer’s industry. Underwriters seem to agree to provide recommendations for issues, which accept higher underpricing. Additionally, non lead-underwriters are also expected to provide analyst coverage of deals which have higher initial returns. Issuers are also likely to switch underwriters for their SEO when they are not satisfied with the lead analyst coverage received during and after the IPO process. Dunbar (2000: 22, 24) confirms that this is important for the intermediaries future underwriter market shares. Also Rees (2003: 19 et. seqq.) considers information momentum, measured by newspaper citations of the issuer. It is shown that pre-IPO investor’s interest leads to higher initial return and trading volume. Higher underpricing in turn results in more post-IPO information production and long-term trading volume of an issue.

IV Discussion It has been shown in the sections above that some theories are refuted, while others are partly confirmed by empirical evidence related to several countries. One possible reason for this is the difference in sample periods, especially concerning hot and cold markets, which determine market conditions as well as IPO characteristics. Both influence the level of underpricing and the explanatory power of the variables. Also several proxies, or measuring methods, are important factors and have to be taken into account when comparing different empirical studies. The difficulty here is in determining the most suitable methods for the examination of underpricing. However, the evidence for theories which consider underpricing as a signalling method, to induce a separating equilibrium between high and low quality firms, is weak. Other factors are better appreciated by the market participants and lead to

THEORY AND EVIDENCE OF IPO UNDERPRICING

47

higher dividend payments or SEOs. Only the participation of previous owners and the issue size of primary shares relative to pre-IPO shares are negatively correlated to the level of underpricing. Moreover, the certification function of intermediaries is not consistently confirmed. A firm’s auditor, as well as earnings and disclosure transparency, seem to reduce the asymmetries between issuer and investors. In comparison, the function of VCs is more questionable and could be analysed for further research. VCs are presumed to signal the firm’s quality, but they also want to exit from their investment with high profits. The different objectives can probably be determined by considering the VC’s position in an IPO in more detail. The VCs providing seed financing over several years behave differently from VCs financing projects at later stages or providing capital to the firm just for going public (bridge financing). This is important for the level of underpricing, due to different investment decisions and expected profits from the exit strategy. It is also interesting to consider overall M&A activity in relation to IPO underpricing. The trade sale of a VC-backed firm is also an important exit channel, and so the market conditions in M&A transactions are likely to influence both the decision to go public and the accepted level of underpricing by VC firms and owners. In this context, the age of the VCs and the distribution of control rights between decision makers are interesting; unfortunately this is more complicated to determine. The role of investment banks is also not clear. When banks are involved in a lending relationship or providing equity to the firm, due to relationships with VC companies, their function and their intentions could cause conflicts, especially when they are chosen as underwriters. On one hand they are supposed to provide some certification of quality for investors; on the other hand they can act to maximize their own profits. The theories based on Rock’s (1986) “winner’s curse” model predict favorable treatment of institutional or regular shareholders. Often a long-term relationship exists between investment banks and their clients, who act as investors in the IPO. They are supposed to give some indication of demand and perception of the potential success of an issue. For example, in pilot fishing the intention is to receive favourable information even before the bookbuilding process. It can be assumed that underwriters aim to reward their clients instead of the issuing firm. Favorable investor relationships could be important for unsuccessful issues and for the future underwriter business, when the choice of investment bank is made by a client. Underpricing is a method to find investors easily and satisfy them with high initial returns, thereby enabling underwriters to make future profit. However, this can also be considered from another perspective: Demand from regular clients can determine the level of underpricing. For example, when managers of mutual funds participate in IPOs, they are expected to consider the related uncertainty and risk

48

THEORY AND EVIDENCE OF IPO UNDERPRICING

of the issue in their portfolio context. Sometimes shares with high expected volatility can be sold easily without excess initial returns, because of their diversification benefits. In hot issue markets, many firms in the same industry go public during a short period of time, but they are not necessarily interesting from the portfolio management aspects because of similarities in firm characteristics. Clients therefore demand higher underpricing before they will participate in the offering and demand shares. This could possibly provide an answer for Ritter (1984), who finds that ex-ante uncertainty about the firm value is not the only determinant for underpricing in hot issue periods. The explanation of average initial returns due to underwriter price support in the secondary market is not really confirmed. In contrast, it can be argued that underwriters, as market makers, do indeed profit from underpricing. Complementary to the overallotment option, the price stabilisation activity can lower the risk and related costs to the investment bank. Higher underpricing therefore should be expected in situations with more market maker activity by underwriters. Additionally, initial returns are likely to encourage analyst recommendations and research comments about the IPO, which in turn attract more investors and increase liquidity. In this context, underpricing seems to support share prices after going public, whereas no conclusion about the quality of the firm can be derived. With these explanations of underpricing, the related long term performance can be considered. Most empirical investigations show that IPOs underperform stock returns of comparable public firms in the long run (3 to 5 years).12 It is also interesting how underwriters and large regular investors can profit from small investors. Those participants are partly driven by irrational expectations, and partly by information momentum. Once again it becomes obvious that underpricing is an instrument for investment banks and their regular clients to profit from inexperienced issuers as well as from small investors. In this context it would be interesting to develop a strategy, which could realise profits from initial returns and thereby reduce the benefits of underpricing for investment banks. The main problem is the underwriter’s ability to ration an issue, which does not guarantee participation and therefore diminishes the success of strategies on underpricing.

12

See for example Ritter (1991) or Burghof/Kraus (2003) for German IPO statistics. However, the long-term underperformance of IPOs is not subject to this dissertation and the explanations for IPO underpricing are not related to the divergent empirical findings for the long term price performance of public firms.

THEORY AND EVIDENCE OF IPO UNDERPRICING

V

49

Conclusion

The country comparisons of USA and Germany are difficult in some parts, due to the lack of empirical investigation of the German market. However, differences in underpricing determinants are found to be partly due to different market conditions. For example, the German empirical studies mainly include samples during the hot issue period of the Neuer Markt, which has been associated with severe underpricing. This means that rational arguments for or against underpricing could not be confirmed. However, in both countries the evidence for the function of underpricing as signalling the quality of the firm is weak. Only the participation ratios of previous owners and the dilution ratio due to primary shares have significant explanatory power, also during hot issue periods. Certification of the firm’s quality of a venture capital company also does not reduce the level of underpricing. Even in Germany during the Neuer Markt, these intermediaries are associated with higher initial returns. Also, the strong banking relationships in Germany cannot be interpreted as an opportunity to signal quality and long term monitoring of a firm: Compared to the USA, the potential exploitation of outsiders and conflicting interests are highly weighted. Also, in both countries the underwriter’s reputation has become less important in reducing the level of underpricing. It is assumed that other functions, such as analyst quality and recommendations, have become critical in the choice of investment banks. Increased competition between banks may, especially in Germany, influence intermediaries’ positions. However, underwriters’ methods of pricing an issue and allocating shares to potential investors do not determine the level of underpricing. Fixed-price offerings and bookbuilding with a best-effort or firm-commitment contract show no significant pattern for the level of initial returns. In this context the mechanism of pilot fishing and the implications of the grey market are relevant in Europe in revealing preliminary opinions about the success of an issue, and are partly included in the offer prices. The underwriter’s activities after the IPO, in terms of price support or the exercise of the overallotment options, seem to benefit the banks in both countries instead of reducing the level of underpricing. Also, a dispersed ownership structure and higher liquidity in the secondary market are more preferable. The participation of small or retail investors driven by irrational expectations, in particular, can be exploited, and also account for high levels of initial return during hot issue periods. Surprisingly the legal environment in USA and Germany does not determine the amount of money left on the table to satisfy investors and to reduce potential liability cost. The results of this chapter show some opportunities for further research. The conflicts between participants in an IPO and the allocation to investors are still an interesting aspect to investigate. Furthermore, the period after Germany’s Neuer Markt should also be considered in more detail, therefore the fol-

50

THEORY AND EVIDENCE OF IPO UNDERPRICING

lowing chapter II investigates agency conflicts and monitoring incentives of IPO participants in hot and cold market cycles.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

51

Chapter II: How Do Pre-IPO Shareholders Determine Underpricing? I

Introduction

One strand of literature has focused on asymmetric information distribution between the participants of the going public process. Divergent information, intentions and resulting agency conflicts between issuer, underwriter and investors are considered to explain initial returns after the first trading day of newly issued shares. Ljungqvist/Wilhelm (2003) summarized previous approaches and hypothesized that the ownership structure of a firm before the IPO is decisive in the level of underpricing. Dispersed ownership of management, financial investors or other shareholders results in higher agency conflicts between pre-IPO owners, and reduces their incentives to monitor the underwriter’s setting of an optimal offer price. They found empirical evidence for this theory before and during the dotcom bubble (1996-2000), which can be considered as the latest hot IPO market period. Also in Germany, average underpricing reached enormous levels during the active period of the Neuer Markt between 1996 and 2001. After the bursting of this bubble in 2001, the IPO volume and average underpricing decreased sharply and remained low in the following years. This leads to the interesting research question, if ownership structure and potential agency conflicts determine IPO underpricing in the same way in extremely different market environments. It is hypothesized that IPO participant’s interests and bargaining incentives change according to IPO market conditions and therefore determine initial returns in distinct ways. Therewith, clustering of pre-IPO shareholders’ stakes is expected to become less relevant explaining the development of underpricing and the public equity market. The results of the empirical investigation with data from the German stock exchange between 1997 and 2007 support the hypothesis. The data from Deutsche Börse AG enables to investigate firms with traditionally high clustering of ownership stakes compared to the USA. The limited abilities of offer price revision within the German IPO process give further interesting insight to shareholder’s bargaining willingness about the optimal offer price. The ownership structure of firms going public has changed during the 11 year period and in different IPO market phases, but does not indicate less dispersed ownership in periods with low average underpricing. Also the determinants of initial returns differ in the sub-

52

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

samples of “hot” and “cold” market periods. Evidence is found that invested financial intermediaries and partly firms’ insiders change their willingness to “leave money on the table” under different market conditions. So this paper contributes to the existing literature about IPO underpricing in accepting previous findings for positive market environments, but also rejecting the general validity of previous theories. The differences of pre-IPO firm’s ownership structures according to the market phases have not been investigated before, allowing this paper to contribute some general ideas to the reasons behind market cycles. To investigate firm’s ownership structure as a determinant of underpricing, the chapter II is structured as follows: Section II analyses previous literature and related theories. Furthermore, the hypothesis regarding the shareholders incentives and the level of underpricing is developed. In section III the research design of this paper is considered, and the proxies discussed are linked to previous studies, and section IV presents the empirical results. First, descriptive statistics about the firm, transaction and ownership characteristics of the IPOs for hot and cold market periods; second, the results of the multivariate regression models are presented to demonstrate support for the hypothesis. Section V concludes and summarizes the most important results.

II Related Literature and Development of Hypothesis One strand of literature explaining IPO underpricing concentrates on the asymmetric information distribution between the participants of the going public process: issuer, underwriter and the investors. The early theories developed by Allen/Faulhaber (1989), Welch (1989) and Grinblatt/Hwang (1989) predict an informational advantage on the part of the issuing firm, compared to investment banks and potential investors. In these models the decision to underprice the offered shares is a deliberate choice by the issuer, to signal the firm’s quality. Only high quality firms are able to “leave money on the table”, because they expect positive capital market reactions on future dividend announcements or higher offer prices at SEO. These future pay-offs cannot be expected by low quality firms, so underpricing would be too costly. The empirical results relevant to these theories are ambiguous (e.g., Michaely/ Shaw (1994: 305 et. seqq.), Jagadeesh/ Weinstein/ Welch (1993: 173 et. seqq.)). Also, Barry (1989: 1101 et.seqq.) states that underpricing becomes more costly for owners the more they sell of their original holdings (secondary shares) and the more new shares are offered (primary shares) in the IPO. High participation of pre-IPO shareholders and high dilution due to an

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

53

increase of shares result in low initial returns. Evidence is found by Habib/Ljungqvist (2001: 449), Ljungqvist (1997: 1316) and by Giudici/Roosenboom (2002: 22) in Europe during the dot-com bubble. Petersen (2007: 8 et.seqq.) tests the theory of underpricing as a deliberate choice by a group of pre-IPO owners for the German IPO market between 1997-2002. He considers the trade-off between costs and benefits of underpricing and finds evidence for the hypothesis. When insiders, including supervisory and board members, managers and their families, are in control of the firm, there is a positive relation to the level of underpricing. However, issuers might be able to reduce the level of underpricing if a third party is involved to certify the quality of the firm. For example venture capitalists (VCs) have the opportunity to monitor a firm over a long period of time prior to the IPO and to give managerial advice. Megginson/Weiss (1991: 880 et. seqq.) developed the theory that this is a signal of quality, which can be recognized by potential investors. Contradictory evidence is found during the last hot issue period: the level of underpricing in VC-backed IPOs did not differ from that in non VC-backed firms (e.g., Bradley/Jordan (2002: 613)). Also Franzke (2003: 20) and Bessler/Kurth (2007: 39) note a positive relation between VC ownership and initial return during the Neuer Markt in Germany. One explanation suggests that VCs are exit driven: A fast exit from the financed firm provides the VC with resources to either disburse fund investors or to reinvest in new promising projects. Another reason is known as “grandstanding”, where young VCs particularly bring firms to the public equity market to build up a positive reputation, although the firms are still associated with high uncertainty about future profits. The position and objectives of the underwriter can also be discussed from different perspectives. Booth/Smith (1986: 262 et. seqq.) argue that they can also certify the quality of the issuing firm. However, Baron (1982) proves that underwriters with informational advantage over the issuer choose lower offer prices, to find investors and distribute shares more easily. A dispersed ownership structure after the IPO also ensures higher liquidity and increases the underwriter’s trading revenue, when becoming a market maker in the secondary market (Booth/Chua (1996: 294 et. seqq.), Boehmer/Fishe (2000: 29 et. seqq.), Ellis/Michaely/O’Hara (2000: 1060)). Also Schultz/Zaman (1993: 204 et. seq.) or Aggarwal (2000: 1082 et. seqq.) show that underwriters often provide price support in the secondary market by repurchasing the newly issued shares and exercising overallotment options. With lower offering prices the implicit costs of price support decrease. Besides this, some studies suggest that underwriters use underpriced issues to favor some of their clients (e.g., Aggarwal/Prabhala/Puri (2002: 1427 et seqq.), Loughran/Ritter (2004: 11 et. seqq.)). Reuter (2006: 2307 et seqq.) finds evidence that

54

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

mutual funds receive favorable allocation of underpriced shares when they provide brokerage commission payments to the underwriter in turn. The association between brokerage payments and the IPO holdings of mutual funds are highest for IPOs with initial returns greater than 20%. However, in most IPOs the investment bank’s compensation for the underwriter business is given as a percentage of gross proceeds.13 Therefore higher offer prices and associated higher gross proceeds would result in higher underwriter fees, so underpricing also comes to a cost to the investment bank. Based on the theories of informational asymmetries and divergent interests of the issuer and the underwriter as well as between pre-IPO shareholders, Ljungqvist/Wilhelm (2003: 724 et. seq.) developed the theory that a fragmented ownership structure could explain the abnormally high underpricing seen in the dot-com bubble. Board members or the CEO have fewer incentives to bargain over the offer price and monitor the underwriter when their stakes in the transaction are smaller and other pre-IPO shareholders have a different focus of interest. Also, more different shareholder groups are subject to “moral hazard in teams” problems, which result in lower monitoring and higher underpricing when going public. Additionally, the selling behavior of the pre-IPO owners is important, as low participation ratios make underpricing less costly and therefore the incentive to negotiate over higher offer prices less immediate. Ljungqvist/Wilhelm (2003: 729 et. seqq.) found evidence for this hypothesis in their empirical study of the development of ownership structure and selling behavior in their effects on underpricing with a sample period from 1996-2000. Giudici/Roosenboom (2002: 17 et. seqq.) developed a similar study for Europe during the period from 1995 to 2001. They found evidence that, in Europe, firms are more often closely held and managed by their founders than in the USA. Also, increasing stakes of different owners are negatively related to the level of underpricing, as the incentives to bargain over the optimal offer price increase. Both papers consider IPOs before and during the boom of the new economy, but lack any evidence of the following years where an enormous downturn of the market could have been recognized. Not only the IPO volume but also the level of underpricing decreased after the bursting of the dotcom bubble. In Germany the mean value of IPO initial returns in this sample dropped from 42.8% (median: 9.99%) during 1997-2001 to 4.6% (median: 1.2%) between 2002 and 2007. With these extreme IPO market changes, the intuitive question comes up, if the ownership structure and potential agency conflicts determine IPO underpricing in the same way in the years following the dot-com bubble. Three different scenarios are possible: The firms which chose to go public 13

See for example Chen/Ritter (2002: 1108 et.seq.), who find evidence that in the USA the spread on IPOs is highly clustered at about 7%.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

55

in the period following the Neuer Markt, could show different ownership structures. They are, presumably, more developed, with dominating pre-IPO shareholders or, for example, fewer venture capital investors. On the other hand, the level of initial returns could simply have shifted downwards for all IPOs, with the explanatory factors and ownership structures remaining the same. Furthermore, it is possible that participants’ interests and the resulting agency conflicts have changed according to the market environment, and so affect underpricing differently. The investigation focuses on German IPOs between 1997 and 2007. In general, firms are held much more closely by a concentrated shareholder group in Germany than in the US, so the investigation of changes in ownership should give very important insights to the pre-IPO agency conflicts in firms. The chapter focuses on the question of whether ownership explains or determines IPO underpricing in every market phase in the same way. The shareholders’ stakes and selling behavior are analyzed during different market conditions to show whether e.g. insiders, financial investors or other blockholders promote the firm’s IPO and offer price. The market environment is categorized into “hot” and “cold” periods according to monthly IPO volume and previous underpricing. It is hypothesized that pre-IPO owners’ willingness to leave money on the table changes according to the market phases. Their bargaining interests and intentions are likely to differ, so that ownership which may not necessarily be dispersed explains agency conflicts and the level of underpricing. Loughran/Ritter (2002: 424 et. seqq.), for example, applied a model of prospect theory to issuer behavior. They state that pre-IPO shareholders’ wealth gains due to a previous upward revision of the offer price are integrated into the money left on the table. The higher the previous unexpected wealth increase, the more issuers are willing to accept initial returns for investors participating in the IPO. Previous positive market return is not fully incorporated in higher offer prices, so underwriters can combine positive information of market development with relatively more money left on the table. Issuers are less concerned about lower than possible proceeds, as they still experience a wealth gain due to a slight upward revision of the offer price. If overall market return decreases before the IPO, underwriters are less likely to revise the offer price, so firm owners do not experience any wealth gains and are more likely to bargain over the optimal offer price and the amount of money left on the table. Additionally, Ljungqvist/Nanda/Singh (2006: 1671 et. seqq.) and Derrien (2005: 490 et. seqq.), show that underwriters and issuers are able to time the IPO to take advantage of highly positive market perception. With increasing optimism on the part of market participants, more firms have an incentive to go public, profiting from the relatively low costs of issuing equity, and so are less concerned about the amount of money left on the table. To refer back to pre-IPO ownership structure, a positive mar-

56

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

ket environment in hot periods is expected to reduce the bargaining incentive of the various shareholder groups. This market phase allows greater flexibility in setting the offer price, which could also increase potential agency conflicts due to differing interests among pre-IPO shareholders. However, in cold markets, the groups of insiders, financial investors and other blockholders are expected to have more concerns about leaving money on the table. Furthermore, less flexibility by the market in terms of offering prices could reduce potential agency conflicts. In summary, the owner’s interest in the optimal offer price and willingness to leave money on the table should change with the market phases. This could also explain the various empirical results presented above for managers signaling the firm’s quality or the certifying function of venture capitalists. Many studies on IPO underpricing have considered the effects of different owner groups, such as banks as shareholders (e.g. Tykvova/Walz (2007: 365) Slovin/Young (1990: 736), Klein/Zoller (2003: 10)) during the period of the Neuer Markt in Germany. Their findings suggest that different ownership before the IPO increases agency conflicts between issuers, underwriters and investors, resulting in higher underpricing. However, another explanation could be that the pre-IPO shareholders were simply less concerned about the money left on the table because the positive market environment provided little incentive to bargain with underwriters about higher offering prices.

III Research Design III.1 Sample Selection and Data Sources Between January 1997 and December 2007: 595 IPOs took place at the Frankfurter stock exchange, which is run by Deutsche Börse AG. This represents the most important stock market in Germany, covering about 90% of the equity market in Germany. This analysis concentrates on this stock exchange because the availability of data from minor exchanges in Germany is limited. Besides the Regulated Market (General and Prime Standard), which is based on public law and requires the highest transparency requirements of the European legislator, the Open Market (Freiverkehr) is an important segment of the German capital market, especially for small and medium-sized firms. The Neuer Markt has been a sub-segment of the Open Market between 1997-2003. In 2005 the Entry Standard has been introduced as the “successor” segment, which is also based on private law, and firms’ shares are traded with lower transparency requirements. The sample includes all initial public equity issues, while private placements and the transfer from one stock exchange or market tier to another are excluded. Also IPOs from Banks and Reits due to differences in financial accounting statements are not taken into considera-

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

57

tion. The chapter’s focus is on pre-IPO ownership structure, so only firms with this information available are included in the sample. Furthermore, initial public offerings are only analyzed, when their offer price has been determined with the bookbuilding process. Finally, the sample consists of 439 IPOs between 19972007. This still represents a large sample for the German market. Previous studies mainly investigated the Neuer Markt and therefore considerd only up to 350 IPOs. Deutsche Börse AG provides information about all offerings in terms of new issues, listings, and exchange transfers. The primary market statistics, available on their website, provide information on IPO dates, offer prices, first prices at the beginning of trading, bookbuilding spans as well as the volume of the issues. The information about the structure of the offering in terms of primary and secondary shares and the size of the overallotment option are obtained from the IPO prospectus. Furthermore, the ownership structure before and directly after the IPO also has been determined with this source. Often additional research has been required for classification of ownership, because owners are often involved, directly or indirectly, with other shareholder groups or companies. Internet research provided some clarification, as well as the paragraphs in the issuing prospectus about the history and development of capital stocks. The firm’s financial and income statements closest to the IPO date are obtained from Reuters Knowledge to receive data on total assets, intangible assets, debt as well as sales or capital expenditure. Another important data source has been Thomson Financial’s Datastream. This database provides the closing price on the first trading day after the IPO and information on percentage price changes and historical volatility of all shares traded at German stock exchanges in the analyzed years. Also the Industry Classification Benchmark (ICB) for the sample IPOs are obtained from this database to determine the firms’ industry sector. III.2 Definition of Variables In order to assess if and how underpricing is determined by the ownership structure before and after going public, the empirical test is oriented on Ljungqvist/Wilhelm (2003) and several ordinary-least-squares (OLS) regression models are estimated to test the hypothesis. Underpricing is generally defined as the difference between the first day closing price and the offer prices, divided by the offer price. Previous studies applied an adjustment to index returns, e.g. NEMAX All Shares, FAZ-index or CDax, to account for market effects on price, in the time span when the offer price is set and beginning of trading (e.g. Wasserfallen/Wittleder (1994:1508), Ljungqvist (1997:1311), Hunger (2001:132 et. seqq.)). However, more recent studies suggest that it is more difficult to find an appropri-

58

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

ate benchmark which does not include the initial returns, but which reflects the industry and firm characteristics (e.g., Ohler/Rummer/Smith (2004: 14), Kurth (2005:313), Kraus (2002:10)). Moreover, Ritter (1984: 217 et. seqq.) states that adjustment by market movement only changes the results of initial returns slightly. Loughran/Ritter (2002: 417) also do not adjust the initial returns, because the return per day of their benchmark averages 0.05% and thus is assumed to have little impact on the conclusion. Also, for our sample period the stock market performance is not included in the calculation of initial returns, but as an explanatory variable. Moreover, to reduce the skewness and kurtosis of the distribution, a logarithmic transformation of underpricing is applied. The dependent variable (UP) is measured as the natural logarithm of underpricing plus one (similar to Hill (2006: 111)). The data on ownership prior to the IPO is collected from the IPO prospectus, and is categorized as relating to: insiders, financial investors and blockholders. The classification of “insider” includes all shares of members of the supervisory and management board, as well as private holdings and trusts these persons have an interest in and can influence the decision making process. Also, employees and family members of firm’s founders often hold shares before the IPO. These private persons are also included in the group of insiders. Overall, the classification takes into account the shares, which are most likely to exercise their voting rights in the interests of firm’s founders and management team. The investments in terms of venture capital, bridge financing for the IPO and private equity are all classified as “investor”, including where private equity of the respective fund has been provided in previous buy-outs. However, the classification is limited to firms and funds with the corporate objective related to private investments where this can be identified from their internet representation or company report. Therewith, firms registered at the German Venture Capital Association e.V. (BVK e.V.), European Venture Capital Association (EVCA), and the National Venture Capital Association (NVCA) are included. Also banks and affiliated corporations which are preIPO shareholder are classified in this group. Particularly in Germany, close banking relationships exist and banks often provide equity via related venture capital or private equity funds as well as being able to act as an underwriter. “Landesbanken”14, for example, often have more than one function in the process of going public. The description “blockholder” is used for shareholders who own more than 25% and are not included in other categories, e.g. company ownership. The limit is set to stakes of more than 25%, because according to German law this enables the shareholder to veto decisions in shareholder meetings or changes to the articles 14

“Landesbanken” are public-sector banks partly owned by German regional governments. They also undertake functions of universal banks.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

59

of incorporation. In the case of a different focus of interests from that of other owners, this shareholder is authorized, due to control and voting rights, to enforce objectives when going public. The owners’ stakes are calculated as their shares in relation to total shares outstanding before the IPO. Furthermore, to have an overall measure for clustering of ownership, the Herfindahl index is calculated, which is defined as the sum of squared pre-IPO ownership stakes of insiders, investors, and blockholders (Giudici/Roosenboom (2002: 18), Ljungqvist/Wilhelm (2003: 733)). The index ranges from zero to one: the value of zero indicates a highly fragmented ownership structure and the value of one indicates only one shareholder prior to the IPO. Other common variables in the underpricing literature are the proxies of “participation” and “dilution”, which capture wealth effects of pre-IPO owners and their potential interests in the offering. Wealth transfer from old shareholders to new ones in terms of initial return is higher when the number of offered shares is high, relative to shares previously outstanding. So dilution is defined as primary shares divided by the total of preIPO shares and is also expected to have a negative effect on the level of underpricing. The overall participation ratio of owners is calculated as the number of secondary shares divided by pre-IPO shares. The selling behavior of pre-IPO owners influence their perception of the amount of money left on the table, as well. As the owners sell more of the pre-IPO shares outstanding higher wealth losses are incurred as a result of underpricing. The effects of shareholders selling behavior and potential conflicts are also captured by measuring sales of each owner classification directly and therewith replacing the variable for participation. The shareholder groups’ sales are measured as the difference between pre- and post-IPO shares divided by the total of pre-IPO shares outstanding (similar to Giudici/Roosenboom (2002:42)). The statistical estimates are repeated with variables of pre-IPO shareholders’ stakes and sales. The regression model includes control variables for the effects of company or transaction characteristics on initial returns, which have been proven in previous studies. For example, higher gross proceeds at the IPO, which are calculated as the natural logarithm of the offer price multiplied with the number of shares issued, are associated with lower investors’ uncertainty about the issue and therefore with lower underpricing (e.g., Ljungqvist (1997: 1316), Löffler/Panther/Theissen (2005: 478)). Often, the standard deviation of daily returns after the IPO is used as a proxy to account for this relation. However, this variable would fail to allow for the uncertainty of market participants, especially in weak issues, because of price stabilization activities in the secondary market of underwriters in Germany. Furthermore, the industry classification is included in the analysis, as firms related to

60

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

the software, internet, media or technology sectors experienced extremely high initial returns in the dot-com bubble. For every sample IPO the ICB categorization is obtained and dummy variables “tech” and “media” equal one, if the firm is classified into these two respective sectors otherwise they are equal to zero. Additionally, market characteristics or publicly available information affect underpricing. For example, Hanley (1993: 246) and Loughran/Ritter (2002: 415) have shown that high market returns before the offer date are only partially included in the offer prices and firms going public in a positive market environment show higher levels of initial returns. Therefore, IPO underpricing is autocorrelated over time and becomes predictable to some extent, so that average cumulative return of all tradable shares in the German stock exchanges during 30 trading days before each IPO is considered as another variable (“return”). The prices for all shares traded at the German stock market between 1997 and 2007 are from Thomson Financial Datastream. This also includes the stocks, which have been delisted over time. So this measure is a better indicator than, for example, the CDax performance index, which only includes share prices of firms, which stocks are still traded at the Regulated Market. Because many firms have been delisted after the bursting of the dot-com bubble, this is likely to be a biased variable to control for public information and market performance in the past. Additionally, the variable “vola” measures the average monthly volatility of these shares in the month of the IPO. This variable is included to provide a control for the valuation uncertainty of the market. For example, Pástor/Veronesi (2005:1720) argue that more firms go public when uncertainty about the future profitability is high but also that this is likely to increase the market valuation of a firm. Instead of aftermarket volatility, previous share price changes are included in the analysis, which can also be considered as publicly available information to potential investors. Further indications of offer prices and initial return are derived from the volume and average underpricing of previous IPOs. For example, Ritter (1984: 219) shows that periods of high average initial returns tend to be followed by periods of high IPO volume. Also Lowry/Schwert (2002: 1171 et. seqq.) and Lowry (2003: 17 et. seqq.) confirm this lead-lag relationship, and suggest that firms are more likely to go public after periods of high initial returns, because increasing first day trading prices are associated with positive (private) information of investors, which are not fully incorporated in the offer price. As a result, firms find it more attractive to go public and consider the related costs as especially low. However, the time a company decides to go public does not necessarily indicate any information about that firm’s underpricing. Although empirical patterns have shown, that periods of high IPO volume are often followed by lower underpriced IPOs. One possible explanation is that more information can be captured in the offer prices over time. The explanatory variables of “volume” (number of IPOs in the month of the offering date) and

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

61

“IR” (initial return of IPOs in the month prior to the IPO) therefore take account of the information of previous IPOs available to the underwriters and investors. Also, in a study of Baker/Wurgler (2006: 1657) both indicators are used as proxies for overall investor’s sentiment, where they are associated with positive market perception. In this context, many previous studies control for the ability of underwriters to reveal information and their customer relations. During the last hot issue period the hypothesis that underwriters provide positive recommendation and analyst coverage, when issuers are willing to accept high underpricing was confirmed (e.g., Aggarwal/Krigman/Womack (2002: 109 et. seqq.), (Loughran/Ritter (2004: 9 et. seqq.)). For this paper also a dummy variable for prestigious underwriter reputation has been calculated, measured in the same way as in Franzke (2003: 14). The relative market share of an underwriter is calculated by equally weighting the relative share of lead managements of IPOs and the relative volume of proceeds of these issues, where only the investment banks within the highest rating category are classified as prestigious. In this sample, 55 different investment banks have been active as a lead underwriter.15 Unfortunately, the variable has been insignificant (economically and statistically) and did not contribute to the explanatory power of the regression models, so the results are reported without this measure. Additionally, previous literature showed that the underwriter variable is likely to be endogenous. When firms expect that they have to leave more money on the table to find investors, they are likely to choose a more prestigious underwriter to mitigate some informational asymmetries. For a better overview, the described variables and definitions are also presented at table I.

15

For underwriter activities see the appendix for chapter II (Appendix II: Table I Underwriter Activity). Underwriter activity of related banks has been grouped together (e.g. Cazenove and JP Morgan). The consolidation phase of banks in 2008 has not been taken into account.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

62

Chapter II: Table I

Definition of Variables Name UP Proceeds Return Vola

Definition Natural logarithm of underpricing plus one. Underpricing is defined as the share price after the first trading day divided by the offer price minus one. Natural logarithm of the offer price multiplied with the number of shares issued at the IPO. Percentage price change of all tradable shares in Germany in the previous 30 days before the IPO.

Volume

Average volatility of all tradable shares in Germany in the month of the IPO. Total number of IPOs in the month of the IPO. Also IPOs which are not included in the sample are included in the IPO counts.

IR

Average underpricing of IPOs in the month before the IPO.

Tech

Dummy variable which equals one if the IPO firm is classified “technology “ by the ICB.

Media

Dummy variable which equals one if the IPO firm is classified “media” by the ICB.

Dilution

Primary shares divided by pre-IPO shares of the firms.

Participation

Secondary shares divided by pre-IPO shares of the firm. Insider shares divided by pre-IPO shares outstanding. "Insider" includes board members, managers, family members and their private holdings. Financial investors’ shares (venture capitalist, private equity fund, bank funds) divided by pre-IPO shares outstanding. Blockholders’ shares divided by pre-IPO shares outstanding. Blockholders are shareholders which own more than 25% of the pre-IPO shares and are not classified in the other shareholder groups. Difference between pre- and post-IPO shares of insiders divided by pre-IPO shares outstanding. Difference between pre- and post-IPO shares of investors divided by pre-IPO shares outstanding. Difference between pre- and post-IPO shares of blockholder divided by pre-IPO shares outstanding. Dummy variable that equals one if an insider is shareholder before the IPO.

Insider Investor Blockholder Insider Sale Investor Sale Blockholder Sale Dum Insider Dum Investor Dum Blockholder

Dummy variable that equals one if a financial investors is shareholder before the IPO. Dummy variable that equals one if a blockholder is shareholder (with more than 25% of the shares) before the IPO.

Dum Insider Sale

Dummy variable that equals one if an insider sells part of his shares at the IPO.

Dum Investor Sale

Dummy variable that equals one if a financial investors sells part of his shares at the IPO.

Dum Blockholder Sale

Dummy variable that equals one if a blockholder sells part of his shares at the IPO.

Herfindahl

Herfindahl index: Sum of squared equity stakes of insiders, investors and blockholders. Dummy variable which equals one if the IPO took place in a month with more than the median value of IPOs per month. Dummy variable which equals one if the average underpricing of the IPOs in the month before the offering date exceeds the median value of underpricing in the sample. Dummy variable which equals one if the offer price is higher than the preliminary price, which is defined as the midpoint of the bookbuilding price range.

HotVolume HotIR HotPrice

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

63

III.3 Definition of Hot and Cold Periods In recent literature several classifications and definitions of hot and cold IPO phases can be found. For example, Helwege/Liange (2004:548 et. seqq.) investigate how firms in both periods differ, and which alternative characterization of hot and cold markets is appropriate. They calculate three month moving average of the number of IPOs, where month with more IPO counts than the top quartile are defined as hot, and the lower third of the sample is defined as cold. Furthermore, they classify the IPO based on its underpricing: Offerings with higher (lower) initial returns than the value of the top (bottom) quartile are defined as hot (cold) IPOs. They find that high underpriced offerings are more distinct from low underpriced IPOs in terms of firm age, proceeds and investments. The firm and offer differences according to IPO volume are not pronounced. They also analyze firm ownership in terms of institutional investors’ holdings after the IPO and find that hot markets’ IPOs (defined by volume) have higher institutional ownership than IPOs in a cold market. However, they do not investigate the firm’s ownership structure before firms go public. The classifications of market cycles, used in this study, are also defined by offering volume and initial returns. When an IPO is completed in a month with more IPO counts than the monthly median value, it is classified as “HotVolume” and the respective dummy variable equals to one. When average underpricing of IPOs in the month prior to the offering exceeds the median value of monthly underpricing across the sample, the firm is categorized as “HotIR”. Because of the smaller total sample size, the classification is orientated at the median values and only two sub-samples are grouped. It has been described before that months with high underpricing and IPO volume often coincide or follow each other, so it is not surprising that in 60% of the sample the “hot” classifications consist with each other. Additionally, the IPOs are split into two sub-samples according to the year of the IPO: 1997-2001 and 2002-2007. After bursting of the dot-com bubble in 2001, the level of underpricing as well as IPO volume decreased sharply. So it is necessary to confirm firm, transaction as well as ownership characteristics in both periods, to discuss the suggestion that owner’s intention are likely to changed with the market characteristics.

IV Empirical Results IV.1 Firm and Transaction Characteristics The complete sample consists of 439 IPOs, and their monthly counts between 1997 and 2007 are shown in figure I. During the first 5 years of the period in ques-

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

64

tion, 257 firms went public, mainly in the stock segment for small and medium sized companies. While the IPO market in Germany was almost inactive in 2002 and 2003, the number of IPOs increased in 2004. The volume averages 4.166 IPOs per month in the last two years of the sample period. The graph in figure II shows average monthly initial returns. In the boom period of Neuer Markt, from 19972001, monthly underpricing averaged 31%16 and was much higher than in the following 6 years. Initial returns also remained much lower, with an average of 6% after 2006, at which point the IPO market seems to have recovered. Chapter II: Figure I

IPOs per Month between 1997-2007 25 20 15 10 5

16

Jan 07

Jan 06

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

Jan 97

0

This is the average for all IPOs in these years, not just IPOs included in the sample.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

65

Chapter II: Figure II

Initial Returns per Month between 1997-2007 2 1,5 1 0,5

Jan 07

Jan 06

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

-0,5

Jan 97

0

-1

Table II provides characteristics of the issuing firms, presenting the mean, median and standard deviations from the values. Additionally, more important insight is obtained from table III, where the differences in the respective variables are analyzed in hot and cold market periods. The mean and median values are presented, as well as the p-value. These p-values denote the probability of accepting the nullhypotheses of both the equality of means t-test and the Wilcoxon-Mann-Whitney test, which assumes that the samples come from the same distribution. The results for the sub-samples, divided by the offering date (panel A: 1997-2001; 20022007), do not show any significant differences in the mean and median values for firms’ total and intangible assets, book value of equity, total debt and sales. Some transaction characteristics differ significantly between sub-periods. Proceeds, defined as the total number of offered shares multiplied by the offering price, are higher in the early market phase (1997-2001) than in the later ones, with a median of € 37.9 mio., compared to € 27.475 mio. The number of secondary shares (sample mean: 1.957 mio.) and primary shares (sample mean: 3.419 mio.) show a higher likelihood of confirmation of the null-hypotheses of the t-test and WilcoxonMann-Whitney test. Overall, more new shares are issued than are sold by pre-IPO owners. The variable “freefloat” accounts for the relationship between publicly traded shares and firm’s total shares after the IPO, and indicates more wealth diversification among pre-IPO shareholders between 2002 and 2007. As suggested above, the differences in the firms’ initial returns in both market phases are very

66

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

high, at 38.2%, and the probability of confirmation of the null-hypotheses are close to zero. Similar conclusions can be drawn when firm- and transaction characteristics in hot and cold phases are considered in terms of IPO volume and underpricing. The ttest and Wilcoxon-Mann-Whitney test confirm that firms going public in months succeeding periods of high underpricing do not significantly differ from IPOs in the other sub-sample (panel B). The transaction characteristics approve the pattern described above, if the period 1997-2001 is considered as a hot phase and 20022007 as a comparatively cold period. IPOs have higher proceeds and initial returns, as well as a lower percentage of freefloat in the shares, in hot months. Also, with this sample distinction in panel B, the mean of pre-IPO owner participation in terms of secondary shares is lower in hot (mean: 0.728 mio.) than in cold months (mean: 3.140 mio.). This suggests that old shareholders prefer to raise more cheap capital in hot issue phases, rather than to sell part of their stock and participate in the offering. Furthermore, in panel C the classification of the sub-samples based on the IPO volume per month show differences in firms’ size, leverage and profitability. The median values are higher in cold months, and the p-values of the Wilcoxon-Mann-Whitney test indicate low probability (up to 5%) of both samples having the same distribution. The variables for IPO characteristics also confirm that the amounts of secondary shares, percentage freefloat and initial returns differ in hot and cold samples. Only the statistical tests for IPO proceeds do not support previous findings that firms raise more equity in an advantageous market environment. Overall, the results are consistent with Helwege/Liang (2004: 558), although they find more distinctions in the sub-samples classified by underpricing, rather than by IPO volume. Hot market periods are a favorable opportunity for young start-ups with greater growth potential to go public, but not only for this group. The differences in the transactions characteristics confirm that this market environment is a window of opportunity for all types of firms and issuers to increase their proceeds, to obtain more relatively cheap equity. In contrast, pre-IPO shareholders’ participation is higher in cold issue periods, and they distribute more of their shares on the public market.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

Chapter II: Table II

Firm and Offer Characteristics The accounting variables are from the IPO firms' financial reporting closest to the IPO date. The book value of equity, total debt, total assets, intangible assets and net sales are denoted in mio. €. "Proceeds" is calculated as the number of offered shares (primary and secondary shares) multiplied with the offer price. Proceeds, primary and secondary shares are denoted in million. "Freefloat" is calculated as the number of total offered shares at the IPO divided by the number of total shares of the firm after the IPO (including primary shares)."Initial Return" is measured as first trading day closing price divided by the offer price minus one. Mean

Median

St.Dev.

Equity Debt Assets Intangible Assets Sales

203.885 174.820 880.48 17.472 308.646

17.895 2.155 36.505 0.380 21.200

1937.679 2230.973 9294.915 103.548 2347.17

Proceeds Primary Shares Secondary Shares Freefloat (in %) Initial Return (in %)

119.00 3.419 1.957 32.071 31.973

37.161 1.500 0.182 29.028 5.131

569.000 14.395 1.403 14.056 77.841

67

Initial Returns

Freefloat

Secondary Shares

Primary Shares

Proceeds

Sales

Debt

Equity

Intangible Assets

Assets

Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median

1.215.362 36.805 13.809 0.420 270.722 17.790 214.310 2.190 385.287 20.340 116.000 37.900 3.550 1.700 1.648 0.176 0.306 0.279 0.428 0.099

276.724 34.900 25.255 0.310 83.590 18.530 104.114 2.095 171.263 24.695 122.000 27.475 3.126 1.300 2.701 0.221 0.352 0.339 0.046 0.012

1997-2001 2002-2007

Panel A: IPO Date 0.356 0.573 0.357 0.851 0.377 0.663 0.651 0.733 0.404 0.607 0.918 0.011 0.779 0.245 0.474 0.149 0.004 0.001 0.000 0.000

P-Value 1.624.433 34.040 16.976 0.430 343.732 16.200 284.381 2.460 490.977 21.080 80.238 40.500 2.910 1.500 0.728 0.175 0.305 0.279 0.405 0.081

Hot 235.904 38.785 17.996 0.370 82.907 19.235 80.087 1.925 151.123 21.340 154.000 32.028 3.920 1.500 3.140 0.200 0.334 0.300 0.227 0.039

Cold 0.154 0.467 0.925 0.600 0.199 0.652 0.382 0.984 0.167 0.896 0.192 0.011 0.465 0.465 0.072 0.939 0.033 0.020 0.015 0.103

P-Value

Panel B: IPO Underpricing

0.787 0.057 0.509 0.839 0.281 0.136 0.352 0.010 0.893 0.027 0.949 0.179 0.536 0.196 0.888 0.040 0.000 0.000 0.175 0.022

Cold P-Value

748.604 1.010.716 33.220 41.175 13.862 21.020 0.380 0.390 314.142 99.981 14.450 19.750 286.449 69.425 1.530 3.295 326.356 293.402 18.320 25.625 120.000 116.000 38.422 32.719 3.836 2.981 1.675 1.400 1.868 2.056 0.162 0.200 0.287 0.355 0.271 0.328 0.705 0.262 0.077 0.037

Hot

Panel C: IPO Volume

See table I chapter II for definition of variables. IPO Underpricing: "Hot" ("cold") defines IPOs where the average underpricing of IPOs in the previous month is higher (lower) than the median value of the sample's monthly underpricing. IPO Volume: "Hot"("cold") defines IPOs occurring in month with higher (lower) total number of IPOs than the median value of monthly IPO counts of the complete sample. The p-value denotes the probability to reject the null-hypothesis of the t-test of equality of means and the Wilcoxon-Mann-Whitney test, if the hot and cold samples have the same distribution.

Firm Characteristics in Hot and Cold Markets

68 HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

Chapter II: Table III

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

69

IV.2 Pre-IPO Ownership Characteristics To gain an insight into pre-IPO shareholders’ intentions in different market periods, the ownership structure is analysed in more detail. Table IV provides evidence for the development of the shareholder classification of insiders, investors, and blockholders, as well as the Herfindahl index. "Mean of IPOs" shows the percentage of IPOs with the considered shareholder group before going public represented in the sample. “Stakes before” is calculated as the ratio of respective owner’s shares to total shares outstanding before the IPO. “Stakes after” is calculated as the ratio of respective owner’s shares after the IPO in relation to the total number of shares after the IPO. "Sales" is calculated as the respective shareholder’s difference in shares before and after the IPO, divided by the number of shares held before the firm went public. The ratios described are calculated solely from the sample of IPOs in which the respective shareholder group is represented. Throughout the sample period, managers, supervisory board, or related persons own stakes in the firm in almost every IPO (mean: 93.8%). The holdings average 68.9%, forming the largest group of pre-IPO owners. When going public, an average of 4.8% of the outstanding shares are sold. Their stakes fall below 50% after the IPO has been completed because of dilution. No major changes are seen in the long term, excepting possibly that in 1997 the insider stakes (mean: 72.7%) and sales were higher (mean: 8.5%) than in the following years. The results for the period 2002-2004 are not conclusive, as there is only a limited amount of data available for these years. The financial investors, such as VCs or private equity funds, are involved in 54.7% of the sample IPOs and hold, on average, 35.4%. Compared to insiders, they participate in more secondary shares at the IPO and sell, on average, 7.2%. This confirms that financial investors use the IPO as an opportunity to exit the investment, while founders use it as an opportunity to raise equity. Development during the sample period after 2004 shows more IPO firms with shares held by these investors than seen in the boom period of the new markets. Surprisingly, the percentage of financial investors’ stakes falls below the mean value during the years with very high IPO volume (1998-2000). Percentage sales are also especially low in 2000 and 2001. The third classification includes all blockholders with more than 25% of the firm’s equity. This group of pre-IPO owners is represented in 16.4% of the sample’s firms, and their stakes average 66.9%. The percentage of sales is also relatively high, at 9.6%, so their stakes decrease by about 20% during the IPO, which is similar to the findings on insider wealth changes. The Herfindahl index varies around a mean value of 0.75, and indicates much greater clustering in ownership structures in the German IPO sample. In the study by Ljungqvist/Wilhelm (2003: 732) in the USA, the Herfindahl index shows a value

70

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

of 0.35. During the 11 year period, the index does not indicate that the ownership structure has changed exceptionally. In 2005 and 2006, relatively low values were achieved, which would indicate higher underpricing than in the hot phases of 1997-2001.

0,164 0.669 0.464 0.096

0.747

Blockholder IPOs Stakes Before Stakes After Sales

Herfindahl

Herfindahl

0,547 0.354 0.241 0.072

IPOs Stakes Before Stakes After Sales

Investors

0,938 0.689 0.468 0.048

IPOs Stakes Before Stakes After Sales

Insider

0.760

0.733 0.486 0.001

0.300 0.172 0.018

0.774 0.548 0.004

Mean Median

1997-2007

0.829

0,273 0.869 0.610 0.187

0,364 0.420 0.228 0.115

0,727 0.796 0.491 0.085

Mean

1997

0.771

0,056 0.483 0.101 0.302

0,528 0.335 0.169 0.115

1,000 0.762 0.554 0.059

Mean

1998

0.792

0,113 0.785 0.552 0.107

0,540 0.316 0.192 0.077

0,927 0.773 0.572 0.068

Mean

1999

0.751

0,208 0.661 0.489 0.048

0,542 0.279 0.185 0.029

0,992 0.700 0.501 0.022

Mean

2000

0,783

0,188 0.757 0.457 0.343

0,500 0.385 0.245 0.019

0,813 0,787 0.526 0.031

Mean

2001

0,635

0,000 0,000 0,000 0,000

1,000 0,542 0,354 0,000

1,000 0,429 0,354 0,000

Mean

2002

/

/ / / /

/ / / /

/ / / /

Mean

2003

0,846

0,000 0,000 0,000 0,000

0,667 0,866 0,678 0,000

1,000 0,422 0,264 0,000

Mean

2004

0,575

0,313 0,569 0,284 0,227

0,688 0,549 0,326 0,107

0,938 0,372 0,247 0,062

Mean

2005

0,656

0,147 0,674 0,476 0,057

0,574 0,419 0,246 0,104

0,912 0,594 0,436 0,044

Mean

2006

0,751

0,286 0,539 0,381 0,005

0,571 0,405 0,223 0,098

0,905 0,586 0,427 0,058

Mean

2007

The ownership data is hand collected from the IPO prospectus. "Insider" includes stakes of board members, top management and related persons. Also related holding companies of these persons are classified as "insiders". "In vestor" includes venture capital, bridge financing and private equity (also buyouts) also provided by banks, or bank related funds. “Other blockholders” are shareholders which own more than 25% of the pre-IPO shares and are not classified in other shareholder groups. The numbers are calculated in percent (%). "IPOs" stands for the percentage of IPOs in this year with this type of omner. "Stakes before" is calculated as the ratio of shares of the respective owner to total shares before the IPO. 'Stakes after" are claculated as the ratio of shares after the IPO of the respective owner in relation to the total number of shares after the IPO (excluding shares from the overallotment option). "Sales" is calculated as the difference of shares before and after the IPO of the respective shareholder divided by the number of shares before the IPO. The ratios are calculated only from the IPOs, in which the respective shareholder group is present. The Herfindahl index measures the ownership concentration: sum of

Ownership Structure

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

Chapter II: Table IV

71

72

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

In table V, the ownership structures in hot and cold IPO periods are presented. The stakes before and after the IPO are calculated, as are the shares sold, as an overall percentage of the sample and not only for the IPO group relating to the particular shareholder. The classification “insider” shows significant differences between hot and cold samples according to offer date and IPO volume. In panels A and C, the management and supervisory board members have larger stakes in the firm before and after going public in hot periods than in cold periods. The p-values indicate low probability, up to 5%, of the mean values being equal, and the samples show the same distribution. These results are not confirmed by the sample differentiation according to underpricing in the prior month of the offering. However, further distinctions can be made considering financial investors’ ownership. In panels A, B, and C, the investor stakes in cold months average 22%, compared to 16% in hot IPOs. The differences are significant according to the t-test. The mean stakes after the IPO are also higher in the cold periods, which is confirmed by the statistical tests in panels A and C. The differences in mean stakes or shares sold in the third category “blockholders” are not extreme. The t-test and Wilcoxon-Mann-Whitney test indicate a probability of accepting the null-hypotheses. Overall, ownership by blockholders is relatively low, if the value is not calculated for a separate sample of IPOs with these shareholders being involved. Lastly, the Herfindahl index indicates differences in the samples. The classifications in terms of offering date and volume (Panel A and C) suggest higher ownership clustering and a higher index value in hot than in cold IPOs. These findings contradict the hypothesis and results presented by Ljungqvist/ Wilhelm (2003). Considering the IPOs according to their market environment, months with presumably higher initial returns indicate higher clustering of pre-IPO shareholders. This means that the argument that more dispersed ownership over time results in higher underpricing cannot be accepted. Furthermore, the results suggest that the insider group, rather than financial investors, promotes the decision to go public in hot phases, because of their higher ownership stakes. The idea of a fast, profitable exit by venture capital or private equity investors from their investment during active months in the IPO market is not necessarily confirmed by their smaller stakes in outstanding shares. They hold considerably larger stakes in IPO firms, and sell more in cold phases. Overall, the descriptive statistics suggest that the ownership structure of IPOs has changed slightly between the periods of 1997-2001 and 2002-2007. The findings are also supported when the IPOs are classified according to the market environment. However, the index of clustering has not increased, which would support the previous argument put forward by Ljungqvist/Wilhelm (2003). Therefore analysis is required into which factors determine IPO underpricing, and whether or not their explanatory power has changed over time.

Herfindahl

Blockholder Sales

Blockholder Stakes After

Blockholder

Investor Sales

Investor Stakes After

Investor Stakes

Insider Sales

Insider Stakes After

Insider Stakes

Mean Median

Mean Median Mean Median Mean Median

Mean Median Mean Median Mean Median

Mean Median Mean Median Mean Median

0.776 0.811

0.264 0.000 0.074 0.000 0.014 0.000

0.160 0.014 0.096 0.000 0.100 0.000

0.708 0.800 0.498 0.567 0.047 0.000

0.681 0.673

0.118 0.000 0.080 0.000 0.016 0.000

0.266 0.081 0.151 0.034 0.126 0.000

0.510 0.520 0.362 0.367 0.045 0.004

0.000 0.002

0.726 0.795 0.782 0.491 0.781 0.918

0.000 0.014 0.001 0.049 0.120 0.136

0.000 0.000 0.000 0.000 0.904 0.164

Panel A: IPO Date 1997-2001 2002-2007 P-Value

0.748 0.777

0.124 0.000 0.085 0.000 0.015 0.000

0.161 0.014 0.099 0.000 0.030 0.000

0.675 0.760 0.473 0.531 0.050 0.002

0.747 0.745

0.098 0.000 0.067 0.000 0.014 0.000

0.221 0.059 0.125 0.029 0.051 0.000

0.623 0.747 0.443 0.518 0.043 0.000

0.956 0.805

0.321 0.588 0.333 0.523 0.901 0.484

0.019 0.176 0.110 0.267 0.031 0.429

0.134 0.206 0.252 0.245 0.564 0.994

Panel B: Underpricing Hot Cold P-Value

0.769 0.805

0.110 0.000 0.079 0.000 0.010 0.000

0.164 0.013 0.100 0.000 0.031 0.000

0.697 0.799 0.498 0.588 0.044 0.000

0.724 0.723

0.112 0.000 0.071 0.000 0.019 0.000

0.222 0.059 0.126 0.022 0.051 0.000

0.596 0.680 0.414 0.475 0.049 0.004

0.080 0.039

0.940 0.795 0.665 0.911 0.288 0.889

0.024 0.115 0.012 0.021 0.041 0.504

0.003 0.043 0.001 0.005 0.409 0.075

Panel C: IPO Volume Hot Cold P-Value

See table IV for definition of variables. Underpricing: "Hot" ("cold") defines IPOs where the average underpricing of IPOs in the previous month is higher (lower) than the median value of the sample's monthly underpricing. IPO Volume: "Hot"("cold") defines IPOs occurring in month with higher (lower) total number of IPOs than the median value of monthly IPO counts of the complete sample. The p-value denotes the probability to reject the nullhypothesis of the t-test of equality of means and the Wilcoxon-Mann-Whitney test, if the hot and cold samples have the same distribution.

Ownership Structure in Hot and Cold Markets

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

73

Chapter II: Table V

74

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

IV.3 Regression Analysis IV.3.1 Variables Explaining IPO Underpricing To demonstrate the influences of shareholder ownership and related agency conflicts on IPO underpricing, several regression equations have been examined using the IPO data from 1997-2007. The ordinary-least-square (OLS) regression models are estimated with White’s (1980) standard errors and covariance, to provide a control for heteroskedasticity of the residuals. The coefficient estimates and the tstatistic values for the null-hypothesises are presented in table VI. First of all, regression UP [1] indicates the importance of the control variables and their effects on the dependent variable of underpricing. The variable of gross proceeds as a proxy for investor uncertainty about an issue shows the expected negative coefficient, although not significant at the relevant levels. Previous performance of all tradable shares on the German stock exchanges during the 30 trading days before the IPO (“return”) has a positive, and the volatility (“vola”) of the shares in the same month a negative, effect on IPO underpricing. Both variables are significant at the 1% level. This supports previous findings: positive market movements are not fully taken into account in higher offer prices, which results in higher initial returns. Higher volatility seems to influence issuers’ and underwriters’ insecurity about investors’ willingness to pay for new shares. Significant t-statistics (at 5%) are found for the variable of “IR”, the average of IPOs’ initial returns one month prior to the offering of the sample firm. Bradley/Jordon (2002: 610) find significant positive results for the average underpricing of the IPOs 30 days before the offering date, and conclude that publicly available information is not fully incorporated in the offering price and so underpricing becomes predictable to some extent. The coefficient is not significant, but shows a positive sign for IPO “volume”. This means that no confirmation is found for the argument that underwriters learn more about the market as IPO volume increases and so information are not completely incorporated into the offering price over time. The dummy variables “tech” and “media”, which equal one if the firm operates in respective industry segments, shows a significant positive correlation to underpricing at levels of 5% and 10%. The explanatory power of these variables, for underpricing, remains unchanged when several proxies for firm ownership are included. However, the R-squares of the following models increase, suggesting dependency of ownership structure and underpricing. The regression estimates for UP [2] and UP [3] provide a control factor for the percentage stakes of the three shareholder classifications, which only show significant statistical values for “blockholder”, at 10%. The larger the holdings of outside blockholders, the lower the initial returns. A negative coefficient is found for the relationship between financial investors’ stakes and first day returns.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

75

The positive sign of the coefficient for insiders’ stakes in UP [2] would predict that the board members’ and managers’ group deliberately accept underpricing, presumably to signal the high quality of the firm. Also Ljungqvist/Wilhelm (2003: 743) find a positive coefficient for the variable of CEO’s stakes in the IPO and estimate equation coefficients of other shareholder groups’ stakes with negative signs. However, when the “participation” and “dilution” variables are applied, to control for shareholder wealth effects, the sign of the coefficient for insiders’ stakes also turns negative. When owners participation in the IPO, by offering secondary shares, is high, and when more primary shares are issued, initial returns decrease. Also, larger shareholders’ ownership stakes result in higher monitoring of underwriters and in increased interest in reducing underpricing. Furthermore, this is supported by the estimates of UP [4], where the explanatory power of the Herfindahl index is tested, rather than ownership stakes separately. The coefficient is negative and significant at the 5% level. Many previous studies have determined the effects of several shareholder groups using dummy variables, which equal one if the respective shareholder group is represented, and otherwise zero. In regression UP [5] and UP [6], the dummy for insiders and financial investors shows a positive relation, while the presence of large blockholders has a negative affect on the dependent variable. The results for the first regression models are similar to Ljungqvist/Wilhelm (2003), meaning that further estimates are completed using owner’s stakes variables instead of dummy variables.

76

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

Chapter II: Table VI

Regression Models on Underpricing (1) For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [1] UP [2] UP [3] UP [4] UP [5] UP [6] Proceeds Return Vola Volume IR Tech Media

-0.016 (-1.466) 0.196 (4.076)*** -1.105 (-2.630)*** 0.003 (1.411) 0.145 (2.323)** 0.121 (2.959)** 0.150 (1.973)*

-0.010 (-1.057) 0.187 (3.861)*** -1.093 (-1.057)*** 0.003 (1.361) 0.151 (2.437)*** 0.109 (2.663)*** 0.148 (1.965)**

Dilution Participation Insider

0.013 (0.195) -0.030 (-0.290) -0.148 (-1.666)*

Investor Blockholder

-0.004 (-0.218) 0.189 (3.892)*** -1.209 (-3.089)*** 0.003 (1.082) 0.145 (2.330)** 0.106 (2.604)*** 0.140 (1.841)* -0.027 (-0.377) -0.221 (-1.623)* -0.001 (-0.006) -0.022 (-0.188) -0.157 (-1.764)*

-0.07 (-0.424) 0.198 (4.095)*** -1.314 (-3.222)*** 0.003 (1.196) 0.138 (2.205)** 0.123 (2.978)*** 0.144 (1.883)* 0.005 (0.085) -0.264 (-2.247)**

Dum Insider Dum Investor Dum Blockholder Herfindahl

R²

0.744 (2.508)*** 0.141

0.659 (2.281)*** 0.151

0.637 (1.587)*** 0.156

-0.143 (-2.124)** 0.815 (2.023)** 0.155

F-Statistic

10.109

7.647

6.566

7.842

Intercept

-0.001 (-0.071) 0.193 (4.001)*** -1.423 (-3.118)*** 0.002 (0.982) 0.135 (2.210)** 0.111 (2.708)*** 0.128 (1.683)* -0.031 (-0.468) -0.234 (-1.744)*

0.001 (0.009) 0.193 (3.959)*** -1.446 (-3.150)*** 0.002 (1.034) 0.136 (2.224)** 0.113 (2.756)*** 0.132 (1.733)* -0.026 (-0.384) -0.252 (-1.909)**

0.148 (1.247) 0.072 (2.020)* -0.044 (-0.790)

0.489 (1.591) 0.168

0.128 (1.112) 0.046 (1.219) -0.062 (-1.096) -0.092 (-1.434) 0.577 (1.779)* 0.170

7.136

6.688

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

77

In table VII, equation estimates for the effects of pre-IPO owners’ sales are presented. Insiders and blockholders are associated with less money left on the table when their participation in the offering is high: The more shares are sold, the lower are the initial returns. The results of the t-statistics also confirm significant effects of these variables (to the 1% level). VCs’ or other financial investors’ selling behavior has a positive correlation to initial returns in UP [7]. However, when the Herfindahl index is included as an additional variable in the regression model (UP [8]), the sign of the coefficient also becomes negative. Surprisingly, when the estimates are repeated with dummy variables for owners’ sales (UP [9], UP [10]), no effects are confirmed at statistically relevant levels. Furthermore, when insiders and financial investors sell shares, higher underpricing is achieved. It is likely that market investors interpret the participation of pre-IPO owners as a negative sign and so demand higher initial returns. The variables for shares sold in relation to total shares outstanding, however, capture the owner’s negative wealth effect of underpricing and therefore have a negative correlation to the dependent variable. Ljungqvist/Wilhelm (2003) also show the expected negative influence of owner participation and increased incentives to bargain about optimal (higher) offering prices.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

78

Chapter II: Table VII

Regression Models on Underpricing (2) For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the tstatistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [7] UP [8] UP [9] UP [10] Proceeds Return Vola Volume IR Tech Media Dilution Insider Sale Investor Sale Blockholder Sale

-0.132 (-0.692) 0.192 (4.002)*** -1.211 (-3.033)*** 0.003 (1.261) 0.146 (2.310)** 0.113 (2.748)*** 0.144 (1.891)** -0.039 (-0.541) -0.300 (-3.367)*** 0.046 (0.275) -0.399 (-2.502)***

-0.010 (-0.586) 0.193 (4.012)*** -1.277 (-3.113)*** 0.003 (1.325) 0.145 (2.298)** 0.118 (2.865)*** 0.148 (1.946)** -0.032 (-0.445) -0.276 (-3.148)*** -0.028 (-0.177) -0.388 (-2.433)***

Dum Insider Sale Dum Investor Sale Dum Blockholder Sale

-0.019 (-1.126) 0.195 (4.010)*** -1.107 (-2.697)*** 0.004 (1.438) 0.153 (2.455)*** 0.119 (2.905)*** 0.149 (1.946)** 0.003 (0.043)

-0.016 (-0.958) 0.194 (3.987)*** -1.168 (-2.805)*** 0.004 (1.502) 0.152 (2.453)*** 0.124 (2.990)*** 0.153 (1.998)** 0.005 (0.078)

0.033 (0.897) 0.058 (0.143)

0.036 (0.995) 0.029 (0.754)

-0.074 (-1.317)

R²

0.766 (1.962)** 0.154

-0.144 (-1.761)* 0.835 (2.079)** 0.159

0.758 (2.078)** 0.152

-0.086 (-1.521) -0.118 (-1.739)* 0.823 (2.209) 0.157

F-Statistic

7.059

6.705

6.925

6.565

Herfindahl Intercept

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

79

IV.3.2 Determinants of IPO Underpricing in 1997-2001 and 2002-2007 The regression results confirm that ownership structures before and after the IPO determine the level of initial returns. Additionally, the descriptive statistics in IV.2 indicate some changes in pre-IPO shareholders over time and in different market cycles. These factors necessitate an analysis of whether the determining factors of underpricing have remained the same over time and whether changes in ownership structure could explain differences in levels of initial returns. Furthermore, a general shift downwards in the level of underpricing could be responsible for the high differences in initial returns seen between 1997-2001 and 2002-2007, meaning that the considered variables have the same influence in both market periods. For this reason, the complete sample is split into the two sub-periods 1997-2001 and 2002-2007, and the OLS-regressions are repeated. The main results are presented in table VIII. For the first period (1997-2001), the coefficients and t-tests are similar to the results described in IV.3.1. The variables of previous return, previous IPO underpricing and industry segments have a positive influence, while volatility decreases initial return on the first trading day. Ownership stakes and proxies for participation and dilution show negative coefficients. The regression model analyzing shareholders’ sales confirms the negative correlation for “insider sales” and “blockholder sales” to underpricing, significant at the 5% and 10% levels. However, financial investors’ sales indicate a positive correlation and suggest that those owners’ concern about money left on the table is inversely proportionate to how much they sell. In the second sample period (2002-2007) the explanatory power of the variables considered in the OLS-regression models changes. In UP [14], only the “participation” and “dilution” proxies have a significant negative coefficient to underpricing. The industry dummy variable for technology firms is also significant at the 10% level, but with a negative sign of the coefficient. In the earlier period, technology-orientated firms dominated the IPO market and experienced enormous initial return. With decreasing IPO volume in this industry segment, investors’ compensation for participating in the respective offerings decreased. IPO firms in the media sector, however, were still associated with more money left on the table. The “return” variable, controlling for previous market performance, shows a high positive coefficient (4.858). Although the t-test does not indicate the significance of this variable at the respective levels, the economic meaning has become more important in explaining “UP”. Further variables in the regression model have changed signs. “Proceeds” and “vola” have a positive effect, while higher values of “volume” and “IR” decrease initial return. Additionally, the coefficients signs for pre-IPO ownership proxies change in UP [14]: higher shareholders’ stakes before the IPO result in higher underpricing when going public. A similar develop-

80

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

ment can be seen in UP [15], where proxies for owners’ selling behavior are included as independent variables. The managers’ and related persons’ group accepts higher underpricing when they sell more of their shares, while financial investors seem to bargain in the opposite direction. “Blockholder sale” is negatively related to the dependent variable, which confirms previous results. Additionally, UP [15] and UP [16] include the variable for ownership clustering before going public (“Herfindahl”). Interestingly, the effects of more clustered ownership structures are positive for the period between 2002 and 2007, although the coefficients are much smaller and not significant. Overall, the F-statistics in UP [14]-UP [16] are relatively small and do not, compared to the previous estimates, indicate high explanatory power on the part of the variables considered in respect of underpricing. The results are also puzzling, because shareholders in IPO firms are expected to be more concerned about money left on the table in negative market environments, whereas the positive coefficients suggest the opposite.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

81

Chapter II: Table VIII

Regression Models on Underpricing: 1997-2001, 2002-2007 For definition of variables look at table I. The regression models use White's (1980) heteroskedasticityconsistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance.

1997-2001

Proceeds Return Vola Volume IR Tech Media Dilution Participation Insider Investor Blockholder

2002-2007

UP [11]

UP [12]

UP [13]

UP [14]

UP [15]

UP [16]

-0.030 (-0.747) 0.174 (3.478)*** -1.470 (-2.506)*** 0.002 (0.465) 0.137 (1.923)** 0.120 (2.280)** 0.153 (1.654)* -0.015 (-0.169) -0.249 (-1.794)* -0.169 (-0.697) -0.190 (-0.620) -0.341 (-1.379)

-0.046 (-1.188) 0.185 (3.820)*** -1.527 (-2.534)*** 0.002 (0.543) 0.133 (1.881)* 0.143 (2.720)*** 0.165 (1.801)* -0.038 (-0.415)

-0.045 (-1.245) 0.189 (3.836)*** -1.627 (-2.775)*** 0.002 (0.468) 0.127 (1.790)* 0.144 (2.734)*** 0.154 (1.657)* 0.017 (0.209) -0.308 (-2.594)***

0.008 (0.977) 4.858 (1.101) 0.128 (0.450) -0.001 (-0.635) -0.051 (-0.430) -0.040 (-1.706)* 0.017 (0.380) -0.127 (-2.126)** -0.253 (-2.146)** 0.019 (0.503) 0.025 (0.504) 0.052 (0.985)

-0.003 (-0.378) 4.701 (1.050) 0.041 (0.140) -0.001 (-0.453) -0.044 (-0.348) -0.051 (-2.019)** 0.010 (0.238) -0.083 (-1.164)

0.008 (0.935) 4.884 (2.654)*** -0.001 (-0.578) -0.001 (-0.578) -0.049 (-0.360) -0.040 (-1.276) 0.016 (0.318) -0.129 (-1.942)** -0.252 (-2.220)**

Insider Sale

R²

1.399 (1.611)* 0.137

-0.318 (-3.352)*** 0.022 (0.088) -0.379 (1.866)* -0.254 (-2.100)** 1.689 (2.085)** 0.153

-0.308 (-2.594)*** 1.757 (2.316)** 0.148

F-Statistic

3.923

4.462

5.192

Investor Sale Blockholder Sale Herfindahl Intercept

-0.086 (-0.465) 0.131

0.046 (0.290) -0.051 (-0.549) -0.023 (-0.136) 0.006 (0.240) 0.130 (0.650) 0.089

0.001 (0.003) -0.061 (-0.322) 0.125

1.464

0.943

1.696

82

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

The regression analysis is repeated for the complete sample, with an interaction term “HotDate” to reassess whether the determinants or relevance of the variables have changed between the two periods. Also, this dummy variable “HotDate” is introduced as a further independent variable and equals one if the IPO took place between 1997 and 2001. Otherwise it is zero. The explanatory variables are also multiplied by this interaction term in the regression models, and the coefficients indicate the differences in these variables between the market periods 1997-2001 (hot) and 2002-2007 (cold). The results are presented in table IX and the tstatistics indicate the significance of the changes. The dummy variables “HotDate” in UP [17] and UP [18] indicate changes of the intercept in the regression equations during hot and cold periods. The variable is positive and significant at the 10% level, which suggests that in the hot period of 1997-2001 underpricing was indeed higher in general. The level of initial returns fell after the active Neuer Markt period. However, the estimates for the interaction term confirm high changes in the respective coefficients in the opposite direction for hot and cold periods, whereas the difference in the variables “tech” and “vola” are significant at the 1% level. The results confirm that the explanatory power of the variables also changed in both periods. In addition to the different intercepts, the slope of the regression models and thereby also the determinants of underpricing differ in the sub-samples.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

83

Chapter II: Table IX

Regression Models on Underpricing: Hot Time Period For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [17] Proceeds

0.008

UP [17] HotDate*Proceeds

(0.999) Return

4.858 0.128

HotDate*Return

-0.001

HotDate*Vola

-0.051 -0.040

HotDate*IR

0.0176

HotDate*Tech

-0.127 -0.253

HotDate*Dilution

0.019

HotDate*Participation

0.025

HotDate*Insider

0.052

0.136 0.111 0.004 -0.189

HotDate*Investor

-0.189

HotDate*Blockholder

-0.393

Tech

(-1.544) HotDate

1.486

HotDate*Vola

-0.001 -0.044 -0.051 0.010 -0.083

HotDate*IR

0.006

HotDate*Tech

0.046 -0.051

HotDate*Dilution

-0.023

HotDate*Herfindahl

0.086

-0.260 (-2.087)**

HotDate*Insider Sale

-0.365 (-1.980)**

HotDate*Investor Sale

0.073 (0.271)

HotDate*Blockholder Sale

(-0.139)

-0.356 (-1.351)

HotDate

(1.660)* Intercept

0.045 (0.391)

(0.091) Blockholder Sale

0.154 (1.519)

(0.297) Investor Sale

0.195 (3.314)***

HotDate*Media

(0.246) Insider Sale

0.178 (1.235)

(-1.191) Herfindahl

0.003 (0.699)

(0.244) Dilution

-1.568 (-2.322)**

HotDate*Volume

(-2.066)** Media

(-0.761)

(1.007)

0.041

-4.516 (-1.031)

(-0.356)

(-0.761)

(0.515) Blockholder

IR

(0.026)

(0.514) Investor

0.161

HotDate*Return

(-0.464)

(1.001)

(-2.195)** Insider

0.188

4.701

-0.042 (-1.045)

(0.143) Volume

(1.305)

(-2.195)** Participation

0.003

(2.769)*** HotDate*Media

(0.389) Dilution

Vola

(1.377)

(1.745)* Media

-1.598

HotDate*Proceeds

(1.074)

(0.710)

(-0.440) Tech

Return

(-2.443)*** HotDate*Volume

(-0.650) IR

-4.684

-0.003

UP [18]

(-0.387)

(-1.086)

(0.461) Volume

Proceeds

(-0.917)

(1.126) Vola

-0.038

UP [18]

1.559 (1.855)*

Intercept

(0.476)

0.130 (0.665)

R²

0.1811

R²

0.194

F-Statistic

3.636

F-Statistic

3.952

IV.3.3 Pre-IPO Ownership and Underpricing in Different Market Phases Not only has ownership structure changed over time and in different market phases, but also the influence of pre-IPO owners on the level of underpricing differs within the 11-year sample period. It is hypothesized that the willingness of preIPO owners to leave money on the table changes according to the market cycles. In this section the different shareholder groups’ bargaining incentives within hot and cold IPOs are investigated in more detail, because the differentiation of the market phases according to the IPO date shows rather unexpected estimates. In table X the results, where a hot market is classified according to average under-

84

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

pricing in the month prior to the offering of the sample firm, are presented. The variable “HotIR” is introduced as a simple dummy variable, and equals one if the firm’s IPO follows a month with higher average underpricing than the median value of the sample, otherwise being zero. The variable “IR” is thereby no longer included in the regression models, as it controls for the same effect. Furthermore, an interaction term is introduced: the dummy variable is multiplied by the variables of insiders’, investors’, and blockholders’ shares before the IPO, and their percentage sales. These coefficients indicate the change in the respective variable and the effect on underpricing when the IPO takes place in a hot month. Regression model UP [19] indicates a positive relation for insider holdings, as well as negative relations for financial investors’ and blockholders’ stakes to initial returns. The same variables with the interaction dummy “HotIR” confirm the coefficient sign for insiders and blockholders, also showing relatively small changes in economic meanings. However, the coefficient for “HotIR*Investor” is high, and also implies a positive correlation of financial investor’s stakes to underpricing in hot market phases. Venture capitalists and private equity investors seem to prefer a fast exit, have lower bargaining incentives over higher offer prices, and leave more money on the table in a positive market environment. Agency conflicts between pre-IPO shareholders could be higher in hot markets and in cases of larger firm ownership by financial investors, which results in higher underpricing. An explanation for divergent interests could be higher flexibility in setting an offer price acceptable to potential buyers. However, this shareholder group owns fewer shares in hot market IPOs and pushes forward the decision to go public in comparatively cold months, which can be concluded from the results of table V. The tstatistics do not indicate high significance of ownership stakes variables or coefficient changes in hot months. Similar estimates are shown in regression models with the independent variables for owner participation (UP [21] and UP [22]). The more pre-IPO shareholders sell in terms of secondary shares, the higher are their wealth losses due to underpricing, and therefore the greater are their incentives to bargain for higher offering prices. In hot months, the coefficients signs for the classification of insiders and investors change. However, the overall effects remain negative, because the positive coefficients of the interaction term are lower than the negative values for “insider sales” and “investor sales”.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

85

Chapter II: Table X

Regression Models on Underpricing: Hot IPO Underpricing For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [19] UP [20] UP [21] UP [22] Proceeds Return Vola Volume Tech Media Dilution Participation Insider Investor Blockholder

0.003 (0.165) 0.186 (3.941)*** -1.367 (-3.527)*** 0.004 (1.566) 0.108 (2.582)*** 0.129 (1.676)* -0.025 (-0.370) -0.256 (-1.854)* 0.021 (0.291) -0.009 (-0.084) -0.124 (-1.407)

HotIR*Insider HotIR*Investor HotIR*Blockholder HotIR

0.045 (1.203)

0.005 (0.295) 0.191 (3.975)*** -1.288 (-3.265)*** 0.004 (1.452) 0.114 (2.670)*** 0.139 (1.776)* -0.016 (-0.231) -0.244 (-1.709)* 0.002 (0.031) -0.103 (-0.703) -0.086 (-0.917) 0.077 (0.459) 0.277 (1.272) -0.201 (-0.114) -0.053 (-0.354)

Proceeds Return Vola Volume Tech Media Dilution

-0.003 -0.204 0.190 (3.978)*** -1.433 (-3.491)*** 0.005 (1.841)* 0.120 (2.845)*** 0.138 (1.805)* -0.036 (-0.491)

-0.002 (-0.095) 0.187 (3.905)*** -1.447 (-3.496)*** 0.005 (1.726)* 0.120 (2.825)*** 0.137 (1.774)* -0.039 (-0.537)

-0.285 (3.208)*** -0.080 (-0.503) -0.405 (-2.568)***

-0.481 (2.293)** -0.093 (-0.433) -0.409 (1.775)* 0.271 (1.191) 0.031 (0.086) -0.030 (-0.121) 0.033 (0.658) -0.113 (-1.713)* 0.769 (1.773)* 0.145 4.778

Participation Insider Sale Investor Sale Blockholder Sale HotIR*Insider Sale HotIR*Investor Sale HotIR*Blockholder Sale

0.524 (1.371) 0.146

Intercept

R²

0.566 (1.410) 0.141

R²

0.047 (1.218) -0.113 (-1.746)* 0.783 (1.947) 0.144

F-Statistic

5.800

4.831

F-Statistic

5.945

Herfindahl Intercept

HotIR Herfindahl

86

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

In table XI the results for the classification of hot and cold markets according to IPO volume in the month of the sample firm’s offering are reported.17 The same conclusion can be drawn regarding the changes in owners’ interest in the different market cycles. The coefficient for financial investors’ stakes changes from negative to positive in months with more counts of IPOs than the median value. The tstatistics in UP [24] do not indicate high significance of these variables in explaining underpricing of shares when a firm goes public. When the sales of the three owner classifications are introduced as explanatory variables, the interaction term changes are not significant at conventional levels (UP [26]). Only the coefficient for “HotVolume*Investor Sale” changes sign from negative to positive, but the overall effect on initial return remains negative. The results confirm that agency conflicts and incentives for financial intermediaries show the most noticeable change in response to market phases. In periods of favorable market environment, these shareholders leave more money on the table when their stakes prior to the IPO are higher. Many previous studies (e.g. Franzke (2003:20), Tykvova/Walz (2007: 364)) have investigated the influence of several financial investors on the level of underpricing during Germany’s Neuer Markt, and have found positive dependency for these variables. These results confirm the findings of previous papers, but also support evidence for the theory that owners are likely to bargain more for higher offering prices with higher monitoring incentives in place in relatively cold or normal periods.

17

In the regression models the variable “volume” is no longer included as an explanatory variable as it controls for the similar market conditions.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

87

Chapter II: Table XI

Regression Models on Underpricing: Hot IPO Volume For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance.

Proceeds Return Vola IR Tech Media Dilution Participation Insider Investor Blockholder

UP [23]

UP [24]

-0.002 (-0.155) 0.187 (3.806)*** -1.252 (-3.139)*** 0.154 (2.487)*** 0.114 (2.729)*** 0.144 (1.868)* -0.033 (-0.455) -0.237 (-1.703)* 0.010 (0.140) -0.016 (-0.137) -0.149 (-1.655)*

0.001 (0.016) 0.186 (3.783)*** -1.211 (-3.031)*** 0.151 (2.448)*** 0.116 (2.766)*** 0.158 (2.001)** -0.037 (-0.529) -0.246 (-1.776)* 0.019 (0.300) -0.084 (-0.610) -0.070 (-0.768) 0.024 (0.083) 0.219 (0.651) -0.106 (-0.348) -0.036 (-0.125)

HotVolume*Insider HotVolume*Investor HotVolume*Blockholder HotVolume

0.012 (0.267)

Proceeds Return Vola IR Tech Media Dilution

UP [25]

UP [26]

-0.010 (-0.556) 0.192 (3.938)*** -1.328 (-3.157)*** 0.155 (2.464)*** 0.126 (2.991)*** 0.151 (1.961)** -0.039 (-0.525)

-0.010 (-0.513) 0.192 (3.923)*** -1.324 (-3.182)*** 0.156 (2.461)*** 0.125 (2.978)*** 0.152 (1.947)** -0.038 (-0.499)

-0.278 (-3.138)*** -0.033 (-0.211) -0.405 (-2.544)***

-0.247 (-1.541) -0.030 (-0.127) -0.338 (-1.654)* -0.062 (-0.311) 0.009 (0.026) -0.241 (-0.912) 0.029 (0.518) -0.111 (-1.698)* 0.879 (2.004)** 0.156 5.213

Participation Insider Sale Investor Sale Blockholder Sale HotVolume*Insider Sale HotVolume*Investor Sale HotVolume*Blockholder Sale

0.588 (1.476) 0.161

Intercept

R²

0.657 (1.584) 0.154

R²

0.023 (0.539) -0.111 (-1.731)* 0.880 (2.124)** 0.156

F-Statistic

6.440

5.398

F-Statistic

6.536

Herfindahl Intercept

HotVolume Herfindahl

88

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

IV.3.4 Pre-IPO Ownership and Underpricing with Positive Investor’s Information The hypothesis, that pre-IPO shareholders’ willingness to leave money on the table changes according to market phases, is not confirmed by any highly significant interaction term variables. Therefore, another classification for market and investor perception of the IPO is introduced, to reassess the results. Loughran/Ritter (2002: 424 et. seqq.) show that, with upward revisions of the offer price, owners profit from an unexpected wealth gain and are willing to leave relatively more money on the table. Hanley (1993:233) has also proven that the final offer price revision is made in response to private investors’ information, revealed during bookbuilding, as well as in response to overall stock market conditions. An upward revision of the preliminary offer price indicates higher initial returns for investors. According to the Benveniste/Spindt model (1989:347 et. seqq.), higher returns are seen as additional compensation for investors, revealing true positive information during the bookbuilding period. Ljungqvist/Wilhelm (2003:736) also consider offer price revision, and predict a positive relation to owner’s stakes and sales in the IPO. Pre-IPO shareholders are likely to bargain more aggressively for higher offer price revision when their wealth gains are higher due to larger stakes. In Germany, the possibility of price revision is limited, and final offer prices are normally set within a given price range (e.g. Löffler/Panther/Theissen (2005: 468), Aussenegg/Pichler/Stomper (2002: 3 et. seqq)). The reason for this is that participation in the bookbuilding process constitutes a binding offer by investors, which is not the case in the USA, and a higher offer price would be related to enormous effort on the part of the underwriter in repeating the whole road show process and so would delay the offering date. In this chapter the relationship between the final offer price and the preliminary offer price is used as an additional indicator of investor perception. To this purpose, the dummy variable “HotPrice” is introduced. This variable equals one if the offer price is higher than the preliminary price, which is defined as the midpoint of the bookbuilding price span, and otherwise it is zero. This sheds more light on offer price revision due to investor’s information signaled to underwriters. Although the scope for increasing the final offer price is very limited, issuers and underwriters are more likely to set the offer price at the upper end of the price range when they receive positive feedback. With somewhat negative or absent information of potential investors, the offer price is set below the issuer’s previous expectations. In this sample, the average bookbuilding range is 19%, limiting unexpected wealth gains for pre-IPO owners. The final price is set above the preliminary price in 60% of the sample IPOs. In table XII, the results for the regression models including the dummy variable and the interaction term with ownership proxies are reported. In UP [27], the con-

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

89

trol variables show the same results as before, and “HotPrice” indicates a positive relation to underpricing, significant at the 1% level, with a t-statistic of 5.944. IPOs with a final offer price above the midpoint of the file price range also show considerably higher initial returns after the first trading day. The R-squared also increases, which indicates greater explanatory power for the model in respect of the IPO underpricing data. The interaction term with the respective dummy variable is analyzed in UP [28]. First of all, the coefficients for the owner’s stakes are negative, including those for the percentage holdings of managers and supervisory board members. Interestingly, this differs from the results of UP [20] and UP [24]. The relationship between insider stakes and underpricing in offerings with positive investor’s perception changes in the opposite, positive, direction. However, the coefficient is relatively small, and so the overall affect remains negative. The findings for the changing bargaining interests of financial investors support the results shown in table X and XI. This investor group is associated with higher underpricing in IPOs with offer prices exceeding the midpoint of the bookbuilding span. Interpreting owners’ bargaining power is less straightforward. It can be presumed that the offer price is not set as high as possible (remaining within the price range), and therefore more money is left on the table, when financial investors own large stakes in the firm. Higher stakes held by blockholders and insiders suggest a tendency to bargain for the highest possible offer price (more unexpected wealth gain) and to monitor the underwriter so that lower initial returns are achieved. Another argument could be that agency conflicts among pre-IPO shareholders increase when financial investors own a large percentage of shares and the offer price is set relatively high in comparison to previous expectations. These financial intermediaries could be willing to benefit future potential customers or signaling investors and forego their own wealth increase. Furthermore, investors with positive perceptions of the IPO may pay more for the shares offered if financial investors own a large part of the firm’s stocks and certify the quality, which results in high price increases. The investigation into owner participation show similar results to the previous classification of hot and cold markets. First of all, the dummy variable for high offer prices is also significant, and is positively related to underpricing in UP [29] and UP [30]. The shareholder sales variables also show negative effects (UP [29]), because they have higher wealth losses due to underpricing when they sell more shares at the IPO. The t-statistics for “insider sale” and “blockholder sale” are significant at the 1% and 5% levels. Regarding the interaction term “HotPrice” in UP [30], monitoring incentives for insiders and blockholders also seem to reduce the amount of money left on the table. The “insider sale” variable is positive and the interaction term of this variable indicates a significant negative change for the

90

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

“HotPrice” IPOs. In the case of high insider sales, offer prices are set as high as possible so that share prices increase less after the first trading day. The opposite proposition can be made for the financial investor’s incentives. When the final offer price is set in the upper range of the bookbuilding span, these shareholders experience (unexpected) wealth gains and are willing to compensate investors by means of higher initial returns.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

91

Chapter II: Table XII

Regression Models on Underpricing: Price Level For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [27] UP [28] UP [29] UP [30] Proceeds Return Vola Volume IR Tech Media Dilution Participation Insider Investor Blockholder

-0.023 (-1.340) 0.179 (3.736)*** -0.728 (-1.865)* 0.001 (0.498) 0.105 (1.717)* 0.106 (2.731)*** 0.122 (1.595) -0.007 (-0.122) -0.176 (-1.329) -0.059 (-0.857) -0.032 (-0.294 -0.183 (-2.178)**

HotPrice*Insider HotPrice*Investor HotPrice*Blockholder HotPrice

0.221 (5.944)***

-0.021 (-1.197) 0.177 (3.665)*** -0.760 (2.050)** 0.001 (0.524) 0.111 (1.813)* 0.107 (2.771)*** 0.121 (1.575) -0.006 (-0.104) -0.201 (-1.427) -0.106 (-1.159) -0.083 (-0.955) -0.088 (-0.924) 0.084 (0.635) 0.107 (0.450) -0.133 (-0.862) 0.157 (1.249)

Proceeds Return Vola Volume IR Tech Media Dilution

-0.029 (-1.662)* 0.182 (3.819)*** -0.765 (-1.862)* 0.001 (0.617) 0.101 (1.642)* 0.114 (2.895)*** 0.125 (1.648)* -0.008 (-0.122)

-0.030 (-1.730)* 0.184 (3.814)*** -0.820 (-1.916)** 0.001 (0.513) 0.105 (1.688)* 0.116 (2.890)*** 0.124 (1.631)* -0.020 (-0.317)

-0.306 (-3.412)*** 0.071 (0.501) -0.306 (-1.903)**

0.023 (0.139) -0.026 (-0.244) -0.130 (-0.540) -0.394 (-2.041)** 0.266 (0.907) -0.371 (-1.422) 0.241 (5.792)*** -0.116 (-1.998)** 0.891 (2.212)** 0.226 7.672

Participation Insider Sale Investor Sale Blockholder Sale HotPrice*Insider Sale HotPrice*Investor Sale HotPrice*Blockholder Sale

R²

0.729 (1.822)* 0.213

0.738 (1.988) 0.218

R²

0.231 (6.769)*** -0.118 (-2.011)** 0.861 (2.141)** 0.222

F-Statistic

8.817

7.331

F-Statistic

9.277

Herfindahl Intercept

HotPrice Herfindahl Intercept

92

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

In the final regression models seen in tables XIII and XIV, owners’ intentions to bargain over the optimal offer price according to investor perception of the IPO and the market environment are presented. The dummy variables for the price level (“HotPrice”) and the hot market periods (“HotVolume” and “HotIR”) are applied to the ownership variables. Pre-IPO owners’ willingness to monitor underwriters in an IPO with high investor demand and in months with high IPO volume or high previous initial returns is estimated. The results in table XIII show estimates for the hot IPO classification according to previous underpricing. The insider stakes variable shows a positive coefficient, while the relation of investors and blockholder stakes to UP is negative. The simple interaction term with “HotIR” also indicates a negative relation between higher ownership stakes and initial returns realized after the firm becoming listed. However, significant positive coefficient changes at the 1% and 5% level of significance are obtained for the “HotPrice*HotIR*Insider” and “HotPrice*HotIR*Investor” variables in UP[31] and UP[32]. This suggests a positive change in the relation between underpricing and insiders’ and investors’ stakes in IPOs with favorable investor perception and high previous underpricing. Also, the high coefficient values indicate considerable economic relevance. On one hand, owners are willing to leave more money on the table under these conditions and do not bargain for the highest possible offer price in the upper half of the price range (or simply the offer price cannot be revised upwards). On the other hand, higher ownership by insiders and financial investors could be interpreted positively by new investors in the given market environment, meaning that they are still willing to pay higher prices than the highest possible offer price, thereby increasing initial returns. The presence of blockholders shows no significant statistic and economic effects in the given interaction terms. Owner participation and willingness to monitor underwriters in setting the optimal offer price are also analyzed. Here the coefficients change from negative to positive for insiders’, investors’ and blockholders’ sales with the two interaction dummy variables in UP [33] and UP [34]. However, only the t-values for the investors’ sales indicate a significant difference at the 5% level in both regression models. Pre-IPO owners are associated with higher underpricing where a high percentage of secondary shares in IPOs is characterized by offer prices in the upper preliminary price range in combination with high previous underpricing. Owners seem to have less incentive to bargain over high offer prices, or they are not able to revise the offer price upwards, resulting in higher initial returns. In table XIV the results for hot categorization according to IPO volume in the given month are presented. The results confirm previous estimates, and the same conclusion can be drawn. Significant changes in the relations of ownership structure and underpricing are found in a positive market environment with positive inves-

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

93

tor’s information. IPOs with a final offer price above the midpoint of the file price range in a hot market have higher initial returns when the owners’ stakes and sales are higher. The relation is negative under less positive market conditions, which is also suggested by Ljungqvist/Wilhelm (2003). Agency conflicts and insiders’ and financial investors’ bargaining interests are especially likely to change. In most of the results, the overall blockholders’ interests are negatively related to initial returns. Firms’ ownership structures have changed over time, but the differences in the IPO markets cannot necessarily be explained by increased clustering of preIPO shareholdings. Evidence is found for the hypothesis that owners’ objectives in terms of offering prices change according to market conditions.

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

94

Chapter II: Table XIII

Regression Models on Underpricing: IPO Underpricing and Price Level For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [31] UP [32] UP [33] UP [34] Proceeds -0.001 -0.001 Proceeds -0.001 -0.002 (-0.023) (-0.031) (-0.008) (-0.124) Return 0.193 0.193 Return 0.189 0.189 (3.924)*** (3.827)*** (3.949)*** (3.939)*** Vola -1.112 -1.115 Vola -1.409 -1.414 (-2.876)*** (-2.858)*** (-3.377)*** (-3.412)*** Volume 0.003 0.003 Volume 0.005 0.005 (1.373) (1.394) (1.755)* (1.775)* Tech 0.117 0.117 Tech 0.121 0.115 (2.888)*** (2.858)*** (2.893)*** (2.655)*** Media 0.137 0.137 Media 0.135 0.130 (1.737)* (1.704)* (1.736)* (1.659)* Dilution 0.001 0.001 Dilution -0.038 -0.046 (0.011) (0.015) (-0.521) (-0.652) Participation -0.221 -0.221 Participation (-1.580) (-1.569) Insider 0.003 0.007 Insider Sale -0.506 -0.486 (0.049) (0.097) (-2.841)*** (-2.298)** Investor -0.107 -0.100 Investor Sale -0.192 -0.078 (-0.873) (-0.679) (1.161) (-0.358) Blockholder -0.082 -0.073 Blockholder Sale -0.447 -0.416 (-1.006) (-0.766) (-2.129)** (-1.797)* HotIR* Insider -0.013 HotIR*Insider Sale -0.033 (-0.074) (-0.079) HotIR*Investor -0.022 HotIR*Investor Sale -0.477 (-0.116) (-1.545) HotIR*Blockholder -0.034 HotIR*Blockholder Sale -0.161 (-0.192) (-0.428) HotPrice*HotIR* 0.192 0.190 HotPrice*HotIR* 0.350 0.356 Insider (2.657)*** (2.209)** Insider Sale (1.779)* (0.947) HotPrice*HotIR* 0.597 0.600 HotPrice*HotIR* 0.814 1.145 Investor (3.482)*** (3.774)** Investor Sale (1.989)** (2.820)** HotPrice*HotIR* Blockholder HotIR

0.080 (0.677) -0.117 (-2.783)***

0.093 (0.720) -0.101 (-0.662)

0.544 (1.417) 0.190

Intercept

R²

0.546 (1.386) 0.190

R²

0.087 (0.358) 0.014 (0.327) -0.095 (-1.492) 0.715 (1.687)* 0.152

F-Statistic

6.602

5.463

F-Statistic

5.053

Herfindahl Intercept

HotPrice*HotIR* Blockholder Sale HotIR Herfindahl

0.203 (0.506) 0.027 (0.527) -0.093 (-1.467) 0.753 (1.712)* 0.154 4.248

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

95

Chapter II: Table XIV

Regression Models on Underpricing: IPO Volume and Price Level For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP [35] UP [36] UP [37] UP [38] Proceeds

HotPrice*HotVolume* Insider

0.264 (4.741)***

-0.005 (-0.333) 0.184 (3.953)*** -1.119 (-2.841)*** 0.111 (1.796)* 0.122 (2.978)*** 0.157 (2.028)** -0.048 (-0.765) -0.211 (-1.518) 0.034 (0.510) -0.079 (-0.510) -0.053 (-0.567) -0.155 (-0.513) 0.039 (0.122) -0.050 (-0.160) 0.303 (5.312)***

HotPrice*HotVolume* Investor

0.524 (3.169)***

0.414 (2.693)***

HotPrice*HotVolume* Investor Sale

0.639 (1.199)

0.857 (1.684)*

HotPrice*HotVolume* Blockholder

0.030 (0.283)

-0.006 (-0.056)

HotPrice*HotVolume* Blockholder Sale

-0.134 (-0.491)

0.131 (0.303)

HotVolume

-0.167 (-4.324)***

-0.067 (-0.228)

HotVolume

0.003 (0.077) -0.100 (-1.563) 0.841 (1.957)** 0.160 5.361

0.029 (0.029) -0.102 (-1.582) 0.872 (1.972)** 0.164 4.559

Return Vola IR Tech Media Dilution Participation Insider Investor Blockholder

-0.007 (-0.406) 0.186 (4.017)*** -1.146 (-2.912)*** 0.117 (1.886)* 0.122 (2.988)*** 0.157 (2.046)** -0.053 (-0.850) -0.210 (-1.482) 0.002 (0.035) -0.062 (-0.490) -0.058 (-0.612)

HotVolume* Insider HotVolume*Investor HotVolume*Blockholder

Herfindahl Intercept R² F-Statistic

Proceeds Return Vola IR Tech Media Dilution

0.655 (1.642)* 0.218 6.505

-0.010 (-0.492) 0.191 (3.926)*** -1.317 (-3.156)*** 0.149 (2.321)** 0.124 (2.929)*** 0.148 (1.875)* -0.052 (-0.710)

-0.365 (-2.589)*** -0.148 (-0.825) -0.400 (-2.106)**

0.113 (0.569)

-0.251 (-1.557) -0.034 (-0.142) -0.349 (-1.693)* -0.503 (-1.052) -0.391 (-1.285) -0.335 (-0.879) 0.490 (1.043)

Participation Insider Sale Investor Sale Blockholder Sale HotVolume*Insider Sale HotVolume*Investor Sale HotVolume*Blockholder Sale HotPrice*HotVolume* Insider Sale

Herfindahl 0.709 (1.750)* 0.216 7.749

-0.007 (-0.379) 0.190 (3.903)*** -1.335 (-3.147)*** 0.150 (2.353)** 0.128 (3.069)*** 0.155 (1.997)** -0.058 (-0.802)

Intercept R² F-Statistic

96

V

HOW DO PRE-IPO SHAREHOLDERS DETERMINE UNDERPRICING?

Conclusion

This chapter offers interesting insights into the pricing decisions and the related levels of underpricing of IPOs during different market periods. The IPO environment is classified into hot and cold market phases according to average underpricing of previous offerings and the number of IPOs per month. The sample is split into two sub-periods (1997-2001 and 2002-2007) to detect relevant developments in the German stock exchanges and newly listed firms over time. First of all, firm characteristics do not indicate considerable differences between hot and cold market phases. Not only between 1997 and 2001 did a large number of young start-up companies in the technology and internet sector go public: The hot periods also offer a window of opportunity for all types of companies to raise relatively cheap equity with comparatively low participation by pre-IPO shareholders. Interesting results are also obtained in terms of ownership structure, which is classified into three categories of shareholders: insiders, investors and blockholders. Insiders, such as managers and supervisory board members, own higher equity stakes in IPOs going public in hot months, whereas financial investors, e.g. venture capitalists or private equity firms, have more shares in cold than in hot IPOs. However, the Herfindahl index shows higher values in hot market phases, which indicates a more clustered ownership in favorable market conditions associated with higher initial returns. This result contradicts the hypothesis put forward by Ljungqvist/Wilhelm (2003). Furthermore, the regression analysis confirms assumptions that determinants and explanatory variables have changed significantly over time; the level of underpricing did not simply decrease in general after the closing of the Neuer Markt. The ownership structure variables indicate changes in bargaining interests in terms of pre-IPO shareholders’ offering prices. Financial investors are particularly willing to leave more money on the table in hot IPOs. When the final offer price in relation to the initial bookbuilding range is considered as another criterion for estimating positive investor perception about the firm going public, a significant positive relation change is estimated. During positive market phases and high offer prices within the given price range, larger insiders’ and investors’ holdings of pre-IPO shares determine higher levels of initial returns. The interpretation of the results is less straightforward, as in Germany, unlike in the USA, an upward offer price revision above the upper limit of the bookbuilding range is not a common mechanism. Although potential shareholder agency conflicts seem to be especially relevant in different market cycles, the newly issued shares and offer price should reflect the firms’ value. The following chapter investigates this step in the IPO process in more detail, and focuses on variability in valuation according to market conditions.

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

97

Chapter III: Do “Herding” Effects on Firm Multiples Determine IPO Valuation? I

Introduction

When firms decide to complete an initial public offering their shares have to be priced to allow new investors to be found. Normally stock prices are the best indicator of investors’ expected firm value, unfortunately these are not available for private firms. Therefore, investment banks use several methods to decide on an appropriate offer price, which reflects the firm’s value as well as the market’s willingness to purchase newly issued shares. Previous literature suggests that in the majority of cases underwriters of IPOs apply comparable multiples of firms already listed, such as market-to-book (MB) and price-earnings (PE) ratios, in conjunction with firm’s accounting information, to determine expected market value. The multiple ratios of publicly traded stocks, however, may vary over time according to their valuation by the market. For example, technology firms during the dot-com bubble showed extremely high market values and multiple ratios. As underwriters and issuers orientate their pricing decisions to industry related public firms, the IPO valuation is expected to change correspondingly. In the German stock market during the boom phase in the technology and internet sector, initial public offerings were also valued very highly in terms of MB and PE ratios. This chapter investigates the relationship between IPO and market valuation, to gain more insight to these observations on the German stock market between 1997 and 2007. The research question should provide a conclusion as to whether changes in IPO multiples can or cannot be explained by industry related market-to-book and price-earnings ratios. The reason for this argument is also analyzed. Underwriters and investors are expected to focus more on comparable firm multiples in periods with favorable market conditions and especially higher IPO volume, as more relevant information is available. They are likely to show some form of “herding” behavior regarding these positive valuation criteria and so neglect the firm’s accounting and profitability characteristics. The focus on multiple ratios also constitutes a common valuation factor for participants, which is likely to reduce information asymmetries between issuers and the market, and to positively affect the IPO decision. These research questions have not been considered and proven in any existing literature, although several theoretical models exist which suggest a form of information spillover in periods with high IPO volume and positive effects on first market prices.

98

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

The results of several empirical tests in this chapter are surprising. The valuation of IPOs does not directly correspond to the MB and PE multiples of publicly traded stocks in the same industry. Neither can any positive valuation effects of previous IPOs and their multiple values be confirmed. The regression models analyze explanatory variables such as financial characteristics, industry multiples and market environment. In summary, in months with only a few public offerings, or in cold periods, firm’s accounting information and industries’ MB and PE ratios can explain the IPO market value. In the results for hot months, firm’s characteristics are less relevant than the value driver, in terms of overall stock market performance before the IPO. Information spillover cannot be related to the multiple ratios under consideration. Instead, the stock price changes, in an environment with high IPO volume, encourage underwriters to increase valuation for newly issued stocks, disregarding the firm’s financial characteristics. Also assumptions about asymmetric information distribution between issuers and investors as well as the decision to go public are considered in a new context and add important findings to previous investigations. Chapter III is structured in four sections: section II presents the related literature and theory on which this paper is based, and the reasons behind the development of the two main questions are presented. In section III, the research design is described. As well as explanations of the sample selection, important variables for the regression models are discussed with reference to previous literature. The empirical results follow in section IV. First, the descriptive statistics of the sample firms are presented. Section IV.2 analyzes multiple values over the time period from 1997-2007 in more detail. In IV.3, the regression models are applied to IPOs’ market values, where differentiation between market phases shows the most relevant estimates (chapter III: IV.3.2). Also in IV.3.3, findings are approved, with assumptions of a different market environment related to asymmetric information distribution.

II Related Literature and Development of Hypotheses Investment banks, acting as IPO underwriters, have to consider all relevant and available information about the firms going public as well as estimating the market’s perception and demand, in order to set an appropriate offer price for new shares. The valuation of firms going public is very complex, as the firm’s characteristics are not the only significant consideration in setting the initial price. The existing literature about IPO valuation considers underwriters’ common tech-

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

99

niques in valuing the issuing firm. For example, Roosenboom (2005: 16 et. seq.) and Deloof/De Maeseneire/Inghelbrecht (2002: 5 et. seq.) find that the discounted cash flow (DCF) model and the dividend discount method (DDM) are very frequently applied in the French and Belgian IPO markets. With the DCF model the firm’s expected cash flow (EBITDA adjusted for changes in working capital and net capital expenditure) over a given planning horizon is discounted by a rate that reflects the firm’s risk. With the DDM the expected dividends, assumed to be a constant portion of net profits, are discounted by the costs of equity. The problem is in appropriately estimating cash flow and dividends over a 5-10 year period. Cogliati/Paleari/Vismara (2008: 14) confirm that with the expected growth rates of cash flows for Europe’s IPOs between 1995 and 2001 were about 20% higher than actually achieved in the 5 years following the IPO. However, the accounting information used most frequently by underwriters is that relating to comparable firm multiples such as the price-earnings ratio, market-tobook ratio and price-to-sales ratio (Kim/Ritter (1999:410)). For example, the price of an issue is decided by multiplying the firm’s earnings per share by the average of PE ratios of comparable firms already traded publicly and valued by the market. The accuracy of PE multiples increases if comparable firms are selected on the basis of industry segments, firm’s related risk and expected earnings growth rates (Alford (1992:.98, 102)). In terms of IPOs, however, Kim/Ritter (1999: 410 et. seq.) find that this seldom leads to precise valuation. Accounting information and comparable multiples can be chosen merely as a benchmark, but additional information about the market’s demand is included in the preliminary and offer prices. On the other hand Beatty/ Riffe/ Thompson (2000: 8 et. seqq) and Bartov/ Mohanram/ Seethamraju (2002: 326 et. seq.) focus on accounting information such as earnings and book value of equity and revenue, finding significant explanatory power for market valuation and offering prices. However, firms’ MB and PE ratios show major movements industry-wide, as they reflect the industry’s growth perception and investors’ valuation of stocks. For example Ofek/Richardson (2002: 269 et. seqq.) show that the PE ratio of internet firms in the dot-com bubble was exceptionally high. Core/Guay/Van Buskirk (2003: 54) also find an increasing MB ratio over time (1975-1999), where the changes cannot be explained by divergent firm characteristics such as income, spending on research and development (R&D), capital expenditures or sales growth. It would be reasonable to assume that this also affects the multiples of IPOs. For example, in Germany, firms going public during the profitable period of the new market or Germany’s Neuer Markt (1997-2001) showed very high MB and PE ratios compared to IPOs in the following years (2002-2007), during which

100

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

IPO firms’ multiples decreased sharply. Interestingly, the firm’s characteristics and important accounting information did not change significantly between 1997 and 2007. The Neuer Markt period was also characterized by enormous initial returns after the first trading day, and by a high volume of IPOs. So, overall, this period of the German stock market has been described as a hot issue period. Since 2002, however, the IPO market has slowed down and turned into a cold phase. As well as the significantly lower PE and MB multiples, the IPO volume and the underpricing of these newly issued shares have also decreased. The reasons for those IPO waves with high IPO volume and high underpricing (or high initial returns) have been investigated in more detail. Pástor/Veronesi (2006:1714) and Boehmer/Ljungqvist (2004:2 et. seqq.) find that firms’ expected profitability, market return and uncertainty all affect the decision to go public. For example technological development can affect the capital market in those respects (also Maksimovic/Pichler (2001:459 et. seqq.)). Several models about the clustering of IPO are also related to the Myers/Majluf (1984:216 et. seq.) model, hypothesizing that more firms go public when the adverse selection costs of equity issue are low, e.g. when investors interpret the IPO decision more favorably or the information discrepancy between issuers and investors is reduced. For example, Subrahmanyam/Titman (1999:1072) model the dependency relationship between the IPO decision and investors’ information acquisition. They come to the conclusion that higher liquidity in capital markets and informational efficiency make public equity markets more attractive for firms. Hoffmann-Buchardi (2001: 355 et seq.) argue that IPO valuation is related to firm- and market-wide factors. Positive information about market perception in offer prices of IPOs affects firms’ decisions to go public. Investors are also able to “free-ride” on information about previous IPOs. Higher market valuation can therefore be achieved by following IPO firms, because they do not have to compensate investors’ information production when the signal from previous IPOs is relatively precise. Alti (2005:1106, 1131) puts forward a similar idea: Hot IPO markets are a result of information spillover. High IPO offering prices are an indicator of investor’s private information, which reduces the information asymmetry between issuers and potential investors. This also reduces uncertainty in IPO valuation and supports the decision to issue equity. However, Froot/ Scharfstein/ Stein (1992:1463) show that information spillovers or investors’ information “herding” can also result in market inefficiencies. They argue that short-term speculators are particularly likely to focus on only one source of information in making their trading decisions. “These informational spillovers can be so powerful that groups of traders may choose to focus on very poor quality data, or even on completely extraneous variables that bear no relation at all to fundamentals.“ Empirical investigations (e.g. Lowry/ Schwert (2002:

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

101

1174), Lowry (2003:31 et. seqq.), Benveniste/ Ljungqvist/ Wilhlem/ Yu (2003: 591)) support the argument that the decision to go public is related to previous IPOs and that firms are more likely to go public if the market valuation is especially high. This idea about information spillover in the IPO decision can also be linked to the IPO valuation theory. As previous literature and market observations indicate, IPO multiples are likely to change with market conditions. The market valuation of IPOs seems higher in a favorable equity market environment, or hot issue phase. Considering these arguments, this chapter intends to answer the first important question: [1] Can changes in IPO valuations be explained by multiples of comparable industry related firms? Furthermore, the second aim of this research is related to the information spillover models presented above: [2] Are there “herding” effects on IPO valuation and comparable multiples in periods with high IPO volumes? This would indicate issuers, underwriters and investors focusing more on high comparable multiples of similar firms when valuing an IPO than on other relevant accounting information and on the firm’s characteristics. During cold issue phases with lower IPO volume, valuation proxies relating to the firm’s financial statements should have more explanatory power than the industries’ MB and PE ratios. Also, because less information of previous IPOs in the same industry is available, investors’ demand is harder to estimate and the corresponding insecurity about the appropriate offering price is greater. Furthermore, this chapter’s investigation is related to the asymmetric information hypothesis put forward by Myers/Majluf (1984). They argue that more firms are likely to issue equity during periods with lower informational asymmetries between managers of a firm and investors. Therefore, in periods with high IPO volume, investors are expected to interpret the firm’s information and decision to go public more favorably due to higher comparable industry multiples. Furthermore, the discrepancy between investors and issuers are assumed to be lower, as common valuation factors carry more weight than firm-specific information. A previous study by Bayless/ Chaplinsky (1996: 271, 274) is closely related to this idea. They investigated the decisions concerning seasoned equity offerings (SEOs), relative to differing market phases. They compare announcement date prediction errors for equity issues in hot and cold market period, finding evidence that identical firms experience less negative market returns where the announcement date is in a hot market, compared to those announced in normal and cold markets. However, the differences are not attributable solely to differences in market conditions, and only in cold periods are firm characteristics significant in explaining prediction errors. Investors seem to give more weight to firm-specific factors in cold markets, while in periods with higher volume of equity issues a greater portion of SEO valuation seems to be re-

102

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

lated to public information available to issuers and investors. Koop/Kai (2001:388, 391) also consider effective market valuation, but they make a distinction between IPOs and SEOs. Their results indicate more efficient pricing in SEOs than in IPOs. However, they do not find evidence of cold/hot issue periods and market conditions, e.g. indicators of economic upturns, changes in stock prices or stock price volatility, having any significant effect on valuation of equity offerings.

III Research Design III.1 Sample Selection and Data Sources The sample contains IPOs from January 1997 to December 2007 listed on segments of the Frankfurter stock exchange and includes all new issues as well as listings accompanied by raising new equity in all stock segments. Private placements and the transfer from one stock exchange or market tier to another are excluded, as well as public offerings of Banks and Reits (11 IPOs) due to differences in financial accounting. Between 1997 and 2007, initial public offerings of 569 firms have been completed. In the early phase (1997-2001) the numbers of equity issues have been considerably higher with 423, compared to 146 IPOs between 2002 and 2007. For a detailed investigation, complete information has been available for 483 IPOs. The information is obtained from Deutsche Börse AG, which provides information about all offerings in terms of new issues, listings, transfers etc. Additionally, the primary market statistics for the Regulated and Open Market deliver the IPO date, offer price, first price at the beginning of trading, bookbuilding span as well as the volume of the issue and market capitalization. Another important source has been Thomson Financial’s Datastream. This database provides the closing price on the first trading day after the IPO and information on percentage price changes, volatility, monthly market-to-book ratios and price-earnings ratios of all shares traded in the German stock market. Also the industry classification (ICB: “Industry Classification Benchmark”) of each IPO was obtained from this database. The information from the financial statements (e.g. net income, assets, intangible assets, equity, total debt, capital expenditure, return on equity, earning per share etc.) is from the Database of Reuters Knowledge, where all firm reports are available and standardised in Euros. The original accounting information closest to the offering date is used for the analysis.

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

103

III.2 Methodology and Definition of Variables To investigate the degree to which accounting information and comparable industry multiples affect IPO valuation, several ordinary least square regression models are analysed in this chapter. The independent variable is the total market value of the IPO firm, which is calculated by multiplying the total number of shares after the IPO (including primary shares) by the following three prices: preliminary price, offer price and share price after the first trading day. Most of the IPOs are sold through the bookbuilding method, where a preliminary price range is set by the issuer and underwriter. As part of the roadshow before the IPO, underwriters offer shares to potential investors, who indicate their demand by their bidding prices within the preliminary range and by the amount of shares they are willing to buy. The preliminary price is calculated as the midpoint of the bookbuilding range, and is the best indicator for the underwriter’s perception of the value of an IPO. Furthermore, the total firm value is calculated with the final offer price. The offer price and its difference from the preliminary price together indicate the investors’ demand for an IPO. A higher offer price means that the information obtained during the bookbuilding phase or roadshow has been favourable. In Germany, however, underwriters are not able to set offer prices higher than the upper price of the bookbuilding range, as the investors make a binding bid during the roadshow period. To set a higher offer price, the whole process would have to be redone, which would delay the IPO. The total market value is therefore also calculated by multiplying all outstanding shares by the share price after the first trading day. As IPOs are often underpriced, meaning that the offer price is deliberately set lower than the price which could be obtained in the market, it is a good indicator of the market’s perception and demand for these newly issued shares. Previous studies have used a variety of proxies to measure the pricing and valuation of an IPO. For example Kim/Ritter (1999: 417) use PE and MB ratios of the IPO firms. Kim/Krinsky/Lee (1995:456) regress the explanatory variables on the share price. In an offering, however, underwriters and issuers are probably concerned with the total valuation of the firm or proceeds and not solely with the offer price. This model necessarily includes the issue proceeds as an independent variable. Trueman/Wong/Zhang (2000: 142) and Core/Guay/Van Buskirk (2003:49 et. seq.) also use the total market value of equity, as well as a deflated model in which market value and explanatory variables are scaled relative to the book value of equity. Beatty/Riffe/Thomson (2000: 10) prove, for a sample of IPOs, that market value deflation by book value of equity or revenue increases the explanatory power of the independent variables (earnings and book value of equity per share) of the regression models (for discussion see: Brown/Lo/Lys (1999: 86 et seqq.)). Furthermore, they apply a log transformation to their value variables and also

104

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

show an improved model fit. This transformation seems appropriate, as a nonlinear or convex relationship between firm value and accounting information, as discussed by Burgstahler/Dichev (1997: 189 et. seqq.), Fischer/Verrecchia (1997: 519 et. seqq.) and Hand (2000: 13 et. seq.) can be assumed. Additionally, the influence of potential outliers and of heteroscedasticity can be reduced by using the natural logarithm. For this reason, this log-transformation is also applied to the three different measures of market value in this study, as well to all value variables included in the regression models. The analysis takes into account the firm’s financial information, in the form of several independent variables on the right hand side of the regression model. First of all, the value of the firm’s total assets is a relevant indicator of its size, level of operations and perceived investor risk in the offering. This proxy is used less often in the valuation models, but the studies related to IPO underpricing often find a significant negative relation between this variable, as a proxy for risk, and the initial return (see Jenkinson/Ljungqvist (2001: 70 et.seq.), Koop/Li (2001:386)). Here the correlation to IPO values is expected to be positive. The amount of intangible assets in relation to total assets of the firm indicates the difficulty of valuing a firm based on accounting variables and substantial positions. Furthermore, the ratio of intangible assets to total assets (“intanratio”) is assumed to be related to the industry or the innovativeness of the firm, as it indicates spending on R&D e.g. on patents or licences. Small and technology orientated IPOs are expected to have a higher proportion of intangibles assets than older companies in the same industry segments. Therefore the effects on market valuation, in this context, can be either positive or negative. Furthermore, the book value of equity is, included in the regression model as an explanatory variable, according to previous studies (e.g. Bartov/Mohanram/Seethamraju (2002: 326)). As the dependent variable reflects the market value of equity, and the influence of comparable industry multiples (MB and PE ratio) is of major interest, the book value is included as a control for differences in size between firms, and for the effects on valuation (see Kim/Ritter (1999: 416 et. seq.)). A related variable, “leverage”, indicates the firm’s amount of debt in relation to its sum of total debt and equity. Firms with higher debt burdens are expected to have a higher default risk and chance of bankruptcy. Additionally, a higher debt burden means higher interest payments, which is likely to reduce the possible dividends for shareholders and is therefore negatively valued by potential investors (see Jensen (1984: 324 et. seqq.), Koop/Li (2001:386)). Additionally, several variables are included to reflect the profitability of the firm and are therefore very important for estimating present and future cash flows, which is also the baseline for DCF methods in IPO valuation. The ratio of earnings

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

105

per share (“EPS”) sets the operating income (before depreciation and amortization) of the firm in relation to total shares outstanding (see also Kim/Krinsky/Lee (1995: 456), Bartov/Mohanram/Seethamraju (2002: 325)). Another ratio is included by introducing the return-on-equity (“ROE”) ratio for the year in which the IPO took place. This ratio sets a firm’s profit (net income) in relation to the book value of common equity. This shows potential shareholders, when previous ratios are extrapolated into the future, the profit that can be achieved with the invested capital. The difference between operating income and net income is calculated by adding/subtracting non-operating income, interest expenses and taxes. Other previous papers also use growth rates of sales, total operating income and earnings as profitability measures and expected value drivers for IPOs (e.g. Beatty/Riffe/ Thompson (2000:11), Core/Guay/Van Buskirk (2003:48), Boehmer/Ljungqvist (2004: 15). Kim/Ritter (1999: 434) also suggest that the comparable firm multiples for IPO valuation should be adjusted to different levels of profitability (e.g. sales and operating cash flows) to increase the predictability of a firm’s market valuation. Furthermore, the accounting information regarding the firm’s capital expenditure is included as an explanatory variable (“capex”). This expenditure is an indicator for future benefits in terms of firm value and expected earnings, as they indicate the firm’s investments in assets. Core/Guay/Van Buskirk (2003:48) and Kim/Krinsky/Lee (1995: 459) include this variable in analyzing (IPO) valuation. The first regression model [1] can be formalized in the following equation, where “MV(Price)” is calculated using the three different prices described above. For a detailed definition of the variables see table I in this chapter. The estimated coefficients are indicated by the respective β and the term ε is the random error variable of the regression model. (1): MV(Price) = β1 + β2(Assets) + β3(Intanratio) + β4(Equity) + β5(Leverage) + β6(Capex) + β7(ROE) + β8(EPS) + ε The major interest of this research is the effect of industry multiples and particularly their influence on valuation and accounting information. For this reason, I introduce the multiples “MB” and “PE” to the second regression model. “MB” stands for the market-to-book value and is defined as the market value of common equity divided by the balance sheet value of common equity. “PE” is the variable for the price-earnings ratio, defined as the price for the stock divided by consensus forecast of earnings per share for the next financial year. Kim/Ritter (1999: 430) have shown that the earnings forecast for the following year is a better multiple value for estimating IPO prices than historical- and current year’s forecasted earnings. These ratios were obtained from Thomson Financial Datastream for every firm whose shares were traded on the German stock market for each month between 1997 and 2007. These firms are categorised by ICB industry, corresponding

106

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

to the classification of the IPOs. For each sample IPO, the multiples are calculated as the average MB and PE ratios of all tradable shares of the same industry in the month before the offering date. As I do not match IPOs and firms by a range of characteristics, this method allows the overall industry firm valuation to be ascertained, as well as the broader effects on IPO offer prices, which is expected to be positively correlated. With the same method two further variables are created: “Vola” and “perform”. The first term, “vola”, is the mean of the 3-month moving average of historical volatility of all tradable firms’ stocks for each industry classification. This variable is included to provide a control for the valuation uncertainty of the market. Pástor/Veronesi (2005: 1720) argue that more firms go public when uncertainty about the future profitability of the industry is high, but also that the valuation in terms of market-to-book value increases with higher volatility of market prices. In an empirical investigation of this model by Boehmer/Ljungvist (2004:9), the volatility of stock returns in the sample firm’s industry is used. Koop/Li (2001:388) introduce the daily S&P return variance prior to the offering as a “misevaluation” factor for efficient IPO and SEO prices. The second term, “perform”, is the variable of the average 3-month percentage price change of all stocks traded in the month before the IPO, also classified by the ICB industries according to the sample firms. Pástor/Veronesi (2005:1720) also suggest that IPO volume is dependent on recent market returns, as it indicates low risk aversion in investors, and higher market valuation of firms. The recent stock price changes are more relevant than the price levels of the markets. The majority of existing literature regarding IPO underpricing shows that higher stock market prices before the offering increase the initial returns on the newly issued shares, after the first trading day (e.g. Loughran/Ritter (2002: 426 et. seqq.)) This variable is therefore included as a potential valuation driver; the effects on accounting information and the comparable firm multiples are especially interesting in terms of “information spillover”. The regression model is expanded in two steps. At the second stage (2) the MB and PE ratios are included as independent variables. The third equation (3) also includes the macroeconomic factors, which are likely to increase valuation within a positive market environment. Furthermore, table I gives on overview of the variables and definitions.

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

107

(2): MV(Price) = β1 + β2(Assets) + β3(Intanratio) + β4(Equity) + β5(Leverage) + β6(Capex) + β7(ROE) + β8(EPS) + β9(PE) + β10(MB) + ε (3): MV(Price) = β1 + β2(Assets) + β3(Intanratio) + β4(Equity) + β5(Leverage) + β6(Capex) + β7(ROE) + β8(EPS) + β9(PE) + β10(MB) + β11(Vola) + β12(Perform) + ε

108

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

Chapter III: Table I

Definition of Variables Name

Definition

Assets

Natural logarithm of total assets.

Intanratio

Intangible assets divided by total assets.

Equity

Natural logarithm of book value of total equity.

Leverage

Total debt divided by the sum of total debt and equity.

Capex

Natural logarithm of capital expenditure.

ROE

Net income divided by book value of common equity.

EPS

Basic earnings per share excluding extraordinary items. Monthly average of price-earnings ratio of all tradable shares in Germany in the related industry sector of the IPO. Average PE ratio one month before the offering date is used. Monthly average of market-to-book ratio of all tradable shares in Germany in the related industry sector of the IPO. Average MB ratio one month before the offering date is used. 3-month moving average of the volatility of all tradable shares in Germany in the related industry sector of the IPO. Average volatility one month before the offering date is used. Percentage price change over last 3 month of all tradable shares in Germany in the related industry sector of the IPO. The average 3-month price change one month before the offering date is used. Natural logarithm of total number of shares (including primary shares) at the IPO multiplied with the midpoint of the bookbuilding range. Natural logarithm of total number of shares (including primary shares) at the IPO multiplied with the offering price. Natural logarithm of total number of shares (including primary shares) at the IPO multiplied with the last price after the first trading day. Dummy variable which equals one if the IPO took place in a month with less IPOs counts than the median, otherwise it is zero.

PE MB Vola Perform MV(Preliminary) MV(Offer) MV(Last) Cold

IV Empirical Results IV.1 Descriptive Statistics of Firm and Market Characteristics An empirical investigation of changes in financial statements, comparable multiples and IPO valuation requires first of all a univariate analysis of the relevant variables. The descriptive statistics show the characteristics of IPO firms and of market environments during different issue periods. Several classifications and definitions of hot and cold IPO phases can be found in recent literature: Periods with high IPO volume (especially in one industry segment) or periods with high underpricing can be considered as “hot”. For example, Helwege/Liange (2004:542 et seqq.) investigate how firms in both periods differ, and which alternative char-

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

109

acterization of hot and cold markets is appropriate. They find that hot market firms are not necessarily small start-up firms clustered in certain industries. However, categorization by the level of IPO underpricing shows more distinct differences in firm characteristics. Bayless/Chaplinsky (1996: 260, 265 et. seqq.) compare SEO announcements in different market cycles, and their classification of hot and cold periods is based on quartile rankings of issue volume. When the highest quartile is exceeded in three consecutive months, the period is defined as “hot”; where the issue volume falls below the lowest quartile in three consecutive months, the period is defined as “cold”. Surprisingly, they do find significant differences in financial information (e.g. free cash flow, return on assets, total assets, leverage) in the firms issuing equity, but no significant differences in market conditions such as stock return and price-earnings ratios of traded shares. Because of the smaller sample size in this paper, the differentiation between hot and cold market periods is based on the median value of IPO volume per month. If the total number exceeds the median value of 10 IPOs per month, then the month is considered as “hot”, otherwise as “cold”. If the offering is in a month with less than 10 other public offerings, it is considered to be an IPO in a cold issue period. The differences in firm’s accounting information and profitability measures are presented in table II.18 The mean and median values for the complete IPO sample as well as for both sub-samples are reported, as are the p-values of the t-test of equality of the mean and the Wilcoxon-Mann-Whitney test. The p-value of the first test indicates the probability of the null hypothesis that the mean of both samples are equal. While the Wilcoxon-Mann-Whitney test (or Wilcoxon rank-sum test) estimates the null hypothesis that the two samples come from the same distribution and the p-value shows the probability of the hypothesis’ validity. The comparison shows that IPOs in hot periods are smaller, measured by total asset value, and have lower debt levels. This supports the suggestion that hot IPOs take advantage of “windows of opportunity” and go public although they are not necessarily under financial constraints and so could increase their debt. The p-values indicate that the differences are significant. Profitability values also show distinctions between the different IPO cycles. The Wilcoxon-Mann-Whitney test confirms that in cold months, firms show higher operating income as well as higher net sales. However, looking at the mean of both variables, the reverse conclusion can be drawn, although this difference is not confirmed by the t-test. This indicates that the standard deviation and diversity of both variables is much higher in the hot sample. “Capital expenditure” only shows significant differences in firms’ spend18

The values are not exactly the same as in chapter II: table III (panel C) because of different IPO samples. However, the estimated results for significant differences in hot and cold markets are corresponding.

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

110

ing on assets when the equality of mean is tested (hot: € 65.218 mio., cold: € 13.634 mio.). In contradiction to this, the median values and differences are found to be much smaller: Hot; € 1.560 mio., cold; € 1.735 mio. However, the firms in the hot period seem to have greater future growth potential, because their asset volume is significantly lower than that found in the cold sample. Furthermore, the performance ratios (ROE and EPS), which are also included in the regression model, show no significant differences in the mean and median values. Neither are intangible assets, nor total book value of equity, distinct in hot and cold market periods. Chapter III: Table II

Firm Characteristics The accounting variables are from the IPO firms' financial reporting closest to the IPO date. Total assets, intangible assets, total equity, operating income, net sales and capital expenditure are denoted in mio. €. Leverage is calculated as total debt divided by the sum of total debt and total equity. ROE is the return (net income) divided by book value of common equity. EPS stands for basic earnings per share excluding extraordinary items. "Hot"("cold") defines IPOs occurring in month with higher (lower) total number of IPOs than the median value. The pvalue denotes the probability to reject the null-hypothesis of the t-test of equality of means and the Wilcoxon-Mann-Whitney test, if the hot and cold samples come from the same distribution.

Assets Intangible Assets Equity Leverage Operating Income Net Sales Capital Expenditure ROE EPS

 

Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median

 

Total

Hot

Cold

P-Value

788.775 36.54 21.668 0.360 193.805 18.250 0.253 0.147 41.413 3.465 312.816 22.645 37.734 1.625 36.790 18.215 2.402 0.390

1445.270 31.605 28.154 0.370 317.007 15.005 0.227 0.120 65.284 2.855 461.630 18.295 65.218 1.560 30.184 19.930 3.732 0.330

212.341 43.235 15.950 0.320 85.620 19.420 0.276 0.175 20.453 4.130 182.150 30.960 13.634 1.735 48.181 15.780 1.234 0.455

0.113 0.010 0.450 0.823 0.154 0.156 0.056 0.074 0.123 0.061 0.162 0.006 0.089 0.854 0.238 0.308 0.231 0.112

 

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

111

The transaction characteristics of the sample taken from both market phases do not allow a clearer conclusion to be drawn (see table III). Only the preliminary price shows a higher mean and median value (with a Wilcoxon-Mann-Whitney test pvalue below 10%), but neither the offer prices nor the price after the first trading day are greatly different between the two periods. Interestingly, the increase in mean values from the preliminary price range to the “last” price is higher in the cold period (€ 11.62) than in the hot (€ 9.53). The number of shares offered, and therefore the value of proceeds, differs significantly; these variables show much higher means and medians in months of high IPO volume. The market value, calculated as the total number of shares multiplied by the offer price, shows a hot period median of € 136.00 mio., in contrast to € 97.48 mio. with a Wilcoxon-MannWhitney test p-value of 0.001. Surprisingly, the mean is much higher in the cold issue phase (p-value: 0.103), which indicates that few very large companies went public in these months, which increased the average market valuation in the cold issue phase. Chapter III: Table III  

Offer Characteristics Preliminary price is the midpoint of the bookbuilding range. The last price is the share price after the first trading day. Proceeds are calculated as the number of offered shares (including shares from an overallotment option) multiplied with the offer price. Market value is calculated as total number of shares outstanding after the IPO multiplied with the offer price. Number of shares, proceeds and market value are denoted in million. "Hot"("cold") defines IPOs occurring in month with high (lower) total number of IPOs than the median value. The p-value denotes the probability to reject the null-hypothesis of the t-test of equality of means and the WilcoxonMann-Whitney test, if the hot and cold samples come from the same distribution.

Preliminary Price Offer Price Last Price Number of Shares Proceeds Market Value

Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median

Total

Hot

Cold

P-Value

22.014 19.000 23.837 19.500 32.602 22.000 5.344 2.050 123.623 39.092 7,560.000 124.000

22.221 19.500 22.890 19.725 31.764 23.500 5.807 2.219 134.400 42.000 556.589 136.000

21.813 17.872 24.774 19.000 33.437 20.100 4.885 1.897 113.000 32.719 14,372.51 97.48

0.807 0.064 0.286 0.781 0.532 0.150 0.502 0.019 0.532 0.010 0.103 0.001

112

DO “HERDING” EFFECTS ON FIRM MULTIPLES DETERMINE IPO VALUATION?

A clearer distinction can be made by looking at firm and industry multiples as well as market characteristics. Panel A of table IV shows the mean and median marketto-book ratios and price-earnings ratios in the month prior to each sample IPO. The MB ratio of all tradable firms in the related industry (MB Ratio (Market)) shows much higher medians in the hot market. Also during these months, IPOs seem to follow previous public offerings which were valued higher than in the cold sample (MB Ratio (Prev. IPO)). Furthermore, the IPOs’ median MB valuation is significantly higher in the hot than in the cold market: 8.232 and 4.185 with a p-value below 0.1%. However, the book value of equity does not differ significantly, as shown in table II. The same conclusion can be drawn in an analysis of the median PE ratios (probabilities of the t-test are lower than 10%). The IPOs have significantly higher price-earnings ratios in hot months, while the earnings per share of these firms vary around a mean of 2.402 (median 0.390). Firms going public in this market cycle also follow much higher PE ratios of previous IPOs and industry averages, which is indicated by the median “PE Ratio (Prev. IPO)” and “PE Ratio (Market)”. The market conditions represented by volatility and by 3-month percentage price change of the traded stocks in the IPO firm’s industry also confirm conclusions drawn in previous literature. IPOs in a month of large numbers of equity issues follow months with a mean share price volatility of 0.476 (median: 0.490) and 3-month percentage price change of 15.082% (median: 6.213%) in the related industry segments, compared to a volatility mean of 0.397 (median: 0.375) and an average positive price increase of 3.450% (median: 2.432%) in a cold month. These differences are significant, according to equality tests. This is also confirmed by data from panel B of table IV. Here, the values are calculated per month, and not for every sample IPO. In more detail: for each hot/cold month the mean/median of MB/PE ratios and the volatility and percentage price changes of all tradable shares in the German stock market are calculated. However, the monthly market-to-book and price-earnings ratios are smaller in a hot month. This suggests that although many firms follow high market valuation of previous public offerings and comparable industry firms, the valuation in the month of the IPO date is not necessarily higher across the market.

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

113

Chapter III: Table IV

Macroeconomic Conditions Panel A: "MB Ratio (IPO)" is the market-to-book ratio of the sample IPOs. "MB Ratio (Market)" is the average market-to-book ratio one month before each sample IPO. "MB Ratio (Prev. IPOs)" is the average market-to-book ratio of IPOs in the month before each sample IPO. Market-tobook ratio is defined as market value of the ordinary (common) equity divided by the balance sheet value of the ordinary (common) equity in the company. The "PE Ration (IPO)" is the priceearnings ratio of the sample IPOs. "PE Ration (Market)" is the average price-earnings ratio one month before each sample IPO. “PE Ratio (Prev. IPOs)” is the average price-earnings ratio of IPOs in the month before each sample IPO. "Volatility" is the 3-month moving firms’ volatility one month before each sample IPO. The "% Price Change" is the percentage share price change over 3 month before each sample IPO. The values are calculated as an average of all tradable shares in Germany related to the industry sector of each IPO. Panel B: The average values per month of all tradable shares across all industries are calculated. "Hot"("cold") defines IPOs occurring in month with higher (lower) total number of IPOs than the median value. The p-value denotes the probability to reject the null-hypothesis of the t-test of equality of means and the Wilcoxon-Mann-Whitney test, if the hot and cold samples come from the same distribution. Panel A MB Ratio (IPO) MB Ratio (Market) MB Ratio (Prev. IPOs) PE Ratio (IPO) PE Ratio (Market) PE Ratio (Prev. IPOs) Volatility % Price Change

Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median Mean Median

Panel B MB Ratio PE Ratio Volatility % Price Change

Mean Median Mean Median Mean Median Mean Median

Total

Hot

Cold

P-Value

60.346 5.016 16.614 5.301 84.040 8.445 157.907 42.989 104.956 31.400 163.056 86.378 0.436 0.424 9.024 4.132

58.738 8.232 7.357 7.912 165.957 11.864 174.112 53.959 55.443 42.489 163.935 96.313 0.476 0.490 15.082 6.213

61.741 4.185 25.676 3.944 13.825 7.059 144.257 34.807 151.848 25.157 162.303 44.951 0.397 0.375 3.450 2.432

0.925 0.000 0.136 0.000 0.000 0.000 0.389 0.001 0.208 0.000 0.930 0.000 0.000 0.000 0.000 0.000

Total

Hot

Cold

P-Value

10.548 8.351 170.050 44.935 0.239 0.202 5.640 4.581

8.267 7.857 66.026 37.710 0.271 0.224 12.038 7.444

10.862 8.386 184.399 46.635 0.234 0.198 4.757 3.903

0.306 0.740 0.192 0.300 0.019 0.788 0.009 0.060

114

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

IV.2 IPO’s and Comparable Firm Multiples The descriptive empirical analysis confirms that the considered multiples of the IPOs differ between market phases. However, the valuation of publicly traded firms does not show the expected low probabilities of the tests of equality. The months of high IPO volume do not show considerably higher MB and PE ratios in shares of firms traded on the German stock exchanges. As this paper is concerned with whether or not IPO valuations can be explained by multiples, the relationships between firms’ and market’s MB and PE ratios need to be investigated in more detail. Figure I shows the average monthly market-to-book ratio of all IPOs and all publicly traded shares in Germany. The IPO MB ratio is much more volatile than the total stock average. The new market period and the dot-com bubble, in particular, show very high market values for initial public offerings. Due to only 6 IPOs having been completed in the two following years (2002-2003), no clear conclusion about valuation changes in these years can be made. After 2006, however, increasing values can be confirmed, with the highest MB ratio in May 2006. The market average indicated by the dotted line shows higher values only between 1998 and 1999, before the increase in IPO valuations. After 1999 the ratios for market valuation of equity remained between approximately 5 and 9. As a control for industry-wide effects, two sub-samples are here examined more closely: the technology and industry sectors. These segments are the largest groups in the sample, according to ICB classification, with 204 technology firms and 94 IPOs related to the sector of industry goods and services. Figure II shows the movement of IPO’s MB ratios and already publicly traded firms in the technology segment. High valuation and volatility in the IPO ratios occurred in the first part of the relevant decade, where the internet and technology firms dominated the IPO market. The average valuation since 2006 has been, by comparison, much lower (except in June 2007). Also, the MB ratios of the technology firms already listed are more stable, varying between a lowest value of 3.6 in January 1997 and a highest value of 18.42 in November 2000. The same conclusion can be drawn from the sample of industry firms (figure III). Interestingly, the average market valuation was negative in the beginning of 1998, and then increased to a MB ratio of 12.34 in February 1999, while the IPOs had its highest market valuation in relation to book value of equity in May 1999.

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

Chapter III: Figure I

MB Ratios of IPOs 250

200

150

100

50

MB IPOs

Jan 07

Jan 06

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

Jan 97

0

MB Market

Chapter III: Figure II

MB Ratios of Technology IPOs 350 300 250 200 150 100 50

MB Market

MB IPO

Jan 07

Jan 06

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

Jan 97

0

115

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

116

Chapter III: Figure III

MB Ratios of Industry IPOs 250

200

150

100

50

MB IPO

Jan 07

Jan 06

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

-50

Jan 97

0

MB Market

In Figure IV the monthly price-earnings ratio of the complete stock sample is compared with the sample firms. No simultaneous movement of the IPOs’ PE ratios or of the market average are seen in this graph. Valuation and changes in the valuation of firms going public are higher in the years between 1997 and 2001. The industry-wide average of all traded firms also shows more volatility compared to the market-to-book ratios, and achieves very high values in January 1999 and October 2001. Interestingly, the two months with the highest ratios of the entire stock market seem to follow high IPO PE ratios. A possible explanation could be that the multiple ratios of new public firms are included in the market average in the following months, which increases the comparable values. After 2004, however, the PE values of IPOs greatly exceed the relatively low PE ratios of all traded firms across the stock market. Figure V shows this relationship for the technology segment. Here the PE ratio of the IPOs corresponds with figure IV. Only price-earnings are lower across all technology stocks, and without exceptionally high peaks in a few months. Between October 2003 and September 2004, higher PE ratios were achieved, with a maximum of 913.91. Unfortunately, during this time only one IPO took place (PE ratio: 100), so the effect on firms going public can not be identified. Firms’ multiples in the industry goods and service sector also increased (to a maximum of 264.52 in December 2002), but one year

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

117

earlier, from November 2002 to November 2003, a period in which no IPO took place in this sector. Overall, figure VI indicates a generally lower level of IPO PE ratios, whereas two high values are obtained in November 2006 and July 2007 driven by a small number of very large public offerings. It can be concluded that the graphs for PE/MB ratios of IPOs and the market do not provide any evidence of their changing in response to each other: Neither can any clear relation between both multiple series be seen across different industry sectors. Chapter III: Figure IV

PE Ratios of IPOs 1200 1000 800 600 400 200

PE IPOs

PE Market

Jan 07

Jan 06

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

Jan 97

0

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

118

Chapter III: Figure V

PE Ratio of Technology IPOs 2500 2000 1500 1000 500

Jan 06

Jan 07 Jan 07

Jan 05

Jan 04

Jan 06

PE Market

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

Jan 97

0

PE IPO

Chapter III: Figure VI

PE Ratio of Industry IPOs 1200 1000 800 600 400 200

PE IPO

PE Market

Jan 05

Jan 04

Jan 03

Jan 02

Jan 01

Jan 00

Jan 99

Jan 98

Jan 97

0

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

119

For this reason, the analysis is expanded here by investigating the crosscorrelation as well as the first-order autocorrelation of these variables. To this end, the 11-year time period of the sample is divided into several sub-periods relating to the number of IPOs (see figure VII). The quartiles of the IPO volume per year are calculated for the complete sample, as well as the technology and industry samples. Years with a number of IPOs falling within the lower quartile are defined as “cold”, while years of IPO volume within the highest quartile are defined as “hot” issue periods.19 The years with a number of issues between the lower and upper quartile are defined as “normal”. Consecutive years with the same classification are grouped together. The years of 1997-1998 are “normal” with an increasing volume, so that 1999-2000 is a period defined as “hot”, with two years of the highest IPO volume in the 11-year sample. 2001 can still be defined as a “normal” period, but with a clearly decreasing number of issues. So that the “cold” period occurred between 2002 and 2004: three years with an IPO volume in the lowest quartile. The three following years are also defined as a “normal” phase. For both, the complete sample and the two sub-samples, the same groups can be recognized over the same time period. Chapter III: Figure VII

IPOs per Year and Industry 160 140 120 100 80 60 40 20 0 1997

1998

1999

2000 Total

19

2001

2002

Technology

2003

2004

2005

2006

2007

Industry

Years with an IPO volume smaller or equal to the lower quartile range are defined as “cold“. Years with an IPO volume larger than the upper quartile range are defined as “hot“.

120

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

First of all the cross-correlation between the MB ratios of the IPOs and the traded shares are estimated for the three samples as shown in figure VIII. Surprisingly, the cross-correlations are negative in the normal (1997-1998) and hot (1999-2000) periods, indicating that higher IPO market-to-book multiples are also more likely to correspond to lower ratios across all traded firms and for the technology and industry sample. With the decreasing IPO volume for the normal year of 2001, the correlation becomes positive (except for industry firms). However, as there are only a few IPOs, with longer time gaps between the offering dates, no clear conclusion can be drawn. The same argument holds for the cold years of 2002-2004. The period between 2005 and 2007, with 131 IPOs, however, also shows a positive cross-correlation between MB ratios of IPOs and the market. Even the normal periods (1997-1998 and 2005-2007) do not indicate the same relation of IPO multiples to market valuation. Chapter III: Figure VIII

Cross-Correlation of MB Ratios of IPOs and Market 0,6 0,5 0,4 0,3 0,2 0,1 0 -0,1 -0,2 -0,3 1997-1998

1999-2000

Total MB-Corr.

2001

2002-2004

Technology MB-Corr.

2005-2007 Industry MB-Corr.

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

121

Chapter III: Figure IX

Cross-Correlation of PE Ratios of IPOs and Market 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 -0,1 -0,2 -0,3 1997-1998

1999-2000

Total PE-Corr.

2001

2002-2004

Technology PE-Corr.

2005-2007 Industry PE-Corr.

Figure IX shows the results for cross-correlation of price-earnings ratios for the sub-periods of the IPO market. Here again, the three samples do not seem to correspond to each other, and show different developments. While the crosscorrelation of the IPO and market ratio is positive in the first normal period (19971998), the industry sample shows a negative relationship. In the hot phase, however, the (small) cross-correlation values become negative for all sub-samples. To draw a conclusion for the next three years is not straightforward, for the reasons outlined above. In the final “normal” phase, only the complete IPO sample shows a positive cross-correlation of PE ratios, while both sub-segments show a small but negative relation between IPO and market valuation. For this figure, no final conclusion can be made, as no clear development across the industry segments and time periods can be recognized. Overall, the correlation values are very small, and no significant interaction (positive or negative) between these variables can be confirmed. The first-order autocorrelation of the IPO multiples is also analyzed, in order to estimate whether the valuation of previous IPOs is more relevant than the market average. In figure X, the continuous line indicates the autocorrelation of MB ratios for the sample including all IPOs. The autocorrelation changes from positive, in 1997-1998, to negative in the three subsequent periods (from hot to cold between 1999 and 2004) and then turns positive again for the last normal market phase.

122

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

The graphs showing the areas of technology and industry, however, show an initial negative autocorrelation in the “normal” phase, which then becomes positive in the hot market. In subsequent years, the value for autocorrelation became negative again, while, in the period between 2005 and 2007, both a positive value for the industry sample and a negative value for technology IPOs can be seen. In “normal” phases, the valuation of an IPO seems to be related to previous public offerings and their multiple values. Looking at the industry segments separately, however, this conclusion can also be drawn for IPOs during the “hot” market of 1999-2000. Chapter III: Figure X

Autocorrelation of MB Ratios of IPOs 0,5 0,4 0,3 0,2 0,1 0 -0,1 -0,2 -0,3 1997-1998 Total MB IPO

1999-2000

2001

2002-2004

Technology MB IPO

2005-2007

Industry MB IPO

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

123

Chapter III: Figure XI

Autocorrelation of PE Ratios of IPOs 0,6 0,4 0,2 0 -0,2 -0,4 -0,6 -0,8 1997-1998

1999-2000

Total PE IPO

2001

2002-2004

Technology PE IPO

2005-2007

Industry PE IPO

Finally, figure XI shows the first-order autocorrelation in terms of IPOs’ priceearnings ratios. Here, the estimates are positive in the normal phase of IPO volume, becoming negative for the period between 1999 and 2000 across the three sub-samples. This indicates that, especially in a month with high numbers of IPOs, the PE multiples of previous IPOs become less relevant in an IPO’s market valuation. The reverse relationship, however, is found in the last normal phase of the market, and corresponds to the first sub-period. Overall, the values for first-order autocorrelation are relatively small and do not confirm that there is any dependence over the month. In the appendix of chapter III, figures I, II, III show the cross- and autocorrelation of the IPOs according to the three sub-samples discussed. However, comparison across the different industry classifications also does not show any clear development of the variables across the market phases. The estimates above also do not show that there is any clear relationship between the market and IPO valuation. The movement and correlation of PE and MB ratios do not correspond to each other. Therefore, the first important question of this paper cannot be answered positively: changes in IPO valuation cannot be explained by the overall market valuation of tradable stocks. This result is surprising, as practitioners and existing literature would suggest (albeit without confirmation) that high IPO market values accompany high investor perception, and therefore high overall market value of stocks.

124

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

IV.3 IPO Valuation IV.3.1 Regression Estimates on IPO Valuation The regression models discussed in III.2 aim to give more insight into estimating the determinants of IPO valuation and the effects of the market environment. As the analysis above shows rather puzzling results, the coefficients and significance of the multiples in the context of firms’ accounting information and market environment are especially interesting. The first OLS-regression estimates MV [1]-[3] in table V show the first model (1), only in terms of firms’ financial and profitability characteristics regressed on the total market value (calculated with preliminary, offer and last prices). The R-squares show the greatest degree of fit to the regression model on preliminary market valuation (R²= 70.3%), and the least percentage, for the market value calculated with the last trading price at the offer day (R²= 65.7%). The F-statistics, indicating the ratio of the explained variability from the regression model (R²) and the unexplained variability, show that the model is useful and the variables (or at least one) have an association with market valuation. The values of the F-statistics also decrease with the different offer prices. Similar statistical characteristics can also be seen in the following regression models.

R² F-Statistic

Intercept

Perform

Vola

MB

PE

EPS

ROE

Capex

Leverage

Equity

Intanratio

Assets

16.315 (86.833)*** 0.703 78.111

0.677 (6.009)*** -0.001 (-1.432) -0.064 (-0.631) -0.569 (-2.096)** 0.059 -1.492 0.001 -1.548 0.013 (7.107)***

16.328 (84.145)*** 0.694 75.043

0.667 (6.016)*** -0.001 (-1.472) -0.056 (-0.561) -0.577 (-2.084)** 0.065 -1.588 0.001 -1.474 0.013 (7.376)***

16.452 (82.751)*** 0.657 63.139

0.748 (8.850)*** -0.001 (-1.433) -0.136 (-1.792)* -0.816 (-2.813)*** 0.053 -1.281 0.001 -1.196 0.013 (7.462)***

MV (Preliminary) [1] MV(Offer) [2] MV(Last) [3]

16.352 (84.313)*** 0.730 69.103

0.674 (5.891)*** -0.001 (-1.475)* -0.068 (-0.656) -0.557 (-2.038)** 0.051 -1.316 0.001 -1.480 0.012 (6.902)*** 0.001 (22.419)*** -0.001 (-3.525)***

MV(Preliminary) [4]

16.352 (84.313)*** 0.732 66.431

0.664 (5.894)*** -0.005 (-1.519) -0.060 (-0.593) -0.566 (-2.036)** 0.058 -1.425 0.001 -1.404 0.013 (7.166)*** 0.001 (22.832)*** -0.001 (-4.146)***

16479 (82.164)*** 0.659 49.305

0.747 (8.831)*** -0.001 (-1.426) -0.141 (-1.846)* -0.829 (-2.857)*** 0.055 -1.305 0.001 -1.165 0.013 (7.385)*** 0.001 (3.714)*** -0.001 (-4.726)***

MV(Offer) [5] MV(Last) [6] 0.626 (5.766)*** -0.001 (-2.723)*** -0.028 (-0.293) -0.427 (-1.565)* 0.045 -1.233 0.001 -1.435 0.006 (3.103)*** 0.001 (23.318)*** -0.001 (-2.327)** 0.092 (0.228) 0.018 (6.443)*** 16.260 (57.975)*** 0.769 69.085

MV(Preliminary) [7] 0.611 (5.781)*** -0.001 (-2.935)*** -0.018 (-0.190) -0.434 (-1.575) 0.052 -1.371 0.001 -1.345 0.006 (3.157)*** 0.001 (23.738)*** -0.001 (-2.691)*** 0.003 (0.080) 0.019 (6.901)*** 16.304 (56.988)*** 0.766 67.637

0.686 (8.388)*** -0.001 (-3.371)*** -0.091 (-1.272) -0.666 (-2.413)*** 0.047 -1.182 0.001 -1.068 0.005 (2.380)*** 0.000 (60.19)*** -0.001 (-2.996)*** 0.111 (0.217) 0.022 (6.873)*** 16.388 (50.011)*** 0.718 52.628

MV(Offer) [8] MV(Last) [9]

For definition of variables look at table I (chapterIII). The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance.

Regression Models on Market Value

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

125

Chapter III: Table V

126

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

In equation (1) the coefficients for the variables of assets, leverage and the EPS of the firm are highly significant, to the 1% level, in explaining IPO valuation, calculated with the number of shares and the three different prices during the offering process. The estimates show that larger firms, in terms of assets, are also valued more highly, a 1% increase in assets resulting in a 0.677% higher market valuation. Higher profitability in terms of earnings per share is also positively correlated with the dependent variable. However, the level of debt and the associated higher bankruptcy risk of the firm on offer is interpreted negatively by IPO participants. The additional variables for intangible assets and book value of equity show negative signs, although these are not significant. The ratio of intangibles seems to reduce the valuation of a firm rather than indicating innovation. Only the result for the equity measure is surprising: the negative relationship indicates that a 1% increase in book value of equity results in a 0.056% reduction in market value calculated with the offer price. Presumably, smaller firms have been the most overvalued. Profitability measures, however, show the expected positive signs: future growth expectations and return on invested capital are valued positively by underwriters and investors. The second regression model in equations MV [4]-[6] (see table V) includes the following additional two multiples: market-to-book and price-earnings ratios of all publicly traded shares in the same industry segment as the sample IPOs. The variables increase the degree of model fit in terms of the R², and are significant at the 1% level. The coefficient of the MB ratio shows a negative indication. A positive relation of IPOs’ market values and industry-related firms had been expected: However, the results from chapter IV.2, or figure VIII, can be confirmed, indicating a negative cross-correlation between these comparable multiples in the first four years of the sample period. The economic magnitude of the MB ratio to IPO valuation is very low, as the coefficient is rounded up to -0.001. A similar minor effect is seen in the comparable PE multiple of previously traded shares, although the coefficient indication is positive. This also supports some of the previous section’s findings in figure IX, where the cross-correlation of IPO’s and market’s PE ratios is positive in the “normal” periods of issue volume in the periods of 19971998 and 2005-2007. The final regression model (3) on market value includes the variables allowing for market effects (see [7]-[9] in table V). The independent variable for volatility (“vola”) is positively related to the IPO valuation as suggested by Pástor/Veronesi (2005:1720), although not a significant factor. Additionally, the “perform” variable shows a positive indication, and is also significant at the 1% level. The measure of the 3-month percentage price change of related industry stocks indicates

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

127

that valuation increases in line with market performance in the months prior to the IPO; also as suggested by Pastor/Veronesi (2005:1720). The coefficient is also much higher than for the PE and MB variables. This confirms that the market return is a more important IPO value driver than the comparable multiples. Interestingly, “intanratio” also becomes significant (at 1% level) in this regression equation, while the interpretation or meaning of “leverage” changes slightly in [7] and [8]. With the exception of this variable, issuers, underwriters and investors can not be said to value or interpret the firms’ and markets’ characteristics differently, because the models’ estimates do not change with the three prices in the IPO process. IV.3.2 IPO Valuation in Hot and Cold Markets The regressions above show that the variables discussed reasonably affect IPO valuation by underwriters and investors. While the explanatory power of comparable multiples seems to be small in comparison to firm’s financial characteristics and market conditions, the PE and MB ratios may explain market valuation to a greater or lesser degree according to market phase. The idea of information spillover would suggest that in phases with high IPO volume firms are valued according to higher market multiples, and firms’ characteristics become less relevant. The reduced information asymmetry between issuers and investors due to the same valuation factors and better positive response to issue announcements would also explain the higher number of firms going public. In order to analyze this hypothesis, the sample is split into two sub-samples, “hot” and “cold”, classified by the median numbers of IPOs per month (see section IV.1 of chapter III). The regression models are applied to the two samples; the changes in coefficients should then indicate which of the valuation variables becomes more or less important.

R² F-Statistic

Intercept

Perform

Vola

MB

PE

EPS

ROE

Capex

Leverage

Equity

Intanratio

Assets

16.435 (66.351)*** 0.781 67.037

0.630 (4.131)*** 0.001 (1.679)* -0.060 (-0.477) -0.785 (-2.3060)** 0.125 (2.381)*** 0.001 -1.459 0.037 -1.072

16.462 (65.490)*** 0.776 65.184

0.613 (4.149)*** 0.001 (0.170)* -0.047 (-0.388) -0.789 (-2.281)* 0.135 (2.521)*** 0.001 -1.365 0.037 -1.082

16.481 (65.606)*** 0.737 52.616

0.764 (6.238)*** 0.001 (0.851) -0.184 (-1.857)* -1.046 (-2.959)*** 0.107 (1.947)** 0.001 (0.957) 0.037 0.903)

16.439 (66.338)*** 0.816 63.881

0.633 (4.042)*** 0.000 -1.433 -0.064 (-0.494) -0.752 (-2.207)** 0.113 (2.246)** 0.001 -1.375 0.033 -1.001 0.001 (25.497)*** -0.001 (-3.903)***

16.471 (65.574)*** 0.812 62.186

0.615 (4.062)*** 0.001 -1.431 -0.052 (-0.417) -0.760 (-2.203)** 0.123 (2.413)*** 0.001 -1.278 0.033 (0.983) 0.001 (25.578)*** -0.001 (-4.594)***

16.523 (65.079)*** 0.741 41.177

0.764 (6.199)*** 0.001 (0.767) -0.191 (-1.915)** -1.074 (-3.026)*** 0.112 (2.001)** 0.001 (0.920) 0.032 (0.795) 0.000 (2.272)** -0.001 (-4.938)***

0.609 (3.820)*** 0.001 -1.487 -0.057 (-0.443) -0.704 (-2.024)** 0.118 (2.313)** 0.001 -1.390 0.046 -1.282 0.001 (16.936)*** -0.001 (-2.611)*** 0.155 (0.251) 0.008 -1.130 16.395 -44.316 0.818 52.036

0.589 (3.832)*** 0.001 -1.499 -0.044 (-0.356) -0.702 (-1.991)** 0.129 (2.500)*** 0.001 -1.298 0.049 -1.333 0.001 (16.801)*** -0.001 (-3.084)*** 0.230 (0.354) 0.009 -1.287 16.399 (43.080*** 0.814 50.841 0.739 (5.611)*** 0.001 (0.800) -0.180 (-1.795)* -0.981 (-2.734)*** 0.121 (2.180)** 0.001 (0.347) 0.055 -1.188 0.000 (1.812)* -0.001 (-2.824)*** 0.547 (0.611) 0.013 -1.374 13.327 (34.186)*** 0.766 34.031

For definition of variables look at table I chapter III. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. MV(PreMV(PreMV(Preliminary) [10] MV(Offer) [11] MV(Last) [12] liminary) [16] MV(Offer) [17] MV(Last) [18] liminary) [13] MV(Offer) [14] MV(Last) [15]

Regression Models on Market Value: Cold Market

128 “HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

Chapter III: Table VI

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

129

In table VI, the OLS-regression estimates for the cold sample are reported. The first model (1) in regressions MV [10]-[12] shows similar results to the complete sample findings. Instead of earnings-per-share, the profitability variable “capex” is significant at the 1% level, and positively affects the market value of IPOs. The indicator for intangible assets in relation to total assets also moves from negative to positive. In months with low IPO volume, the value placed on innovation outweighs insecurity about the valuation of assets in place. The results for the second regression model MV [13]-[15] are similar to the estimates from regression MV [4]-[6]. The variables for assets, leverage and capital expenditure are significant at the 1% and 5% levels, and coefficients show the expected signs. Again, the multiple values of the market are significant in explaining IPO valuation, where “MB” shows a negative and “PE” a positive correlation. For both variables, however, the economic magnitude is very low, because the coefficients show values slightly below +/- 0.001. The results for the complete regression equation are more interesting. In contrast to the results in table V, the variables “perform” and “vola” have no significant influence on IPO valuation. Particularly, the proxy for previous market performance shows lower coefficients and lower t-statistics. The comparison of market conditions also confirms that previous market price changes are higher in hot than in cold IPOs. Therefore, this result suggests that market return becomes less relevant in IPO valuation when performance is low. Consideration of the “hot” sample should support these findings. The second sub-sample includes all IPOs completed in a month with more than the median number of initial public offerings. The results are shown in table VII, and similar to the “cold” sample, the intercept of the regression models being highly significant with a coefficient in the region of 16.00. This means that the regression models applied in this chapter cannot explain completely the market valuation of IPOs. As this dissertation is concerned with the effects of specific market characteristics, and as the intercept coefficients (β1) are closely similar, the two subsamples can be compared.20 Compared to the estimates of the complete and the cold IPO samples, the regressions MV [19]-[21] in table VII show changes in the explanatory power of variables. Although the value of a firm’s total assets is significant, the level of debt no longer has a negative effect on IPO valuation. The profitability measures “ROE” and “EPS” show significant positive coefficients and high t-statistics. In contrast 20

Appendix III: Table I shows the results for the regression models with an intercept dummy “cold” on the complete sample. This confirms the argument, and shows significant changes in two variables, whereas almost all other variables change signs. Hot and cold market samples and the considered independent variables change, and this is not due to the absence of variables of the regression equation. The differences are more pronounced in the models on the two sub-samples.

130

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

to the cold regression estimates, the “capex” variables are no longer significant, even changing from positive to negative signs. The second model (2) also includes the market or industry multiples for each IPO (MV [22]-[24]). The same firmspecific accounting measures are related to IPO valuation as shown in model (1). However, the PE ratio is also significant, and affects the IPO valuation positively, although the coefficients are very small. Furthermore, the MB ratio does not show high enough values of t-statistics to indicate significance at the relevant levels. The third regression model on market value is applied in equation MV [25]-[27], and provides this chapter’s central insight into changes in hot and cold market phases. In addition to the positive correlation of assets, the ratio of intangible assets to total assets is highly significant and it occurs with a negative correlation to the dependent variable. In the cold market regression analysis, this variable showed a positive influence in market value (without significance). Leverage and capital expenditure of the firms going public do not effect the valuation in a hot market environment differently than a cold issue phase. Additionally, the other profitability proxies, EPS and ROE, which were significant in the first two regression models (1) and (2) in the hot phases, lose their explanatory power. Compared to the complete and cold sub-samples, the multiple variables PE and MB also do not show any significant relationship to the market value in the regression model (3). However, the proxies for market environment can be expected to indicate more relevance. The volatility measure is not significant: However, the coefficient is much higher in regression MV [25] and MV [26] than in the previous result, and turns negative in MV [27]. This suggests that in periods with high IPO volume, investors in particular are concerned about volatility in the market. Greater uncertainty about market development could result in reduced willingness to buy shares, thereby affecting prices after the first trading day negatively. A more pronounced effect is seen with the variable “perform”, which is significant at the 1% level. Higher market valuation follows a positive price change in traded stocks in the three months prior to the offering. Interestingly, during the cold market months, this variable has no effects on IPO valuation, as seen in table VI. The estimates for [16]-[18], however, show significant effects of MB and PE multiples of industryrelated firms for each IPO, which cannot be confirmed for the hot issue phases ([25]-[27]).

R² F-Statistic

Intercept

Perform

Vola

MB

PE

EPS

ROE

Capex

Leverage

Equity

Intanratio

Assets

16.201 (50.906)*** 0.641 23.566

0.753 (6.727)*** -0.001 (-1.221) -0.130 (-0.882) -0.144 (-0.301) -0.048 (-0.664) 0.006 (3.077)*** 0.010 (3.335)***

16.213 (49.548)*** 0.629 22.298

0.757 (6.427)*** -0.001 (-1.209) -0.134 (-0.900) -0.176 (-0.351) -0.048 (-0.632) 0.006 (3.130)*** 0.011 (3.407)***

16.293 (47.728)*** 0.593 19.162

0.712 (5.630)*** -0.001 (-1.142) -0.072 (-0.472) -0.227 (-0.422) -0.040 (-0.535) 0.009 (3.982)*** 0.013 (3.973)***

15.714 (40.553)*** 0.672 20.506

0.756 (7.186)*** -0.001 (1.653)* -0.116 (-0.836) 0.226 (0.464) -0.086 (-1.129) 0.005 (2.600)*** 0.009 (3.006)*** 0.07 (2.381)*** 0.011 (0.501)

15.716 (39.276)*** 0.660 19.438

0.758 (6.854)*** -0.001 (-1.660)* -0.118 (-0.843) 0.203 (0.398) -0.086 (-1.081) 0.005 (2.644)*** 0.009 (3.084)*** 0.007 (2.446)*** 0.008 (0.375)

15.893 (35.301)*** 0.617 16.137

0.705 (5.855)*** -0.001 (-1.536) -0.049 (-0.331) 0.113 (0.207) -0.072 (-0.909) 0.007 (3.320)*** 0.011 (3.606)*** 0.007 (2.042)*** -0.006 (-0.264)

0.719 (5.249)*** -0.001 (-2.153)** -0.111 (-0.761) -0.206 (-0.428) -0.069 (-1.037) 0.001 (0.086) 0.003 (0.980) 0.001 (0.283) 0.002 (0.145) 0.444 (0.573) 0.020 (3.922)*** 16.029 (35.962)*** 0.741 22.922

0.712 (4.931)*** -0.001 (-2.214)** -0.110 (-0.742) -0.271 (-0542) -0.064 (-0.935) -0.001 (-0.044) 0.003 (0.931) 0.001 (0.243) 0.001 (0.092) 0.293 (0.374) 0.022 (4.316)*** 16.142 (34.764)*** 0.641 23.566

0.639 (4.444)*** 0.001 (-2.170)** -0.035 (-0.239) -0.467 (-0.192) -0.040 (-0.564) 0.001 (-0.453) 0.004 -1.192 -0.004 (-0.106) -0.009 (-0.511) -0.028 (-0.034) 0.026 (4.629)*** 16.578 (31.087)*** 0.717 20.296

For definition of variables look at table I chapter III. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the tstatistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. MV(PreMV(Pre-liminary) MV(PreMV(Offer) liminary) [19] MV(Offer) [20] MV(Last) [21] [22] MV(Offer) [23] MV(Last) [24] liminary) [25] [26] MV(Last) [27]

Regression Models on Market Value: Hot Market

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

Chapter III: Table VII

131

132

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

The results indicate changes in IPO valuation and varying influences of firm and market characteristics. In a cold environment, assets, leverage and capital expenditure of the firm going public are relevant: However, during phases with more IPO volume, the value of total and intangible assets is more important for underwriters and investors in setting offer prices. However, the hypotheses about information spillover in terms of favourable industry multiples are not confirmed. “Herding” on available public information of market’s MB and PE ratios is not seen in phases with a higher number of equity issues. In cold periods, these variables are more important in explaining IPO valuation. During hot markets the previous performance of the stock markets indicates such behavior. In months with high IPO volume, the performance of already traded shares is also high, and regression estimates indicate that this positively affects the market value of newly issued shares. This is not the case for the other sub-sample of IPOs. Issuers, underwriters and investors, then, give more weight to market performance when it is favorable and when more IPOs are taking place. The variable “perform” may include market information on which participants rely in preference to firms’ accounting information and comparable multiples. The price change may become more important than the average level of market valuation. In contrast to previous suggestions, MB and PE multiples can be considered as firm related information, and have only minor influences. “Herding” occurs in connection with aspects related to positive market performance. The results pertaining to the second question with which this study is concerned do not show the expected effects of multiple values, but suggest some form of herding in respect of public information. Overall, the relevance of MB and PE ratios on IPO valuation is not greatly significant. IV.3.3 Information Asymmetries and IPO Valuation Another aspect to be considered is the information asymmetry between issuers and investors. As proposed by Myers/Majluf (1984: 216 et. seqq.), more firms issue equity if agency costs are low. For example, common valuation factors of issuers, underwriters and potential investors reduce information asymmetries and related costs. Therefore, it would be reasonable for periods with high IPO volume to also have more equality of information distribution between participants. Indicators of this IPO environment are, for example, the initial returns after the first trading day. As there are several theories regarding explanations of IPO underpricing, it can be argued that increased information asymmetries between issuers, underwriters and investors require higher initial returns as a form of compensation for investors to participate in the offering (e.g. Rock (1984:187 et. seqq.), Beatty/Ritter (1986:213 et seqq.)). However, periods of higher IPO underpricing coincide, or are often followed by periods of high IPO volume (e.g. Lowry (2003), Lowry/Schwert (2002), Ibbotson/Jaffe (1975)). Also in this sample, hot periods in terms of IPO volume

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

133

have a median value of initial returns of 7.94%, compared to 3.7% in cold issue periods, a significant difference according to the Wilcoxon-Mann-Whitney test. The total sample mean is 33.91% (median: 5.29%).21 This does not necessarily indicate that high IPO volume periods are the result of reduced agency costs for the issuer of equity. In order to gain more insight into the distinct valuations of IPOs in relation to information asymmetries between participants, the regression models in periods with high and low underpricing should be examined. The complete IPO sample is once more split into two sub-samples by initial returns after the firm’s shares have been listed for the first time. IPOs with underpricing lower than the median value of 5.29% are grouped together as “UP: Low” and those with initial returns above the median are grouped as “UP: High”. The estimates for regression model (3) on market value calculated with the offering price are shown in table VIII. Interestingly, the results are similar to hot and cold sub-samples MV [17] and MV [26]. The valuation equation MV [28] for the sample of low initial returns shows a significant positive coefficient for the variables “assets” and “capex”. Larger (or less risky) firms with higher growth perspectives have a higher market valuation. However, firms with relatively lower book values of equity receive a higher offering price. Debt levels also negatively affect the IPO valuation, significant at the 1% level. In the estimates for IPOs in cold months, the same variables (other than “equity”) showed significant explanatory power for the relevant variable. Additionally, the multiple variables of market-to-book and price-earnings-ratio are highly significant and show a negative/positive indication of the coefficients. This is also consistent with the results for firms in cold issue periods, in which these variables also affect IPO valuation. Additionally, the average 3-month performance of industry related stocks determines the market value in regression MV [28], which was not found in previous estimates. These results confirm, however, that the valuation of IPOs with lower initial returns is similar to that of cold issue firms. The intercept of the regression models is also very close. Accounting information and multiple ratios explain IPO market value when the information asymmetry between investors and issuer (underwriter) is assumed to be low. The assumption that in cold periods the firm is valued according to financial statements, so that the informational differences between participants is low and underpricing is reduced, is therefore reasonable. Before a final conclusion is reached, however, the estimates for the sub-sample of IPOs with higher underpricing should be discussed. Regression estimates MV [29] show significant coefficients for the variables “assets” and “intanratio”. Higher levels of firms’ assets in place increase the market valuation of IPOs, while an in21

These values are calculated with the sample of IPOs of chapter III.

134

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

creased ratio of intangible to total assets seems to increase uncertainty, and therefore has a negative effect on the dependent variable. Another significant explanatory proxy is “perform”, indicating the 3-month average stocks price change before the IPO. The economic impact of stock market performance is almost twice as high as it is in low initial return IPOs in MV [28]. These estimates correspond to the findings of regression MV [26] considering the hot market IPOs, where the same variables were found to be significant with the same sign of coefficients. The valuation of IPOs with considerably higher underpricing, and presumably higher information asymmetries between participants, is the same as in hot issue phases. This suggests that in IPOs where valuation is driven by market return, rather than by comparable multiples or firm characteristics, the information asymmetry between issuers and investors is also higher. While in this chapter it has previously been argued that IPOs seem to be valued by the common factor of market performance, this factor does not influence the information gap. When valuation is determined by the market, initial returns are also higher for investors in these IPOs. In the case of IPO valuation by profitability and industry-related firms, the compensation of investors by allocation of underpriced shares is low. This presumably reflects the reduced uncertainty for an investor bidding for newly issued shares which seem to be priced appropriately in a market context. The results do not necessarily confirm a close relationship between reduced information asymmetries and a higher number of IPOs. Lowry (2003: 6) summarizes in her investigation that adverse selection costs are only marginally significant in explaining IPO volume fluctuation, rather the firm’s demand for capital and changes in the investor optimism determining the decision to go public.

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

135

Chapter III: Table VIII

Regression Models on Market Value: Controlling for Asymmetric Information Regression [28]: Sample of all IPOs which have lower initial returns than the median value. Regression [29]: Sample of all IPOs which have higher initial returns than the median value. Initial return is calculated: last share price after the first trading day divided by the offer price minus one. For definition of variables look at table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance. UP: Low MV(Last) [28]

UP: High MV(Last) [29]

0.787

0.490

R²

(7.417)*** -0.000 (-0.086) -0.263 (-3.146)*** -1.017 (-3.493)*** 0.077 (1.529)** 0.000 (0.403) 0.043 (1.253) 0.001 (6.142)*** -0.001 (-2.648)*** -0.251 (-0.401) 0.014 (2.836)*** 16.612 (40.634)*** 0.776

(2.310)** -0.001 (-3.024)*** 0.154 (1.092) 0.118 (0.185) -0.014 (-0.218) 0.002 (1.603) 0.009 (3.127) -0.003 (-0.653) -0.005 (-1.143) 0.638 (0.866) 0.026 (3.432)*** 16.207 (27.195)*** 0.701

F-Statistic

38.912

19.621

Assets Intanratio Equity Leverage Capex ROE EPS PE MB Vola Perform Intercept

136

V

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

Conclusion

This chapter considers IPO valuation with comparable firm multiples of marketto-book and price-earnings ratios. Changes in the explanatory variables are analyzed according to IPO market phases with high and low equity issue volumes. With the sample seen of German IPOs between 1997 and 2007, no significant changes in firm and transaction characteristics can be confirmed in “hot” and “cold” market periods. However, the market environment for firms going public in months with high IPO volume is much more favorable. These IPOs do not only have higher MB and PE values, but they also follow months of higher multiple ratios in previous equity issues and industry related public firms. Previous volatility as well as average percentage price change of the stock market is also higher than that seen in the IPOs in cold markets. However, this paper’s first important research question cannot be answered as expected. The IPO multiple ratios do not correspond to the overall market valuation of publicly traded firms. Higher monthly MB and PE ratios in newly issued stocks are not correlated to the monthly multiple values of industry related firms. Some sub-periods of the investigated period show a positive and others a negative, cross-correlation of these values. Across industry segments another, different, development is found. Furthermore, the autocorrelations of the monthly IPO multiple ratios do not confirm the theory that previous IPOs provide more important information for firm valuation then previously traded shares. The second aim of this chapter, relating to models of IPO waves and information spillover effect, also shows interesting results. The regression estimates of the cold sample, defined by the number of IPOs in a month, confirm that financial information, profitability measures and multiple ratios are all relevant in market valuation of firms going public. Only the MB and PE ratios seem to have minor economic effects. The market-to-book ratio of previously traded firms in the same industry also has a negative sign of the coefficient. The hot sample regressions, however, prove that in addition to the firm’s size, the valuation is also driven by the 3month average percentage price change before the IPO. The analysis indicates that hot IPOs follow months with high market performance, but that this also significantly affects valuation. The “herding” effects on multiple values in periods with high IPO numbers cannot be confirmed, as these variables have no significant influence in explaining the dependent variable. Underwriters and investors are more likely to show “herding” behavior in relation to information included in the stock price changes. Additionally, the expected lower information asymmetries with common valuation factors such as MB and PE ratios are not found. When initial returns after the first trading day are used as an indicator of the information gap

“HERDING” EFFECTS IN FIRM MULTIPLES AND IPO VALUATION

137

between issuers/underwriters and investors, months with high IPO volumes also show higher asymmetries. Interestingly, market valuation in IPOs with high/low initial return is very similar to valuation in hot/cold market phases. This means that firms do not necessarily choose to go public because of reduced agency costs in equity issues. The results of this chapter do not support several assumptions made in previous research papers and theoretical models, but still form a major contribution to the existing literature about the valuation of publicly traded firms and the decision to go public.

CONSLUSION

139

Conclusion Developing a private firm into a public one is a widely discussed process in corporate finance literature. The decision and the process of raising external capital from a public equity market is very complex and is probably a controversial issue in the lifecycle of the firm, as several parties with divergent interests and intentions are involved. The initial public offering requires the issuing firm’s owners, potential investors and the investment bank as a financial intermediary, and all participants try to profit in the short and the long term. These differences of interest or of status of information offer many different theoretical explanations for the puzzle of underpricing in initial public offerings. Empirical investigations find varying results and partly support or negate these early hypotheses, leading to another strand of literature and explanations developing over time. For these reasons, the literature overview in chapter I can be classified into research areas based on asymmetric information distribution as well as explanations with the assumption of symmetric information of issuers, investors and underwriters. In summary, many arguments and concepts are observed which affect offer prices and the amount of money left on the table in a plausible way. Some explanations cannot be discussed individually, and several determining factors have to be included in the analysis. Some institutional mechanisms can reduce underpricing, while others are likely to affect initial returns in a positive way and so overall no clear distinction of the effects and resulting levels of underpricing can be made, which makes the country comparison between Germany and USA very difficult. Furthermore, it becomes obvious that different market cycles also have to be considered in the decision to go public. Chapter II empirically investigates the preIPO ownership structure of a firm over hot and cold market periods. Several ideas related to informational asymmetries and monitoring activities are summarized within the hypothesis that bargaining interest regarding previous shareholders’ offer prices are likely to change within the market environment. It is shown that financial investors especially, such as venture capitalists or private equity partners, allow higher initial returns on the first trading day in a positive market environment. Previous research results for the German Neuer Markt are confirmed, which have also proven that increased numbers of financial investors bring increased agency conflicts and thereby increase the percentage price increase of new stocks. However, the results also support the theory of higher monitoring incentives on the part of pre-IPO shareholders with higher stakes in the firm. Higher clustering of shareholder groups in German firms compared to that seen in the USA does not seem to affect the findings for underpricing of IPOs necessarily.

140

CONSLUSION

Excluding potential agency conflicts between owners and other participants, IPO offer prices should reflect the firm’s value, which is discussed in chapter III. One common technique among underwriters is comparison of accounting information and multiple ratios, such as price-earnings ratios or market-to-book ratios, of IPOs and already publicly traded firms. The research hypothesis of chapter III should indicate whether the developments of IPO valuation can be explained by industry average multiples. Surprisingly, the valuation multiples of the stock market are unreliable indicators of market values of IPOs. The underwriters seem to increase valuations in response to positive stock price changes, rather than focusing on multiple values of firms within the same industry segment. Furthermore, accounting and profitability measures become less important in valuing the IPO during hot periods and where stock market performance is positive. The results show the impact and developments of IPO volume, firm valuation and underpricing in the German stock market. A larger sample size from a more active stock market could also help to support the results and prove further findings. Furthermore, it could be interesting to reassess the results from a more stable environment or from several hot and cold market cycles over time. However, the German market offers no further opportunities to extend the study in this direction, because IPO activity has slowed down, with only two new issues in the Regulated Market segment in 2008. The requirements for improving and ensuring firms’ accessibility to private and public financing are obvious, and the German stock market is expected to be restructured in this direction in the future. However, institutional changes are not expected to be able to solve the underpricing puzzle. This dissertation shows that the issuing firm is not alone in deciding about the amount of money left on the table. Investors indicate their demand for newly issued shares as well as influencing the overall market sentiment and perception of publicly traded stocks. Additionally, investment banks want to profit from their business activities, and presumably try to favor their long-term clients, who are predominantly not the IPO firms. In conclusion, the process of going public is very complex, and the determining factors for IPO valuation and underpricing are highly ambiguous, even for practitioners and participants.

REFERENCES

141

References Aggarwal, Rajesh K./Krigman, Laurie/Womack, Kent L. (2002): Strategic IPO underpricing, information momentum, and lockup expiration selling, Journal of Financial Economics, Vol. 66, pp. 105-137. Aggarwal, Reena (2000): Stabilization Activities by Underwriters after Initial Public Offerings, The Journal of Finance, Vol. 55, No. 3, pp. 1075-1103. Aggarwal, Reena/Prabhala, Nagpurnanand/Puri, Manju (2002): Institutional Allocation in Initial Public Offerings: Empirical Evidence, Journal of Finance, Vol. 57, No. 3, pp. 1421- 1442. Akerlof, George A. (1970): "The market for lemons": Quality uncertainty and the market mechanism, Quarterly Journal of Economics 84, pp. 488-500. Alford, Andrew A. (1992): The Effect of the Set of Comparable Firms on the Accuracy of the Price-Earnings Valuation Method, Journal of Accounting Research, Vol. 30, No. 1, pp. 94-108. Allen, Franklin/Faulhaber, Gerald R. (1989): Signaling by the underpricing in the IPO market, Journal of Financial Economics 23, pp. 303-323. Alti, Aydogan (2005): IPO Market Timing, Review of Financial Studies, Vol. 18, No. 3, pp. 1106-1138. Aussenegg, Wolfgang/Pichler, Pegarte/Stomper, Alex (2002): Sticky Prices: IPO Pricing on the Nasdaq and Neuer Markt, Working Paper. Available at SSRN: http://ssrn.com/abstract=302917 or DOI: 10.2139/ssrn.302917 Bachmann, Ralph (2003): The Pricing of Initial Public Offerings when IPO Size and Investment Choice are Endogenous, Working Paper. Available at SSRN: http://ssrn.com/abstract=387680 or DOI: 10.2139/ssrn.387680 Baker, Malcolm/Wurgler, Jeffrey (2006): Investor Sentiment and the CrossSection of Stock Returns, The Journal of Finance, Vol. 61, No.4, pp. 16451680. Balvers, Ronald J./McDonald, Bill/Miller, Robert E. (1988): Underpricing of the New Issues and the Choice of Auditor as a Signal of Investment Banker Reputation, The Accounting Review, Vol. 63, No. 4, pp. 605-622.

142

REFERENCES

Baron, David R. (1992): A Model of the Demand for Investment Banking Advising and Distribution Services for New Issues, The Journal of Finance, Vol. 37, No. 4, pp. 955-976. Barry, Christopher B. (1989): Initial Public Offering Underpricing: The Issuer's View - A Comment, The Journal of Finance Vol. 44, No. 4, pp. 10991103. Bartov, Eli/Mohanram, Partha/Seethamraju, Chandrakanth (2002): Valuation of Internet Stocks - An IPO Perspective, Journal of Accounting Research, Vol.40, No. 2, pp. 321-346. Bayless, Mark/Chaplinsky, Susan (1996): Is There a Window of Opportunity for Seasoned Equity Issuance?, The Journal of Finance, Vol. 51, No. 1, pp. 253-278. Beatty, Randolph P. (1989): Auditor reputation and the pricing of the initial public offering, The Accounting Review, Vol. 64, No. 4, pp. 693-709. Beatty, Randolph P./Riffe, Susan/Thompson, Rex (2000): IPO Pricing with Accounting Information, Working Paper. Availabe at http://scholar.google.de Beatty, Randolph P./Ritter, Jay R. (1986): Investment Banking, Reputation, and the Underperformance of Initial Public Offerings, Journal of Financial Economics 15, pp. 213-232. Beatty, Randolph P./Welch, Ivo (1996): Issuer Expenses and Legal Liability in Initial Public Offerings, Journal of Law and Economics, Vol. 39, pp. 545602. Benveniste, Lawrence M./Busaba, Walid Y./William, Wilhelm J. Jr. (1996): Price stabilization as a bonding mechanism in new equity issues, Journal of Financial Economics 42, pp. 223-255. Benveniste, Lawrence M./Ljungqvist, Alexander/Wilhelm Jr., William/Yu, Xiaoyun (2003): Evidence of Information Spillover in the Production of Investment Banking Services, The Journal of Finance, Vol. 58, No. 2, pp. 577- 608. Benveniste, Lawrence M./Spindt, Paul A. (1989): How investment bankers determine the offer price and allocation of new issues, Journal of Financial Economics 24, pp. 343-361. Benveniste, Lawrence M./Wilhelm, William J. (1990): A comparative analysis of IPO proceeds under alternative regulatory environments, Journal of Financial Economics 28, pp. 173-207.

REFERENCES

143

Bessler, Wolfgang/Kurth, Andreas (2007): Agency Problems and the Performance of Venture-backed IPOs in Germany: Exit Strategies, Lock-up Periods, and Bank Ownership, European Journal of Finance, Vol. 13, Issue 1, pp. 29-63. Bessler, Wolfgang/Thies, Stefan (2002): Initial public Offering, Subsequent Seasoned Equity Offerings, and Long-Run Performance: Evidence from IPO in Germany, Working Paper. Available at SSRN: http://ssrn.com/abstract=302357 or DOI: 10.2139/ssrn.302357 Boehmer, Ekkehart/Fishe, Raymond P.H. (2000): Do underwriters encourage stock flipping? A new explanation for the underpricing of IPOs, Working Paper. Available at SSRN: http://ssrn.com/abstract=228434 or DOI: 10.2139/ssrn.228434 Boehmer, Ekkehardt/Ljungqvist, Alexander (2004): On the decision to go public: Evidence from privately-held firms, Working Paper. Available at SSRN: http://ssrn.com/abstract=266993 or DOI: 10.2139/ssrn.266993 Booth, James R./Chua, Lena (1996): Ownership Dispersion, Costly Information, and IPO Underpricing, Journal of Financial Economics 41, pp. 291-310. Booth, James R./Smith, Richard L. (1986): Capital Raising, Underwriting and the Certification Hypothesis, Journal of Financial Eonomics 15, pp. 261-281. Boulton, Thomas J./Smart, Scott B./Zutter, Chad J. (2007): Earnings Quality and International IPO Underpricing, Working Paper. Available at SSRN: http://ssrn.com/abstract=972694 Boulton, Thomas J./Smart, Scott B./ Zutter, Chad J. (2006): IPO Underpricing and the Market for Corporate Control, Working Paper. Available at SSRN: http://ssrn.com/abstract=928464 Bradley, Daniel E./Jordan, Bradford D. (2002): Partial Adjustment to Public Information and IPO Underpricing, Journal of Financial and Quantitative Analysis, Vol. 37. No. 4, pp. 595-616. Brau, James C./Fawcett, Stanley E. (2006): Initial Public Offerings: An Analysis of Theory and Practice, The Journal of Finance, Vol. 61, No. 1, pp. 399436. Brav, Alon/ Gompers, Paul A. (2003): The Role of Lockups in Initial Public Offerings, The Review of Financial Studies, Vol. 16, No. 2, pp. 1-29. Brennan, Michael J./Franks, Julian R. (1997): Underpricing, ownership, and control in initial public offerings of equity securities in the UK , Journal of Financial Economics 45, pp. 391-413.

144

REFERENCES

Brown, Stephen/Lo, Kin/Lys, Thomas (1999): Use of R² in accounting research: measuring changes in value relevance over the last four decades, Journal of Accounting and Economics 28, pp. 83-115. Burghof, Hans-Peter/ Kraus, Thilo (2003): Post IPOPerformance and the Exit of Venture Capitalist, Working Paper. Available at SSRN: http://ssrn.com/abstract=407846 or DOI: 10.2139/ssrn.407846 Carter, Richard B./Dark, Frederik H./Singh, Ajai K. (1998): Underwriter Reputation, Initial Returns, and the Long-Run Performance of the IPO Stocks, The Journal of Finance, Vol. 53, No. 1, pp. 285-311. Carter, Richard/Manaster, Steven (1990): Initial Public Offerings and Underwriter Reputation, The Journal of Finance, Vol. 45, No. 4, pp. 1045-1067. Chemmanur, Thomas J. (1993): The Pricing of Initial Public Offerings: A Dynamic Model with Information Production, The Journal of Finance, Vol. 48, No. 1, pp. 285-304. Chen, Hsuan-Chi/Ritter, Jay R. (2000): The Seven Percent Solution, The Journal of Finance Vol. 55, No. 3, pp.1105-1131. Cliff, Michael T./Denis, Davis J. (2004): Do Initial Public Offering Firms Purchase Analyst Coverage with Underpricing?, The Journal of Finance , Vol. 59, No. 6, pp. 2871-2901. Cogliati, Giordano/Paleari, Stefano/Vismara, Silvio (2008): IPO pricing: growth rates implied in the offer price, Working Paper. Available at SSRN: http://ssrn.com/abstract=965450 Core, John E./Guay, Wayne R./Van Buskirk, Andrew (2003): Market valuations in the New Economy: an investigation of what has changed, Journal of Accounting and Economics 34, pp. 43-67. Cornelli, Francesca/David Goldreich (2001) : Bookbuilding and Strategic Allocation, The Journal of Finance, Vol. 56, No. 6, pp. 2337-2369. Cornelli, Francesca/David Goldreich/Ljungqvist, Alexander (2006): Investor sentiment and Pre-IPO Markets, The Journal of Finance, Vol. 61, No. 3, pp. 1187-1216. Corwin, Shane A./Schultz Paul (2005): The Role of IPO Underwriting Syndicates: Pricing, Information Production, and the Underwriter Competition, The Journal of Finance, Vol. 60, No. 1, pp. 443-486. Deloof, Marc/De Maeseneire, Wouter/Inghelbrecht, Koen (2002): The Valuation of IPOs by Investment Banks and the Stock Market: Empirical Evidence,

REFERENCES

145

Working Paper. Available at SSRN: http://ssrn.com/abstract=298602 or DOI: 10.2139/ssrn.298602 Der Bundesverband Deutscher Kapitalbeteiligungsgesellschaften - German Private Equity and Venture Capital Association e.V. (BVK: 2007a): BVK Statistik: Das Jahr 2006 in Zahlen, http://www.bvkev.de/media/file/110.BVK_Jahresstatistik_2006_200207.pdf. Der Bundesverband Deutscher Kapitalbeteiligungsgesellschaften - German Private Equity and Venture Capital Association e.V. (BVK: 2007b): BVK Special: Venture Capital in den USA 2006, http://www.bvkev.de/media/file/141.Special_USA_2006.pdf. Derrien, Francoise (2005): IPO Pricing in "Hot" Market Conditions: Who Leaves Money on the Table?, The Journal of Finance, Vol. 60, No. 1, pp. 487-521. Dichev, Ilia C./Burgstahler, David C. (1997): Earnings, Adaptation, Equity Value, The Accounting Review, Vol. 72, No. 2, pp. 187-215. Dorn, Daniel (2003): Does sentiment drive retail demand for IPOs?, Working Paper. Available at SSRN: http://ssrn.com/abstract=472660 or DOI: 10.2139/ssrn.472660 Draho, Jason (2001): The Effects of Uncertainty on the Underpricing of IPOs, Working Paper. Available at SSRN: http://ssrn.com/abstract=271368 or DOI: 10.2139/ssrn.271368 Drake, Philip D./Vetsuypen, Michael R. (1993): Underpricing and Insurance Against Legal Liability, Financial Management 22, pp. 64-73. Dunbar, Craig G. (2000): Factors effecting investment bank initial public offering market share, Journal of Financial Economics 55, pp. 3-41. Ellis, Katrina/Michaely, Roni/O'Hara, Maureen (2000): When the Underwriter Is the Market Maker: An Examination of Trading in the IPO Aftermarket, The Journal of Finance, Vol. 55, No. 3, pp. 1039-1074. Ellul, Andrew/Pagano, Marco (2006): IPO Underpricing and After-Market Liquidity, The Review of Financial Studies, Vol. 19, No. 2, pp. 381-421. Field, Laura Casares /Sheehan, Dennis P. (2004): IPO Underpricing and Outside Blockholdings, Journal of Corporate Finance, Vol. 10, Issue 2, pp. 263280. Fischer, Christoph (2000): Why Do Companies Go Public? Empirical Evidence from Germany's Neuer Markt, Working Paper. Available at SSRN: http://ssrn.com/abstract=229529 or DOI: 10.2139/ssrn.229529

146

REFERENCES

Fischer, Paul E./Verrecchia, Robert E. (1997): The Effect of Limited Liability on the Market Response to Disclosure, Contemporary Accounting Research, Vol. 13, No. 3, pp. 515-541. Franzke, Stefanie A. (2003): Underpricing of Venture-Backed and Non VentureBacked IPOs: Germany's Neuer Markt, RICAFE Working Paper, No. 003. Available at SSRN: http://ssrn.com/abstract=482544 or DOI: 10.2139/ssrn.482544 Franzke, Stefanie A./Schlag, Christian (2003): Over-Allotment Options in IPOs on Germany's Neuer Markt: An Empirical Investigation, CFS Working Paper No. 2002/16. Available at SSRN: http://ssrn.com/abstract=384120 or DOI: 10.2139/ssrn.384120 Froot, Kenneth A./Scharfstein, David S./Stein, Jeremy C. (1992): Herd on the Street: Informational Inefficiencies in a Market with Short-Term Speculation, The Journal of Finance, Vol. 47, No. 4, pp. 1461-1482. Garfinkel, Jon A. (1993): IPO Underpricing, Insider Selling and Subsequent Equity Offerings: Is Underpricing a Signal of Quality?, Financial Management 22, pp. 74-83. Georgen, Marc/Khurshed, Arif/Renneboog, Luc (2006): Why are the French so different from the Germans? Underpricing of IPOs on the Euro New Markets, European Corporate Governance Institute, Finance Working Paper No.122/2006, International Review of Law and Economics (Forthcoming). Available at SSRN: http://ssrn.com/abstract=899071 Giudici, Giancarlo/Roosenboom, Peter (2002): Pricing Initial Public Offerings in Europe: What has changed?, Working Paper. Available at SSRN: http://ssrn.com/abstract=314275 or DOI: 10.2139/ssrn.314275 Gompers, Paul/ Lerner, Josh A. (1996): Grandstanding in the Venture Capital Industry, Journal of Finance 42, pp. 133-156. Gouldey, Bruce K. (2006): Uncertain Demand, Heterogeneous Expectations, and Unintentional IPO Underpricing, The Financial Review 41, pp. 33-54. Grinblatt, Mark/Hwang, Chuan Yang (1989): Signalling and the Pricing of New Issues, The Journal of Finance, Vol. 44, No. 2, pp. 393-420. Habib, Michael A./Ljungqvist, Alexander P. (2001): Underpricing and Entrepreneurial Wealth Losses in IPOs: Theory and Evidence, The Review of Financial Studies, Vol. 14, No. 2, pp. 433-458. Habib, Michael A./Ljungqvist, Alexander P. (1998): Underpricing and IPO proceeds: a note, Economic Letters 61, pp. 381-383.

REFERENCES

147

Hand, John R.M. (2000): Profits, losses and the non-linear pricing of Internet stocks, Working Paper. Available at SSRN: http://ssrn.com/abstract=204875 or DOI: 10.2139/ssrn.204875 Hanley, Kathleen (1993): The Underpricing of Initial Public Offerings and the Partial Adjustment Phenomenon, Journal of Financial Economics, Vol. 34, pp. 231-250. Hanley, Kathleen/Kumar, Arun A./Seguin, Paul J. (1993): Price stabilization in the market for new issues, Journal of Financial Economics 34, pp. 177-197. Hanley, Kathleen/Wilhelm, William J. (1995): Evidence on the strategic allocation of initial public offerings, Journal of Financial Economics 37, pp. 239-257. Helwege, Jean/Liang, Nellie (2004): Initial Public Offering in Hot and Cold Markets, Journal of Financial and Quantitative Analysis, Vol. 39, No. 3, pp. 541-569. Hill, Paula (2006): Ownership Structure and IPO Underpricing, Journal of Business Finance & Accounting 33, pp. 102-126. Hoffmann-Burchardi, Ulrike (2001): Clustering of initial public offerings, information revelation and underpricing, European Economic Review 45, p. 353-383. Hogan, Chris E. (1997): Cost and Benefits of Audit Quality in the IPO Markets: A Self Selection Analysis, The Accounting Review, Vol. 72, No. 1, pp. 6786. Hopp, Christian/Dreher, Axel (2007): Do Differences in Institutional and Legal Environments Explain Variations in IPO Underpricing?, CESIfo Working Paper Series No. 2082. Available at SSRN: http://ssrn.com/abstract=1006987 Hughes, Patricia J./Thakor, Anjan V. (1992): Litigation Risk, Intermediation, and the Underpricing of Inital Public Offerings, The Review of Financial Studies, Vol. 5, No. 4, pp. 709-742. Hunger, Adrian (2001): IPO-Underpricing und die Besonderheiten des Neuen Marktes - Eine ökonomische Analyse börsenrechtlicher Marktsegmentierung, Europäische Hochschulschriften, Reihe V, Volks- und Betriebswirtschaft, Frankfurt. Ibbotson, Roger G. (1975): Price Performance of Common Stock New Issues, Journal of Financial Economics 2, pp. 235-272. Ibbotson, Roger G./Jaffe, Jeffrey F (1975): Hot issue markets, Journal of Finance 30, pp. 1027–1042.

148

REFERENCES

James, Christopher (1992): Relationship Specific Assets and the Pricing of Underwriter Services, The Journal of Finance, Vol. 47, No. 5, pp. 1865-1885. Jegadeesh, Narasimhan/Weinstein, Mark/Welch, Ivo (1993): An empirical investigation of IPO returns and subsequent equity offerings, Journal of Financial Economics 34, pp. 153-175. Jenkinson, Tim/Ljungqvist, Alexander (2001): Going Public: The Theory and Evidence on How Companies Raise Equity Finance, Oxford University Press, Second Edition. Jenson, Michael C. (1986): Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers, The American Economic Review, Vol. 76, No. 2, pp. 323329. Johnson, James/Miller, Robert (1988): Investment Banker Prestige and the Underpricing of Initial Public Offerings, Financial Management 17, pp. 19-29. Kahnemann, Daniel/Tversky, Amos (1979): Prospect Theory: An Analysis of Decision Under Risk, Econometrica 47, pp. 263-291. Kim, Jeong-Bon/Krinsky, Itzhak/Lee, Jason (1995): The role of financial variables in the pricing of the Korean initial public offerings, Pacific-Basin Finance Journal 3, pp. 449-464. Kim, Moonchul/Ritter, Jay R. (1999): Valuing IPOs, Journal of Financial Economics 53, pp. 409-437. Klein, Peter G./Zoeller, Kathrin (2003): Universal Banking and Conflicts of Interest: Evidence from German Initial Public Offerings, University of Missouri, Working Paper No. 2003-06. Available at SSRN: http://ssrn.com/abstract=269797 Koop, Gary/Kai, Li (2001): The valuation of IPO and SEO firms, Empirical Finance 8, pp. 375-401.

Journal

of

Kraus, Thilo (2002): Underpricing of IPOs and the Certification Role of Venture Capitalists: Evidence from Germany's Neuer Markt, Working Paper. Available at SSRN: http://papers.ssrn.com/sol3/papers.cfm? abstract_id=301431 Kurth, Andreas (2005): Agency-Probleme und Performance von Initial Public Offerings: Eine empirische Untersuchung von Unternehmen des Neuen Marktes, Deutscher Universitäts-Verlag. Leite, Tore (2004): Excess initial returns in IPOs, Journal of Financial Intermediation 13, pp. 359-377.

REFERENCES

149

Leland, Hayne E./Pyle, David H. (1977): Informational Asymmetries, Financial Structure, and Financial Intermediation, The Journal of Finance, Vol. 32, No. 2, pp. 371-387. Leone, Andrew/Rock, Steve/Willenborg, Michael (2007): Disclosure of intended use of proceeds and underpricing in initial public offerings, Journal of Accounting Research, Vol. 45, Issue 1, pp. 111-153. Lin, Timothy H./Smith, Richard L. (1998): Insider reputation and selling decision: the unwinding of venture capital investments during equity IPOs, Journal of Corporate Finance 4, pp. 241-263. Liu, Laura Xiaolei/Shermann, Ann E./ Zhang, Yong (2008): The Role of Media in Initial Public Offerings, Working Paper. Available at SSRN: http://ssrn.com/abstract=890602 Ljungqvist, Alexander P. (1999): IPO Underpricing, Wealth Loss and the Curios Role of Venture Capitalists in the Creation of Public Companies, OFRC Working Paper Series 1999fe04. http://www.finance.ox.ac.uk/file_links/ finecon_papers/1999fe04.pdf Ljungqvist, Alexander P. (1997): Pricing initial public offerings: Further evidence from Germany, European Economic Review 41, pp.1309-1320. Ljungqvist, Alexander/Nanda, Vikram/Singh, Rajdeep (2006): Hot Markets, Investor Sentiment, and IPO Pricing, Journal of Business, Vol. 79, No. 4, pp. 1667-1702. Ljungqvist, Alexander P./Wilhelm, William J. Jr. (2005): Does Prospect Theory Explain IPO Market Behavior?, The Journal of Finance, Vol. 60, No. 4, pp. 1756-1790. Ljungqvist, Alexander/Wilhelm, William J. Jr. (2003): IPO Pricing in the Dot-com Bubble, The Journal of Finance, Vol.58, No. 2, pp. 723-752. Löffler, Gunter/Panther, Patrick F./Theissen, Erik (2005): Who knows what when? The information content of pre-IPO market prices, Journal of Financial Intermediation 14, pp. 466-484. Logue, Dennis E. (1973): On the pricing of unseasoned equity issues:1965-1969, Journal of Financial and Quantitative Analysis, Vol. 8, No. 1, pp. 91-103. Loughran, Tim/Ritter, Jay R. (2004): Why has IPO Underpricing Changed Over Time?, Financial Management, Autumn 2004, pp. 5-37. Loughran, Tim/Ritter, Jay R. (2002): Why Don't Issuers Get Upset About Leaving Money on the Table in IPOs?, The Review of Financial Studies, Special 2002, Vol. 15, No. 2, pp. 413-443.

150

REFERENCES

Loughran, Tim/Ritter, Jay R./Rydqvist, Kristian (1994): Initial Public Offerings: International Insights, Pacific-Basin Finance Journal, Vol. 2, pp. 165-199. Lowry, Michelle (2003): Why Does IPO Volume Fluctuate So Much?, Journal of Financial Economics, Vol. 67, No. 1, pp. 3-40. Lowry, Michelle/Schwert, William G. (2002): IPO Market Cycles: Bubbles or Sequential Learning?, The Journal of Finance, Vol. 57, No. 3, pp. 1171-1199. Lowry, Michelle/Shu, Susan (2002): Litigation risk and IPO underpricing, Journal of Financial Economics 65, pp. 309-335. Maksimovic,Vojislav/Pichler, Pegaret (2001): Technological Innovation and Initial Public Offerings, Review of Financial Studies, Vol. 14, No. 2, pp. 459494. Megginson, William L./Weiss, Kathleen A. (1991): Venture Capitalist Certification in Initial Public Offerings, The Journal of Finance, Vol. 46, No.3, pp. 879-903. Michaely, Roni/Shaw, Wayne H. (1994): The Pricing of Initial Public Offerings: Test of Adverse Selection and Signaling Theories, The Review of Financial Studies, Vol. 7, No. 2, pp. 279-319. Muscarella, Chris J./Vetsuypens, Michael R. (1989): A simple test of Baron's model of IPO underpricing, Journal of Financial Economics 24, pp. 125135. Myers, Stewart C./ Majluf, Nicholas S. (1984): Corporate financing and investment decisions when firms have information that investors do not have, Journal of Financial Economics 13, pp. 187-221. Oehler, Andreas/Rummer, Marco/Smith, Peter N. (2004): IPO Underpricing and the Relative Importance of Investor Sentiment - Evidence from Germany, Working Paper. Available at SSRN: http://ssrn.com/abstract=592302 Ofek, Eli/Richardson, Matthew (2002): The Valuation and Market Rationality of Internet Stock Prices, Oxford Review of Economic Policy, Vol. 18, No. 3, pp. 265-287. Petersen, Markus (2007): Is underpricing a deliberate choice? Evidence from Germany, Working Paper. Available at SSRN: http://ssrn.com/ abstract=1017352 Pástor, Lubos/ Veronesi, Pietro (2006): Rational IPO Waves, The Journal of Finance, Vol. 60, No. 4, pp. 1713-1757.

REFERENCES

151

Rees, William A. Jr. (2003): IPO Underpricing, Trading Volume, and Investor Interest, Working Paper. Available at SSRN: http://ssrn.com/abstract=115348 or DOI: 10.2139/ssrn.115348 Reuter, Jonathan (2006): Are IPO Allocation for Sale? Evidence from Mutual Funds, The Journal of Finance, Vol. 61, No. 5, pp. 2289-2324. Reilly, Frank K. (1973): Further Evidence on Short Run Results for New Issue Investors, Journal of Financial and Quantitative Analysis, Vol. 8, No. 1, pp. 83-90. Ritter, Jay R. (2003): Differences between European and American IPO Markets, European Financial Management, Vol. 9, No. 4, pp. 421-434. Ritter, Jay R. (1991): The Long-Run Performance of Initial Public Offerings, The Journal of Finance, Vol. 46, No.1, pp. 3-27. Ritter, Jay R. (1984): The "Hot Issue" Market of 1980, Journal of Business, Vol. 57, No. 2, pp. 215-240. Ritter, Jay R./Welch, Ivo (2002): A Review of IPO Activity, Pricing and Allocation, The Journal of Finance, Vol. 57, No. 4, pp.1795-1828. Rock, Kevin (1986): Why New Issues are Underpriced, Journal of Financial Economics 15, pp. 187-212. Roosenboom, Peter (2005): How Underwriters Value IPOs?: An Empirical Analysis of the French IPO Market, Working Paper. Available at SSRN: http://ssrn.com/abstract=675242 Ruud, Judith S. (1993): Underwriter price support and the IPO underpricing puzzle, Journal of Financial Economics 24, pp. 135-151. Schenone, Carola (2004): The Effect of Banking Relationships on the Firm's IPO Underpricing, The Journal of Finance, Vol. 59, No. 6, pp. 2903-2958. Schrand, Catherine/Verrecchia, Robert E. (2005): Information Disclosure and Adverse Selection Explanations for IPO Underpricing, Working Paper. Available at SSRN: http://ssrn.com/abstract=316824 or DOI: 10.2139/ssrn.316824 Schultz, Paul H./Zaman, Mir A. (1994): Aftermarket support and the underpricing of the initial public offerings, Journal of Financial Economics 35, pp. 199219. Sherman, Ann E./Titmann, Sheridan (2002): Building the IPO Order Book: Underpricing and the Participation Limits with Costly Information, Journal of Financial Economics, Vol. 65, Issue 1, pp. 3-29.

152

REFERENCES

Slovin, Myron B./Young, John E. (1990): Bank Lending and Initial Public Offerings, Journal of Banking and Finance 14, pp. 729-740. Spence, Andrew M. (1974): Market Signalling, Cambridge, Mass.: Harvard University Press. Stoughton, Neil M./Zechner, Josef (1998): IPO-mechanisms, monitoring and ownership structure, Journal of Financial Economics, 49, pp. 45-77. Subrahmanyam, Avanidhar/Titman, Sheridan (1999): The Going-Public Decision and the Development of Financial Markets, The Journal of Finance, Vol. 64, No 3, pp. 1045-1082. Tinic, Shea M. (1988): Anatomy of Initial Public Offerings of Common Stock, The Journal of Finance, Vol. 43, No. 4, pp. 789-822. Titman, Sheridan/Trueman, Brett (1989): Information Quality and the Valuation of New Issues, Journal of Accounting and Economics, Vol. 8, pp. 159-172. Truemann, Brett/Wong, M. H. Franco/Zhang, Xiao-Jun (2000): The Eyeballs Have It: Searching for the Value in Internet Stocks, Journal of Accounting Research, Vol. 38 Supplement, pp. 137-162. Tykvová, Tereza/Walz, Uwe (2007): How important is participation of different venture capitalists in German IPOs?, Global Finance Journal 17, pp. 350378. Wasserfallen, Walter/Wittleder, Christian (1994): Pricing initial public offerings: evidence from Germany, European Economic Review 38, pp. 1505-1517. Welch, Ivo (1992): Sequential Sales, Learning, and Cascades, The Journal of Finance, Vol. 47, No. 2, pp. 695-732. Welch, Ivo (1989): Seasoned Offerings, Imitation Costs, and the Underpricing of Initial Public Offerings, The Journal of Finance, Vol. 44, No. 2, pp. 421449. Willenborg, Michael (1999): Empirical Analysis of the Economic Demand for Auditing in the Initial Public Offering Market, Journal of Accounting Research, Vol. 37, No. 1, pp. 225-238. Zingales, Luigi (1995): Insider Ownership and the Decision to Go Public, The Review of Economic Studies, Vol. 62, No. 3, pp. 425-448.

APPENDIX CHAPTER I

153

Appendix: Chapter I The following tables give a brief summary of discussed literature in chapter I.

AI Information Asymmetries between Issuer and Investor A I.1 Signalling Theory Model: Future Performance and Seasoned Equity Offerings Source

Sample size (Country)

Time

Variable

Results

Bessler/Thies (2002: 12 et. seqq.)

218 (Germany)

19771995

Subsequent seasoned offerings during 36month after the IPO

Firms with high underpricing are more likely to make an SEO.

Garfinkel (1993: 549 (USA) 198075, 82) 1983

SEO within seven years after IPO

No effect on likelihood of reissue and abnormal return to the announcement of SEO

James (1992: 1871, 1875)

Public offerings (common stock, straight debt, convertible debt, common stock with warrants)

Mean IPO percentage underpricing is higher for firms that do not make a subsequent offer.

520 (USA) 19801983

APPENDIX CHAPTER I

154

Jegadeesh/ Weinstein/ Welch (1993: 157)

1985 (USA)

19801986

SEO within three years after IPO, returns on SEOs

Likelihood, size, timing of SEO are less predictable with underpricing than with aftermarket returns. Market reacts less unfavourably to SEO announcements by firms with high underpricing

Michaely/Shaw (1994: 283, 303 et. seqq.)

947 (USA) 19841988

Dividend yield, time after IPO to announcement, excess return on three days surrounding announcement

No indication that greater underpricing is positively related to subsequent dividend policy. Market does not react to the announcement of dividend payments.

Michaely/Shaw (1994: 283, 307 et. seqq.)

947 (USA) 19841988

Size of seasoned equity or debt issue in relation to IPO, market reaction to announcement

Higher underpricing is not associated with higher likelihood of future reissuance of debt or equity: Unfavorable price reaction at the announcement of SEO, regardless of the level of underpricing.

Welch (1989: 443 et.seqq.)

1028 (USA)

Public SEO within sample period

Timing of SEO related to IPO, IPO used to advertise for SEO

19771982

APPENDIX CHAPTER I

155

Model: Insider Ownership Source

Sample size (Country)

Time

Variable

Results

Franzke (2003: 10, 20)

353 (Germany)

19972002

Participation (secondary shares divided by pre-IPO shares), dilution (primary shares divided by pre-IPO shares)

No significant relationship between participation/ dilution and underpricing

Giudici/ Roosenboom (2002: 10, 22)

1245 (Euro New Markets)

19952001

Participation, dilution

Significant negative relationship of underpricing and dilution/ participation ratio

Habib/ 1376 Ljungqvist (USA) (2001: 439, 449)

19911995

Participation, dilution

Significant negative correlation between participation and underpricing

Ljungqvist (1999: 10,16)

1421 (USA)

19961999

Participation, dilution

Significant negative correlation between participation and underpricing

Ljungqvist (1997: 1310, 1315)

189 (Germany)

19701993

Fraction of share capital retained by insiders after the IPO

Positive correlation between insider ownership and underpricing

Michaely/Shaw (1994: 283, 311 et. seqq.)

947 (USA)

19841988

Percentage held by insiders, risk (var. after 60 days), firm value (ROE value for 2 years)

No significant explanation of insider ownership for underpricing and firm value

156

APPENDIX CHAPTER I

Wasserfall/ 92 Wittleder (1994: (Germany) 1507, 1511)

19611987

Retention rate (ratio of numbers of shares held by old owners to total shares after the IPO)

Insignificant positive correlation of retained shares to underpricing

A I.2 Certification of Quality Model: Auditor Certification Source

Sample size (Country)

Time

Variable

Results

Beatty (1989: 696, 707)

2567 (USA)

19751984

Auditor fee (cash compensation, including auditing, printing, legal-, other miscellaneous fees)

Significant negative relation between auditor fees and underpricing

Boulton/ Smart/Zutter (2007: 7, 13)

7306 (34 countries)

Earnings quality (manipulation opportunities, accounting standards, earnings aggressiveness, earnings opacity)

Manipulation of earnings, weak accounting standards, earnings aggressiveness, earnings opacity are positively correlated to underpricing

Hogan (1997: 70, et. seqq.)

692 (USA)

Auditor fee (considering auditor choice as endogenous variable), Big 6 auditors

Issuer selects auditor which minimizes the costs of underpricing and compensation.

19901992

APPENDIX CHAPTER I

157

Hopp/Dreher (2007: 11, 16, 23)

29 countries

19882005

Accounting rating (measured by inclusion of 90 accounting items in the balance sheet)

Accounting rating of countries has negative impact on underpricing.

Hunger (2002: 126, 191)

270 (Germany)

19971999

Disclosure requirements of stock segments

Higher underpricing in “Neuer Markt”

Leone/ Rock/ Willenborg (2006: 11, 15)

787 (USA)

19931994

Detail of use of Increase in dollar deproceeds (detail regarding the use tail in dollars) of proceeds is associated with less underpricing.

Schrand/ 129 (USA) Verrecchia (2005: 7, 10, 18)

19901999

Disclosure (announcements 90 days before to 180 days after IPO)

Significant negative relationship between informative disclosures and IPO underpricing

Willenborg 280 (USA) (1999: 227, 229, 234)

19931994

Auditor fees (distinguishing between startup/non start-up and IPO proceeds)

Start-ups are associated with insurance signal of auditors, non-start-ups with insurance and certification signal.

Model: Venture Capitalist Certification Source

Sample size (Country)

Time

Variable

Results

Bessler/Kurth (2004: 35 et. seq., 40)

307 (Germany)

19972001

Percentage of VC equity prior to IPO (independent VCs, 4% stock ownership)

Insignificant higher underpricing for VCbacked IPOs

APPENDIX CHAPTER I

158

Bessler/Kurth (2004: 35 et seq., 44)

307 (Germany)

19972001

Percentage of VC equity holding prior to IPO (independent venture capitalists, 4% stock ownership)

Underpricing for firms with min. lockup of 6 months: 45.7%. Underpricing of firms with voluntary extended lockups: 22.1%

Brav/Gompers (2003: 6, 9)

2871 (USA)

19881996

Venture capital backed, lockup length, postIPO insider shares locked

High quality underwriters and VC have shorter lockup periods; high risk and high-growth companies utilize longer lockup length

Bradley/Jordan 3325 (2002: 607, 613) (USA)

19901999

VC-backed offers (controlling for overhang, partial adjustment of offer prices, hot issue periods)

No significant difference between VCbacked and non-VCbacked issues, after controlling for industry effects

Franzke (2003: 10, 20)

353 (Germany)

19972002

Percentage of Significant higher unVC equity derpricing for prestigholding prior to ious VC-backed IPOs IPO

Georgen/ Khurshed/ Renneboog (2006: 10, 14)

265 (Germany)

19962000

VC-backed offers

Insignificant lower underpricing for VCbacked IPOs

Georgen/ Khurshed/ Renneboog (2006: 10,19)

265 (Germany)

19962000

VC-backed offers

No stringent relation between lockup agreements and underpricing

APPENDIX CHAPTER I

159

Kraus (2002: 10, 22)

323 (Germany)

19972001

VC backed offers (official institution, preIPO ownership above 5%)

No impact of VC financing, when controlling for underwriter prestige and exante uncertainty

Lin/Smith (1998: 247, 261 et.seq.)

2634 (USA)

19791990

Percentage of VC equity holding prior to IPO, during 3 years afterwards controlling for VC reputation

Underpricing is highest when investors with established reputation sell and lowest when they do not. Certification of VCs reduces underpricing.

Megginson/ Weiss (1991: 880 et. seq., 889)

640 (USA)

19831987

VC-backed offers (comparing non VCbacked offers)

Less underpricing, lower underwriter compensation for VCbacked firms

Time

Variable

Results

Universal bank (commercial and investment banking), Specialist (primary SIC code, no lending, no commercial- or retail-banking).

Universal bank underwriting is related to higher than average underpricing.

Model: Bank Certification Source

Sample size (Country)

Klein/Zoeller (2003: 5,10)

270 1997(Germany) 1999

160

APPENDIX CHAPTER I

Schenone (2003: 306 2912, 2915, (USA) 2922)

19982000

Slovin/Young (1990: 732 et.seq.)

316 1980(Germany) 1984

Relationship banks (lending relationship or debt issues) as underwriters or switching, controlling for ability to underwrite

Firms that cannot go public with their relationship bank as an underwriter have highest underpricing compared to switchers and nonswitchers.

Bank debt, credit lines

Lower underpricing of firms with credit lines and debt

AII Information Asymmetries between Investors A II.1 The Winner’s Curse Model: Rock (1986) Source

Sample size (Country)

Time

Variable

Results

Aggarwal/ Prabhala/ Puri (2002: 1425, 1427)

174 (USA)

19971998

Allocation and profits for institutional investors

Higher profits for institutional investors compared to retail investors due to favorable allocation

19831988

Initial return for percentage of shares allocated to institutional investors

Institutional investors receive higher proportion of under- and overpriced issues.

Hanley/Wilhelm 38 (USA) (1995: 242, 246)

APPENDIX CHAPTER I

161

Michaely/Shaw (1994: 285, 289 et. seqq.)

58 (USA)

19841988

Average underpricing of MLP IPOs

Underpricing significantly different from zero for IPO of MLPs

Tinic (1988: 795, 796)

84 (USA)

19681969

Institutions’ investments in under/overpriced issues

No significant result of institutional investors receiving disproportionately large amounts of underpriced IPOs

Model: Higher Uncertainty Source

Sample size (Country)

Time

Variable

Results

Beatty/Ritter (1986: 218, et.seq., 223)

1028 (USA)

19771983

Proceeds in prospectus, reciprocal of gross proceeds

Positive relation between proxies of ex ante uncertainty and underpricing

Ljungqvist (1997: 1310, 1315)

189 (Germany)

19701993

Inverse of real Significant positive gross prorelationship between ceeds inverse of real gross proceeds and underpricing

Ritter (1984: 216, 224, 228)

1075 (USA)

19771982

Annual sales, standard deviation of aftermarket return (20 days)

Significant positive relationship between sales and standard deviation and underpricing

Wasserfallen/ Wittleder (1994: 1507, 1513)

92 (Germany)

19611987

Standard deviation of returns (1 month), annual sales, proceeds

Significant positive relationship between standard deviation of returns during first month and underpricing

APPENDIX CHAPTER I

162

Model: Underwriter Reputation Source

Sample size (Country)

Time

Variable

Results

Balvers/ McDonald/ Miller (1988: 614 et. seq., 618)

1182 (USA)

19811985

Underwriter reputation (“Top 25” reported by Institutional Investor); auditor reputation (Big Eight)

Investment bank and auditor reputation have significant negative effect on underpricing.

Beatty/Welch (1996: 556, 561, 584)

952 (USA)

19811984, 19921994

Ranking by market share 3 months before IPO

Lower underpricing with higher quality underwriters in 1980s, reversed relation in 1990s

Carter/ Dark/Singh (1998: 287, 289, 294)

2292 (USA)

19791991

Four-tier ranking (Johnson/ Miller (1988)), relative market share (Megginson/ Weiss (1991)), tentier ranking by tombstone announcements (Carter/ Manaster (1990))

Significant negative coefficients for separate regression of reputation measures: Carter/Manster measure is the only significant measure when introduced simultaneously in the regression analysis

19791983

Ten-tier ranking by tombstone announcements

Issues with prestigious underwriters have less underpricing than those of non-prestigious

Carter/ 501 (USA) Manaster (1990: 1053 et. seq., 1062)

APPENDIX CHAPTER I

Cliff/Denis (2004: 2877, 2885)

1050 (USA)

163

19932000

Ten-tier ranking by tombstone announcements (Carter/ Manaster (1990))

Positive relationship between underpricing and reputation of underwriter

Franzke (2003: 353 10, 14, 22) (Germany)

19972002

Lead managements, relative volume of proceeds

No significant correlation

Giudici/ Roosenboom (2002: 10, 24)

1245 (Euro New Markets)

19952001

Underwriter market share of gross proceeds in the local market

Significant lower underpricing for issues with prestigious underwriters

Habib/ Ljungqvist (2001: 445, 449)

1376 (USA)

19911995

Promotion costs (auditor fee, legal advisor, road show, exchange, printing, accountable/no accountable underwriter expenses)

Higher promotion expenses result in lower underpricing

APPENDIX CHAPTER I

164

Hopp/Dreher (2007: 11, 16, 23)

29 countries

19882005

Credit regulation, foreign bank assets, foreign banks/total banks, ownership restriction, security business restriction, deposit insurance scheme, Supervisor

Significant results for security business restriction (+), foreign banks (-), deposit insurance scheme (-), credit regulation (-) on underpricing

A II.2 Information Revelation Model: Benveniste/Spindt (1989) Source

Sample Time size (Country)

Variable

Results

Bradley/ Jordan (2002: 607, 604)

3325 (USA)

19901999

Pre-offer price adjustment, final adjustment

Highest underpricing for offers with positive price revision, prior positive adjustment of price range is also significant for underpricing

Cornelli/ Goldreich (2001: 2339, 2353 et. seqq.)

39 (USA)

19951997

Bid size, bid type

Favorable allocation to price-bids and larger bid size

Corwin/ Schultz (2005: 450 et. seq., 467)

2146 (USA)

19972002

Syndicate size, size of price revision

Higher underpricing and price revision with larger syndicate. Public information included in price revision, not related to syndicate size.

APPENDIX CHAPTER I

165

Hanley 1430 (19943: 234 et. seqq.)

19831987

Offer price Higher offer size and revision, offer underpricing for issues amount with positive price revision

Hopp/Dreher (2007: 11, 16, 24)

19882005

Dummy for fixed-price, bookbuilding, auction

Pricing mechanism not significantly related to variations in underpricing

Liu/Shermann/ 3627 Zhang (2008: (USA) 5, 11)

19802004

Media attention (the number of articles mentioning the firm from one day after filing date to one day before the offering date)

With positive price revision, media intention is significantly correlated to underpricing.

Löffler/ Panther/ Theissen (2005: 472 et. seqq.)

357 (Germany)

19982001

Pre-IPO (grey Positive relation bemarket) bidtween grey market ask spreads prices and offering price

Loughran/ Ritter/ Rydqvist (1994: 167, 172 et. seqq.)

25 countries

19701990

Best-effort, firmcommitment contracts

29 countries

Contracts are not significantly related to variations in underpricing.

APPENDIX CHAPTER I

166

AIII Information Asymmetries between Issuer and Investor Model: Baron (1982) Source

Sample size (Country)

Time

Variable

Results

Muscarella/ Vetsuypens (1989: 128,130)

38 (USA)

19781987

Investment bank’s underpricing

Underpricing significantly different from zero at 5% level

AIV Symmetric Information Model: Underwriter Price Support Source

Sample size (Country)

Time

Variable

Results

Ellis/Michaely/ O’Hara (2000: 1046)

306 (USA)

19961997

Inventory size of underwriters as market makers, exercise of overallotment option

Inventory size is highest for issues without overallotment, market making is profitable and trading profits increase with underpricing

19972001

Calculation of overallotment arrangements, frequency of market price below offer price, market adj. performance 20 days after trading

No reduced underpricing, no price stability, no improved performance with overallotment arrangements

Franzke/Schlag 352 (2003: 12, 20 et. (Germaseqq.) ny)

APPENDIX CHAPTER I

167

Hanley/Kumar/ Seguin (1993: 182 et. seq.)

1523 (USA)

19821987

Bid-ask spread (closing bid price/floor price; BlackScholes put option value)

Stabilizing bids during the first 10 trading days

Ruud (1993: 143, 144)

468 (USA)

19821983

Distribution of initial returns

Large disparity between mean and median of initial return, suppressed left tail (negative) initial returns

1992

Inside bid and Underwriter’s market ask quotes makers are significantly more often on inside bids than asks, compared to other market makers.

Schultz/Zaman 72 (USA) (1993: 205, 210)

Model: Litigation Risk Source

Sample size (Country)

Time

Variable

Results

Drake/ Vetsuypen (1993: 64, 71)

93 (USA)

19691990

Sued IPO firms for misstatement in offering prospectus

No correlation between underpricing and expost liabilities

Hopp/Dreher (2007: 11, 21)

29 countries

19882005

Dummies for class action lawsuits, ability to recover losses, criminal sanctions

No significant correlation between underpricing and legal environment

APPENDIX CHAPTER I

168

Loughran/ 25 countRitter/ Rydqvist ries (1994: 167, 174)

19601992

Suggestion of more underpricing in USA, because of class action lawsuits

Insurance no major determinant of underpricing, as underpricing is not significantly higher in USA

Lowry/Shu (2002: 314 et. seq., 331)

1841 (USA)

19881995

Sued/nonsued firms, with litigation probability as an endogenous variable

Firms with higher litigation risk underprice more, greater underpricing lowers litigation risk (marginally significant coefficient).

Tinic (1988: 803 et. seqq.)

204 (USA)

19231930 19661971

Sample before and after Securities Act 1933

Underpricing is higher after 1933 with 11.6% compared to 5.17% before the Securities Act 1933.

Model: Ownership Structure Source

Sample size (Country)

Time

Variable

Results

Boehmer/Fishe (2000: 26, 30 et.seq.)

110 (USA)

19971998

Flipping transactions, volume weighted bidask spreads

Positive relation between flipping and market maker revenue

Boulton/Smart/ Zutter (2006: 9, 13, 17)

3956 (USA)

19902000

Pre-IPO M&A activity, ownership structure after IPO

Significant positive relation between underpricing and pre-IPO M&A activity, higher probability of takeover for firms with institutional investors

APPENDIX CHAPTER I

169

Brennan/ Franks (1997: 397 et. seqq.)

64 (UK)

19861989

Oversubscription, percentage of firm sold, allocation of underpricing costs between directors, nondirectors

Cost of underpricing higher for non-director shareholders, level of oversubscription increases with underpricing, discrimination against large applicants is significant.

Ellul/Pagano (2006: 396 et. seqq., 413)

337 (UK)

19982002

Liquidity, variability of liquidity: PIN by Easley et. al. 1996, adverse selection component of spread by Lin, Sanger and Booth (1995), effective spread

Positive correlation between underpricing and liquidity and liquidity risk.

Field/Sheehan (2004: 266, 269) Fischer (2000: 4, 22)

953 (USA)

19881992

Pre-IPO blockholder

Blockholders already exist before IPO.

163 (Germany)

19951997

Management equity/residual blockholder before IPO, allocation of issues (subsample)

Higher selling of outside blockholders relative to management/owners, average allotment of 3.8% of issued shares

Georgen/ Khurshed/ Renneboog (2006: 10, 33)

265 (Germany)

19962000

Percentage shares owned by underwriters

Significant positive correlation between underpricing and underwriter ownership

APPENDIX CHAPTER I

170

Petersen (2007: 14, 25, 32 et.seq.)

385 (Germany)

19972002

Insider/Outsider control (more than 25% of voting rights), cost of underpricing in relation to preIPO holdings valued with issue price

Insider ownership is positive related to underpricing, marginal costs for insiders negatively correlated to underpricing, no significant result for outsiders

Model: Behavioural Finance Source

Sample size (Country)

Time

Variable

Results

Cornelli/ Goldreich/ Ljungqvist (2006: 1198, 1203 et. seqq.)

486 (Europe)

19952002

Last grey market price before IPO, midpoint price range, aftermarket trading price, trading volume

Positive correlation of grey market prices and aftermarket prices, positive correlation between grey market prices and underpricing, aftermarket trading volume is higher with higher grey market prices

Dorn (2003: 8, 18 et seqq.)

289 (Germany)

19982002

Media attention (newspaper articles), transaction volume (12 month around IPO) of retail IPO purchases

Higher media attention results in higher preIPO purchase volume and aftermarket demand of retail investors. Higher underpricing is related to higher purchase volume of retailers.

APPENDIX CHAPTER I

Ljungqvist/ 3435 (US) Wilhelm (2005: 1766 et. seq., 1785)

19932000

171

Switching underwriter for SEO, issuer’s satisfaction in terms of wealth loss (underpricing) and wealth gain (revaluation of retained shares)

Issuers’ satisfaction (underpricing and price revaluation) is determinant for switching underwriters for SEOs, after controlling for reputation and analyst coverage.

Model: Information Momentum Source

Sample size (Country)

Time

Variable

Results

Aggarwal/ Krigman/ Womack (2002: 116, 124 et. seqq.)

618 (USA)

19941999

Timing, quantity of research recommendation and comments, broker comments, number of mentioning in First Calls from IPO to one month after lockup expiration

Underpricing leads to higher overall research coverage, greater underpricing leads to higher insider selling at the expiration of lockup period.

Bessler/Kurth (2004: 35 et. seq., 40)

307 (Germany)

19972001

Bank ownership pre-IPO, buy recommendation

Insignificant better performance of recommended issues during 6 month after IPO

APPENDIX CHAPTER I

172

Cliff/Denis 1050 (2004: 2877 et. (USA) seq., 2883)

19932000

Positive correlation Lead underbetween analyst coverwriter recommendation age and underpricing during 1 year after the IPO, underwriter as Institutional Investor’s allstar analyst team

Dunbar (2000: 12, 24)

3591(USA) 19841994

Overall bank’s analyst ranking (Institutional Investor)

Positive correlation between analyst ranking and underwriter market share, negative relation between change in analyst ranking/ industry specialisation and market share

Loughran/ Ritter (2004: 12, 25)

6391 (USA)

19802003

Carter/Manast er ranking for all-star analyst (rank of 8 or higher)

Positive correlation between the top-tear underwriter and underpricing during 1990s, negative/insignificant coefficient in 1980ies/post-bubble

Rees (2003: 11, 20, 22 et.seq.)

3541 (USA)

19831993

Trading volume divided by number of shares issued over three years, newspaper references by 25 major US newspapers

Pre-IPO newspaper references influence trading volume and initial returns, high underpricing leads to post IPO newspaper references, newspaper references result in higher trading volume.

APPENDIX CHAPTER II

173

Appendix: Chapter II The following table gives an overview of underwriter activity in the analyzed sample of IPOs. Appendix II: Table I

Underwriter Activity (1) Underwriter ranking is similar to Franzke (2003: 14). "Relative Number" stands for the investment bank's number of lead management IPOs in relation to total sample IPOs. "Relative Proceeds" accounts for the volume of underwritten proceeds of the bank in relation to total proceeds of the sample IPOs. Relative Number ABN Amro Bank N.V. Atlas Acquisition Holdings Corp. Baader Wertpapierhandelsbank Bank J. Vontobel & Co. AG Bankgesellschaft Berlin AG Bankhaus Hermann Lampe KG Bayerische Landesbank Berenberg Bank KG Berliner Effektenbank AG BHF Bank AG BNP Paribas Bank Group CC Bank AG Citigroup Inc. Commerzbank AG Concord Effekten AG Conrad Hinrich Donnerbank AG Credit Suiss Group Deutsche Bank AG Dresdner Bank AG DZ Bank AG Equinet AG FleetBoston Robertson Stephens International Ltd. Gebhardt & Co. Wertpapierhandelsbank AG Goldman Sachs Group Inc. Gontard & MetallBank AG Hamburgische Landesbank Hauck & Auffhäuser Privatbankiers KGaA HSBC Trinkaus & Burkhardt KGaA HWAG Hanseatisches Wertpapierhandelshaus AG Hypovereinsbank AG/Unicredit ICE Securities Limited J. Henry Schroder & Co. Limited JP Morgan Chase & Co. K/L/M Equity AG Kling Jelko Wertpapierhandelsbank AG

0.0044 0.0133 0.0288 0.0066 0.0022 0.0022 0.0044 0.0044 0.0110 0.0288 0.0155 0.0022 0.0133 0.0664 0.0354 0.0001 0.0221 0.1084 0.0907 0.0973 0.0177 0.0088 0.0022 0.0243 0.0265 0.0022 0.0022 0.0265 0.0022 0.0575 0.0022 0.0088 0.0288 0.0022 0.0066

Relative Proceeds 0.0042 0.0192 0.0093 0.0018 0.0019 0.0199 0.0024 0.0004 0.0060 0.0107 0.0049 0.0004 0.0010 0.0876 0.0673 0.0001 0.0159 0.4215 0.0773 0.1197 0.0075 0.0026 0.0476 0.0108 0.0157 0.0043 0.0199 0.0078 0.0043 0.0287 0.0002 0.0040 0.0230 0.0004 0.0034

174

APPENDIX CHAPTER II

Underwriter Activity (2) Landesbank Baden-Württemberg Landesbank Hessen-Thüringen Landesbank Rheinland-Pfalz Lang & Schwarz Wertpapierhandelsbank AG Lehman Brothers Inc. M. M. Warburg & Co. KGaA 'Merck, Finck & Co. Privatbankiers Morgan Stanley Norddeutsche Landesbank Girozentrale Quirin Bank AG Raiffeisen Zentralbank Österreich AG SAB AG Sal. Oppenheim jr & Cie. KGaA Société Générale S.A. Trigon Wertpapierhandelsbank AG UBS AG VEM Bank AG Viscardi AG Weserbank AG Westdeutsche Landesbank

0.0531 0.0022 0.0022 0.0022 0.0111 0.0265 0.0044 0.0288 0.0199 0.0133 0.0022 0.0022 0.0531 0.0022 0.0022 0.0265 0.0310 0.0044 0.0022 0.0487

0.0496 0.0000 0.0039 0.0003 0.0158 0.0079 0.0004 0.0165 0.0153 0.0328 0.0001 0.0003 0.0234 0.0013 0.0009 0.0435 0.0276 0.0005 0.0018 0.0159

APPENDIX CHAPTER III

175

Appendix: Chapter III The following figures show auto-/cross-correlation of IPO and market MB/PEratios for the complete sample as well as the technology and industry sector. Additional regression model results are presented in Appendix III Table I. Appendix III: Figure I

Auto/ Cross-Correlation of MB/PE Ratios of all IPOs 0,8 0,6 0,4 0,2 0 -0,2 -0,4 -0,6 -0,8 1997-1998

1999-2000

2001

2002-2004

Total MB IPO

Total PE IPO

Total MB-Corr.

Total PE-Corr.

2005-2007

APPENDIX CHAPTER III

176

Appendix III: Figure I

0,4

Auto/Cross-Correlation of MB/PE Ratios of Technology IPOs

0,3 0,2 0,1 0 -0,1 -0,2 -0,3 1997-1998

1999-2000

2001

2002-2004

2005-2007

Technology MB IPO

Technology PE IPO

Technology MB-Corr.

Technology PE-Corr.

Appendix III: Figure II

Auto/Cross-Correlation of MB/PE Ratios of all Industry IPOs 0,6 0,5 0,4 0,3 0,2 0,1 0 -0,1 -0,2 1997-1998

1999-2000

2001

2002-2004

Industry MB IPO

Industry PE IPO

Industry MB-Corr.

Industry PE-Corr.

2005-2007

APPENDIX CHAPTER III

177

Appendix III: Table I

Regression Models on Market Value (Hot/Cold) (1) For definition of variables look chapter III: table I. The regression models use White's (1980) heteroskedasticity-consistent standard errors and covariance. The values of the t-statistics are denoted in parentheses. Significance is indicated with * for 10% level of significance, ** for 5% level of significance, *** for 1% level of significance.

Assets Intanratio Equity Leverage Capex ROE EPS

MV(Peliminary) [1]

MV(Offer) [2]

MV(Last) [3]

0.694

0.695

0.656

(4.904)***

(4.776)***

(4.819)***

-0.001

-0.001

-0.001

(-2.109)*

(-2.211)**

(-2.341)**

-0.107

-0.107

-0.038

(-0.713)

(-0.711)

(-0.264)

-0.256

-0.306

-0.432

(-0.541)

(-0.623)

(-0.869)

-0.054

-0.054

-0.049

(-0.874)

(-0.850)

(-0.759)

-0.001

-0.001

0.001

(-0.050)

(-0.145)

(0.550)

0.003

0.003

0.004

(0.945)

(0.916)

(1.283)

PE

0.001

0.001

-0.001

(0.188)

(0.176)

(-0.041)

MB

0.004

0.003

-0.011

(0.299)

(0.199)

(-0.595)

0.120

0.066

0.195

(0.175)

(0.096)

(0.268)

Vola Perform Cold*Assets Cold*Intanratio Cold*Equity Cold*Leverage Cold*Capex Cold*ROE Cold*EPS Cold*PE

0.0120

0.022

0.026

(4.096)***

(4.480)***

(4.641)***

-0.066

-0.093

0.060

(-0.322)

(-0.457)

(0.337)

0.001

0.001

0.001

(2.282)**

(2.347)**

(1.767)*

0.450

0.060

-0.139

(0.224)

(0.305)

(-0.783)

-0.441

-0.390

-0.554

(-0.748)

(-0.645)

(-0.906)

0.166

0.179

0.174

(2.147)**

(2.269)**

(2.108)**

0.001

0.001

-0.001

(0.431)

(0.516)

(-0.246)

0.041

0.044

0.051

(1.137)

(1.204)

(1.109)

-0.001

-0.001

0.001

(-0.141)

(-0.128)

(0.048)

APPENDIX CHAPTER III

178

Regression Models on Market Value (Hot/Cold) (2) Cold*MB Cold*Vola Cold*Perf Intercept

-0.005

-0.004

0.009

(-0.351)

(-0.260)

(0.534)

0.184

0.268

0.249

(0.250)

(0.361)

(0.323)

-0.012

-0.012

-0.013

(-1.337)

(-1.359)

(-1.188)

16.276

16.315

16.408

(55.963)***

(54.467)***

(44.858)***

R²

0.791

0.788

0.750

F-Statistic

37.250

36.641

27.464

Studienreihe der Stiftung Kreditwirtschaft an der Universität Hohenheim Bände 1 - 11 sind nicht mehr lieferbar. Band 12: Axel Tibor Kümmel: Bewertung von Kreditinstituten nach dem Shareholder Value Ansatz, 1994; 2. Aufl.; 1995. Band

13: Petra Schmidt: Insider Trading. Maßnahmen zur Vermeidung bei US-Banken; 1995.

Band

14: Alexander Grupp: Börseneintritt und Börsenaustritt. Individuelle und institutionelle Interessen; 1995.

Band

15: Heinrich Kerstien: Budgetierung in Kreditinstituten. Operative Ergebnisplanung auf der Basis entscheidungsorientierter Kalkulationsverfahren; 1995.

Band

16: Ulrich Gärtner: Die Kalkulation des Zinspositionserfolgs in Kreditinstituten; 1996.

Band

17: Ute Münstermann: Märkte für Risikokapital im Spannungsfeld von Organisationsfreiheit und Staatsaufsicht; 1996.

Band

18: Ulrike Müller: Going Public im Geschäftsfeld der Banken. Marktbetrachtungen, bankbezogene Anforderungen und Erfolgswirkungen; 1997.

Band

19: Daniel Reith: Innergenossenschaftlicher Wettbewerb im Bankensektor; 1997.

Band

20: Steffen Hörter: Shareholder Value-orientiertes Bank-Controlling; 1998.

Band

21: Philip von Boehm-Bezing: Eigenkapital für nicht börsennotierte Unternehmen durch Finanzintermediäre. Wirtschaftliche Bedeutung und institutionelle Rahmenbedingungen; 1998.

Band

22: Niko J. Kleinmann: Die Ausgestaltung der Ad-hoc-Publizität nach § 15 WpHG. Notwendigkeit einer segmentspezifischen Deregulierung; 1998.

Band

23: Elke Ebert: Startfinanzierung durch Kreditinstitute. Situationsanalyse und Lösungsansätze; 1998.

Band

24: Heinz O. Steinhübel: Die private Computerbörse für mittelständische Unternehmen. Ökonomische Notwendigkeit und rechtliche Zulässigkeit; 1998.

Band

25: Reiner Dietrich: Integrierte Kreditprüfung. Die Integration der computergestützten Kreditprüfung in die Gesamtbanksteuerung; 1998.

Band

26: Stefan Topp: Die Pre-Fusionsphase von Kreditinstituten. Eine Untersuchung der Entscheidungsprozesse und ihrer Strukturen; 1999.

Band

27: Bettina Korn: Vorstandsvergütung mit Aktienoptionen. Sicherung der Anreizkompatibilität als gesellschaftsrechtliche Gestaltungsaufgabe; 2000.

Band

28: Armin Lindtner: Asset Backed Securities – Ein Cash flow-Modell; 2001.

Band

29: Carsten Lausberg: Das Immobilienmarktrisiko deutscher Banken; 2001.

Band

30: Patrik Pohl: Risikobasierte Kapitalanforderungen als Instrument einer marktorientierten Bankenaufsicht – unter besonderer Berücksichtigung der bankaufsichtlichen Behandlung des Kreditrisikos; 2001.

Band

31: Joh. Heinr. von Stein/Friedrich Trautwein: Ausbildungscontrolling an Universitäten. Grundlagen, Implementierung und Perspektiven; 2002.

Band

32: Gaby Kienzler, Christiane Winz: Ausbildungsqualität bei Bankkaufleuten – aus der Sicht von Auszubildenden und Ausbildern, 2002.

Band

33: Joh. Heinr. von Stein, Holger G. Köckritz, Friedrich Trautwein (Hrsg.): E-Banking im Privatkundengeschäft. Eine Analyse strategischer Handlungsfelder, 2002.

Band

34: Antje Erndt, Steffen Metzner: Moderne Instrumente des Immobiliencontrollings. DCF-Bewertung und Kennzahlensysteme im Immobiliencontrolling, 2002.

Band

35: Sven A. Röckle: Schadensdatenbanken als Instrument zur Quantifizierung von Operational Risk in Kreditinstituten, 2002.

Band

36: Frank Kutschera: Kommunales Debt Management als Bankdienstleistung, 2003.

Band

37: Niklas Lach: Marktinformation durch Bankrechnungslegung im Dienste der Bankenaufsicht, 2003.

Band

38: Wigbert Böhm: Investor Relations der Emittenten von Unternehmensanleihen: Notwendigkeit, Nutzen und Konzeption einer gläubigerorientierten Informationspolitik, 2004.

Band

39: Andreas Russ: Kapitalmarktorientiertes Kreditrisikomanagement in der prozessbezogenen Kreditorganisation, 2004.

Band

40: Tim Arndt: Manager of Managers – Verträge. Outsourcing im Rahmen individueller Finanzportfolioverwaltung von Kredit- und Finanzdienstleistungsinstituten, 2004

Band

41: Manuela A. E. Schäfer: Prozessgetriebene multiperspektivische Unternehmenssteuerung: Beispielhafte Betrachtung anhand der deutschen Bausparkassen, 2004.

Band

42: Friedrich Trautwein: Berufliche Handlungskompetenz als Studienziel: Bedeutung, Einflussfaktoren und Förderungsmöglichkeiten beim betriebswirtschaftlichen Studium an Universitäten unter besonderer Berücksichtigung der Bankwirtschaft, 2004.

Band

43: Ekkehardt Anton Bauer: Theorie der staatlichen Venture Capital-Politik. Begründungsansätze, Wirkungen und Effizienz der staatlichen Subventionierung von Venture Capital, 2006.

Band

44: Ralf Kürten: Regionale Finanzplätze in Deutschland, 2006.

Band

45: Tatiana Glaser: Privatimmobilienfinanzierung in Russland und Möglichkeiten der Übertragung des deutschen Bausparsystems auf die Russische Föderation anhand des Beispiels von Sankt Petersburg, 2006.

Band

46: Elisabeth Doris Markel: Qualitative Bankenaufsicht. Auswirkungen auf die Bankunternehmungsführung, 2010.

Band

47: Matthias Johannsen: Stock Price Reaction to Earnings Information, 2010.

Band

48: Susanna Holzschneider: Valuation and Underpricing of Initial Public Offerings, 2011.