1
The Influence of Venture Capital
and Institutional Environment Quality
on IPO Underpricing
Does a high quality institutional environment work as a complement to
venture capital influence to reduce IPO underpricing?
Master thesis Rijksuniversiteit Groningen Business Administration – Finance
Name: P.W.L. Bröcker Student number: 1481320 Supervisor: C.L.M. Hermes
Date: July 14th , 2011
Abstract:
2
Table of contents
1. Introduction 3
2. Literature Review 4
2.1 Venture capital influence on underpricing 4
2.1.1 Venture capital and method to exit 5
2.1.2 Venture capital and underpricing 5
2.1.3 Venture capital and underpricing studies 7 2.2 Institutional environment and underpricing 8 2.3 Institutional environment quality and venture capital financing 10 2.3.1 Institutional environment and financing obstacles 10 2.3.2 Institutional development and venture capital behaviour 11
2.4 Hypotheses development 12
3. Methodology 14
3.1 Dependent variable 14
3.2 Independent variables 14
3.2.1 Venture capital variable 14
3.2.2 Worldwide Governance Indicators 14
3.2.3 Control variables 16
3.3 Regression models 18
4. Data description 20
4.1 Data collection process and data source 20
4.2 Descriptive statistics 20
4.2.1 Full data sample 21
4.2.2 VC backed and non-VC backed data sample 23 4.3 Indicators of institutional environment quality 25
5. Results 28
5.1 Venture capital influence on underpricing 28 5.2 Institutional environment and underpricing 29
5.3 Interaction model 32
5.3.1 Robustness check: VC vs. non VC backed sample 32 5.3.2 Venture capital influence and law system origin 35
6. Discussion 38
7. Conclusion and recommendations 39
3
1. Introduction
It is often seen that shares issued in an Initial Public Offering (IPO) show a high initial return on the first trading day, also called IPO underpricing. Underpricing can be defined as the difference between the offer price and the closing price at the end of the first trading day (Daily et al., 2003). When the closing price is higher than the offer price, it is said that ‘there is a lot of money left on the table’. The selling firm could have raised more money by reducing the level of underpricing. In the previous decades, a lot of research has been done on IPO underpricing providing us with different theories to explain underpricing. According to Ljungqvist (2007) the reasons for underpricing can be divided in four categories, namely information asymmetry reasons, institutional reasons, ownership and control reasons and behavioral reasons. The focus in this study is on the information asymmetry reasons, since the level of information asymmetry can be influenced by both the venture capital firm and the institutional environment quality in a country. During the process of an IPO, there are three main stakeholders involved. These are the selling firm, the investors willing to buy and the underwriting bank. Information asymmetry occurs when one of these stakeholders knows more than the others.
In this study, the relationship between venture capital influence, institutional environment quality and underpricing is investigated and discussed. The existing literature relates venture capital influence and institutional environment quality to underpricing separately. In most of the existing literature, information asymmetry is a key reason to explain these relationships. This study extends current literature by investigating the interaction effect between venture capital influence and institutional environment quality to reduce the level of information asymmetry. Institutional environment quality is measured using six indicators, explained in more detail in the methodology section. To investigate the interaction effect between venture capital influence and institutional environment, the six indicators are interacted with the venture capital variable. As a robustness check, the relationship between the six indicators and underpricing is tested for the venture capital backed and non venture capital backed sample separately.
4
2. Literature Review
In this study, several instruments are discussed that can reduce the level of information asymmetry and thus the level of underpricing. In the first part of this section, the relationship between venture capital influence, information asymmetry and underpricing is discussed. Two hypotheses often mentioned in the literature are the certification hypothesis and the grandstanding hypothesis (Dissanaike and Amel-Zadeh, 2007). This study focuses on the certification hypothesis, because the grandstanding hypothesis explains underpricing by looking at the age of a venture capital firm, while the certification hypothesis explains venture capital influence from an information asymmetry perspective. In the second part of this section, the relationship between the institutional environment, information asymmetry and underpricing is discussed. A high quality institutional environment can reduce the level of information asymmetry and thus the level of underpricing. The contribution of this study is to investigate the interaction between venture capital influence and institutional environment quality. The question of importance is if they work as a substitute or as a complement to reduce the level of underpricing. The different relationships relevant in this study are summarized in figure 1.
Figure 1. Research model
2.1 Venture capital influence on underpricing
According to Gladstone and Gladstone (2002), a venture capital investment is characterized by high risk. It is of less importance how the investment is structured, both an equity investment and a high risk loan can be seen as a venture capital investment. The way in which venture capital differs from financing firms is the level of management assistance. Venture
5
capital firms are actively involved in the business they invest in and they therefore can be described as a business partner for the firm they invest in. Another characteristic of venture capital given by Gladstone and Gladstone (2002) is the fact that venture capital firms invest in new firms and business ideas. Venture capital money is often invested in start-ups or first, second or third round financing. It is therefore not surprising that venture capital investment can result in high returns or great losses. In the first part of this section, different methods for a venture capital investment to exit will be presented. In the second part, the role of venture capital in the IPO process will be discussed, explaining the role of venture capital in underpricing from the certification hypothesis. In the last part of this section, empirical results on the role of venture capital in underpricing will be presented.2.1.1 Venture capital and methods to exit
In their study on optimal exit decisions in venture capital finance, Bascha and Walz (2001) state that one of the main characteristics of venture capital finance is the limited investment period. This limited investment period is an instrument for the investors in a venture capital fund to control the managers of the fund to ensure the distribution of their investment returns. Due to this limited investment period, the venture capital firm has to exit the investment at some time to ensure a return for the investors involved. Bascha and Walz (2001) mention five different methods to terminate the financial relationship between the venture capital firm and the firm they invested in. These methods are liquidation, share repurchases by the founder, selling shares to institutional investors, selling the firm as a whole to another firm and an IPO. For this study, the relevant method to exit is selling the firm in an IPO.
2.1.2 Venture capital and underpricing
According to Dissanaike and Amel-Zadeh (2007) the two most prominent explanations for the variation in underpricing with venture capital involvement are the certification and the grandstanding hypothesis. The certification hypothesis suggests that venture capital involvement reduces underpricing. Bascha and Walz (2001) state that venture capitalist play a active governance role, using a wide range of instruments including direct monitoring, to reduce information asymmetries between the issuing firm and the investors.
6
and the potential investors. This certification role of venture capital should reduce IPO underpricing. In their study Megginson and Weiss (1991) discuss three criteria for a third party to be reliable as a monitor. The first one is the reputation of the venture capital firm. They state that the third party must have a reputational capital at stake that could be damaged if they would certify an offering as fairly priced while in fact it was over-valued. Venture capital firms have a strong incentive to build a reliable reputation in order to maintain access to the IPO market on good terms. It is also important for the venture capital firm to build strong relationships with pension fund managers and other institutional investors, since they are important as investors when a venture capital firm brings a firm public. Next to that, the value of the venture capital firm’s reputational capital must exceed the largest possible benefits that could be obtained by certifying falsely. According to Megginson and Weiss (1991) venture capital firms invest in their reputational capital in order to remain competitive in the venture capital industry and in the capital markets. A venture capital firm can profit from a false certification when they sell all their shares in the IPO at a high price. In their study, Cumming and MacIntosh (2003) distinguish between a full exit and a partial exit. They define a full exit as a sale of all of the venture capitalists holdings within one year of the IPO. A partial exit involves only a sale of part of the holdings within that same period. The disadvantage of a full exit is the fact that it could be interpreted as a negative signal to the investors. When a venture capital firm retains part of the shares, they signal that the shares offered are good shares. Therefore, retaining some of their shares after the IPO can therefore be used as a mechanism for credible certification. Finally, it must be costly for the issuing firm to purchase the services of the venture capital firm. Megginson and Weiss (1991) state that the services venture capital firms provide to the firm are very costly and hard to obtain. Firms therefore hand over large holdings of shares to venture capital firms in exchange for relatively small cash investments. The results of Barry et al. (1990) and Megginson and Weiss (1991) support the certification hypothesis, as they found lower underpricing for IPOs backed by venture capital. Another study that explains the role of venture capital from the certification hypothesis is the study of Wang et al. (2003), who studied the influence of venture capital on IPO underpricing in Singapore for the period 1987-2003. Their results showed significant lower underpricing for venture capital backed IPOs and significant lower underpricing for older venture capital firms, supporting the first criteria of reliable third party certification.7
2.1.3 Venture Capital and underpricing studies
8
Table 1. Overview of underpricing studies for VC-backed IPOs
Year Authors Period Region Underpricing
1990 Barry et al. 1978-1987 USA Lower
1991 Megginson, Weiss 1983-1987 USA Lower
1999 Espeblaub et al. 1992-1995 UK No sign. difference
2000 Hamao et al. 1989-1995 Japan Higher
2001 Francis, Hasan 1990-1993 USA Higher
2002 Bradley, Jordan 1990-1999 USA Higher
2002 Barnes et al. 1992-1999 UK No sign. difference
2003 Frankze 1997-2002 Germany Higher
2003 Wang et al. 1987-2001 Singapore Lower
2004 Bessler, Kuhrt 1998-2001 Germany No sign. difference
2004 Lee, Wahal 1980-2000 USA Higher
2004 Brau et al. 1990-1996 USA No sign. difference
2005 Dolvin 1986-2000 USA Higher
2005 Jelic et al. 1964-1997 UK No sign. difference
2007 Klaassen, Von Eije 1994-2005 Netherlands No sign. difference
2007 Tykova, Walz 1997-2002 Germany No sign. difference
2009 Goergen et al. 1996-2000 France/Germany No sign. difference
2.2 Institutional environment and underpricing
In this part, the relationship between the institutional environment, information asymmetry and underpricing is discussed.
9
Capital market governance
Daouk et al. (2005) investigated the effect of capital market governance on different dimensions of market performance, including pricing efficiencies. The capital market governance index used in their study captures three dimensions: the degree of earnings opacity, the enforcement of insiders laws and the effect of removing short-selling restrictions. They associate higher pricing inefficiencies with higher levels of underpricing and state that improvements in capital market governance helps firms to access capital markets on betters terms, leading to more efficient market pricing, correspondingly lower costs of capital and lower levels of underpricing. Their results confirm their hypothesis, an increase in their capital market governance index of one standard deviation results in a decrease of 5.2 percent in the level of underpricing.
Government regulation
10
Legal framework and shareholder protection
Engelen and Van Essen (2010) examined the impact of country-specific institutional characteristics on IPO underpricing and found that the quality of a country’s legal framework reduces the level of underpricing significantly. In their study, they discuss two ways in which the legal framework of a country can increase the ex ante uncertainty about the value of a firm. At first, weaker shareholder protection leads to more uncertainty with respect to post IPO management decisions that may negatively affect the value of a firm. Secondly, weaker shareholder protection leads to more uncertainty about the future distribution of the realized firm value among the different shareholders. When shareholder protection is weaker, managers or majority shareholders have the opportunity to extract private benefits out of the firm at the expense of the minority shareholders. This ex post expropriation increases the ex ante uncertainty when a firm goes public. Due to this higher uncertainty regarding post IPO management decisions and the expropriation risk, IPOs have to be more underpriced on average in countries with weaker shareholder protection to induce investors to subscribe to the offer. Following the results of Engelen and Van Essen (2010), higher legal framework quality reduces the ex ante uncertainty and the level of underpricing.
2.3 Institutional environment quality and venture capital financing
In order to explain and predict the relationship between institutional environment quality and venture capital influence, the literature concerning institutional environment and financing of firms is discussed in this part. In the first part, the relationship between institutional environment and financing obstacles is discussed. In the second part, the relationship between institutional development and venture capital behaviour is discussed.
2.3.1 Institutional environment and financing obstacles
11
underlying institutions are the most important country characteristic to explain the financing obstacles of firms.Demirgüç-Kunt and Maksimovic (1998) investigated how differences in legal and financial systems affect the access to external financing for firms. They state that market imperfections, including information asymmetries between corporate insiders and investors, constrains firms in their ability to obtain external capital. These imperfections can be reduced by an effective legal and financial system. Their results show that firms in countries that have better developed stock markets and more effective legal systems are able to obtain external funds and grow faster. These results are in line with the results of Beck et al. (2003).
Overall it can be concluded that institutional environment quality helps to reduce market imperfections and lowers the financing obstacles for firms, which results in increased access to external financing, including venture capital financing.
2.3.2 Institutional development and venture capital behaviour
12
banks are not allowed to own such stocks and the ownership structures in these countries are wider dispersed. Bruton et al. (2005) state that in bank centered countries, many venture capital firms are partially or fully owned by banks, resulting in less risk taking en investing more in later-stage firms. In stock market centered countries, there is more early-stage investing and venture capital firms can obtain higher returns and take more risk.2.4 Hypotheses development
This paper contributes to the literature on IPO underpricing by researching different factors to resolve the problem of information asymmetry, thereby reducing the level of underpricing. Following the certification hypothesis, venture capital influence is associated with lower levels of underpricing (Megginson and Weiss, 1991). For this paper a negative relationship is expected between IPOs backed by a venture capital firm and the level of underpricing. Hence, the following hypothesis is formulated:
Hypothesis 1: Venture capital backed IPOs are less underpriced than non venture capital
backed IPOs
Following the different theories to explain the role of the institutional environment to resolve the information asymmetry problem, a negative relationship is expected between country-level institutional environment quality and the country-level of underpricing. Hence, the following hypothesis if formulated:
Hypothesis 2: IPOs in countries with higher institutional environment quality are less
underpriced than IPOs in countries with lower institutional environment quality
13
it is unclear from the existing literature if venture capital influence can be seen as a substitute for a low quality institutional environment. It is therefore expected for this study that institutional environment quality works as a complement to venture capital influence, where venture capital influence is more effective to reduce the level of IPO underpricing in a high quality institutional environment. Hence, the following hypotheses are formulated:Hypothesis 3: The negative relationship between venture capital influence and IPO
14
3. Methodology
3.1 Dependent variable
The dependent variable in this study is the level of underpricing. Underpricing can be defined as the difference between the offer price and the closing price at the end of the first trading day (Daily et al., 2003). A firm is underpriced if the percentage of underpricing is positive. The level is underpricing is calculated with the following equation:
Underpricing = ( Closing price day 1 – Offer price ) / Offer Price
3.2 Independent variables
In this part, the independent variables used in this study are introduced. The venture capital variable is used to test hypothesis 1 and the Worldwide Governance Indicators are used to test hypothesis 2. The venture capital dummy is interacted with the Worldwide Governance Indicators to test hypothesis 3. The control variables related to underpricing are added to all regression models.
3.2.1 Venture capital variable
Following the studies mentioned in table 1, a dummy is introduced for IPOs backed by a venture capital firm.
Venture capital dummy
Since this study focuses on information asymmetry as a reason for underpricing, the certification hypothesis is followed and a negative relationship is expected. Issues will be assigned a value of 1 if backed by a venture capital firms and zero otherwise.
3.2.2 Worldwide Governance Indicators
15
quality tend to focus on legal, regulation and accounting indicators (Hopp and Dreher, 2007) or capital market governance (Daouk et al., 2005), where the WGI measures include measures on voice and accountability and political stability. In their latest paper, Kaufman, Kraay and Mastruzzi (2010) define governance as:“The traditions and institutions by which authority in a country is exercised. This includes (a) the process by which governments are selected, monitored and replaced; (b) the capacity of the government to effectively formulate and implement sound policies; and (c) the respect of citizens and the state for the institutions that govern economic and social interactions among them.”
The WGI measures six indicators and covers over 200 countries. These measures are based on hundreds of individual variables from a wide variety of data sources. For this study, all six indicators are used for the countries included.
The six indicators of governance included for this study are defined, following the definition of Kaufman, Kraay and Mastruzzi (2010):
Voice and Accountability (VA)
This variable measures the extent to which a country's citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media. Freedom of expression, association and media can have a positive effect on the quality and availability of information. Buckley, Duer and Mendel (2008) state that independent and free media contribute to good governance and accountability by providing quality information, including investigative journalism and informational programming. A high score on VA is therefore associated with lower levels of underpricing.
Political Stability and Absence of Violence/Terrorism (PV)
16
Government Effectiveness (GE)
Measures the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies. Following Barth et al. (2004), a more effective government is expected to correct market failure such as information asymmetry and to reduce the level of underpricing.
Regulatory Quality (RQ)
Indicates the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development. Following the helping-hand approach of Barth et al. (2004), higher regulatory quality is expected to reduce the level of information asymmetry and the level of underpricing.
Rule of Law (RL)
This variable measures the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence. Because a higher quality legal framework reduces the ex ante uncertainty (Engelen and Van Essen, 2010), a negative relationship is expected with the level of underpricing.
Control of Corruption (CC)
This indicator captures perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as capture of the state by elites and private interests. It is expected that better control of corruption results in less uncertainty and thus lower level of underpricing (Engelen and Van Essen, 2010).
3.2.3 Control variables
For this study, different control variables used in earlier studies are included that are related to the dependent variable, underpricing. In this part, the control variables and their expected relationship with underpricing are discussed.
Deal size
17
Retention
According to Barry (1989) there is a relationship between the number of shares offered by former shareholders and the level of underpricing, where former shareholders are defined as the entrepreneurs and venture capitalists selling their stake in a firm. Barry (1989) states that the former shareholders have more financial incentives to reduce the level of underpricing when selling a higher proportion of their shares and they will care less about the level of underpricing when only selling one share. Therefore, former shareholders will try to influence the offer price of an issue when their wealth can be negatively affected by the price setting. Barry (1989) assumes a positive relationship between the level of retention and the level of underpricing. Therefore, a positive relationship is expected between the level of retention and underpricing for this study.
Country dummy
To test the different levels of underpricing in Europe, country dummies will be added to the regression of this study to check for differences between the levels of underpricing in the countries included. A country dummy will be introduced for all countries included in the full data sample except for Austria, since one country dummy has to be excluded. Since earlier results on underpricing in Europe and difference between countries are scarce or not significant, no expectation is set for this study.
IPO year
Klaassen and Von Eije (2007) state that more recent literature finds greater underpricing for VC backed IPOs. Following Klaassen and Von Eije (2007), a time variable for the IPO year is included, with an expected positive relationship. The first year in the period of investigation (2000) will be assigned a value of 1, the last period (2010) a value of 11.
Technology dummy
18
Hot market dummy
In their research on IPOs in hot and cold markets, Helwege and Liang (2004) described several characteristics of hot and cold markets. They describe hot IPO markets as having a high volume of issues, severe underpricing, frequent oversubscription of issues and concentrations in particular industries. For this study, the different years in the investigation period will be assigned a value of 1 if it is considered to be a hot market and zero otherwise. To decide which years can be considered hot, the first description of Helwege and Liang (2004) will be followed, the volume of issues. Therefore, a year is considered as a hot market when the numbers of IPOs in a particular year are above mean during the investigation period. For this study, the years 2000, 2005, 2006 and 2007 are considered to be hot. Following the link between hot markets and higher levels of underpricing discussed by Helwege and Liang (2004), a positive relationship between hot markets and underpricing is expected.
3.3 Regression models
In order to test the relationship between the dependent variable introduced in part 3.1 and the independent variables introduced in part 3.2, an Ordinary Least Square (OLS) regression is used as described by Brooks (2008):
In this model γ is the dependent variable, the level of underpricing. The constant in this model is α, which is expected to be positive for this study assuming that there will always be underpricing on average. The betas represent the independent variables, where the results of the OLS show if they are significant. The ε is the error term, which has a zero mean and has to be normally distributed.
To empirically test the influence of the independent variables, including the VC backed dummy, on the level of underpricing, the following regression will be tested:
Regression 1: Underpricing = β0 + β1*Deal size+ β2*Retention+ β3*Country+ β4*IPO year
19
To empirically test the influence of the independent variables, including the indicators of institutional environment quality, on the level of underpricing, the following regression will be tested:Regression 2: Underpricing = β0 + β1*Deal size+ β2*Retention+ β3*Country + β4*IPO year
+ β5*Technology + β6*Hot Market + β7 *VA + β8 *PV + β9 *GE + β10 *RQ + β11 *RL
+ β12*CC
To empirically test the interaction effect between venture capital influence and the indicators of institutional environment quality on the level of underpricing, the following regression will be tested:
Regression 3: Underpricing = β0 + β1*Deal size+ β2*Retention+ β3*Country + β4*IPO year
+ β5*Technology + β6*Hot Market + β7 *VC backed + β8 *VA + β9 *PV + β10 *GE + β11 *RQ
+ β12*RL + β13*CC + β14 *VC*VA + β15*VC*PV + β16*VC*GE + β17*VC*RQ
+
β18*VC*RL20
4. Data description
This section gives an overview of the data used for this study. In the first part, the data collection process and the data sources will be presented. In the second part, the descriptive statistics of the total sample and the sub samples will be presented. In the third part, the data on the institutional environment indicators to explain cross-country variations in underpricing will be presented.
4.1 Data collection process and data source
In order to investigate the relationship between the different independent variables in this study and the level of underpricing, a set of data is constructed of European IPOs for the period 2000-2010. The data is collected from the database of Dealogic. Dealogic is a platform used by investment banks and capital market professionals to collect data, with a particular focus on equity capital market data. Dealogic gives the option to specifically search for IPOs in the period and geographical area chosen and to select which information about this product has to be included. A first scan of European IPOs resulted in database of 3,625 IPOs, including 217 venture capital backed IPOs. In order to make some reliable statistical inferences and to make a comparison between venture capital influences between countries, IPOs included in the final dataset had to meet some criteria. At first, IPOs in countries where no venture capital backed IPOs took place were excluded in order to compare the difference between venture capital and non venture capital backed IPOs on a country-level. Next to that, IPOs with a deal size of less than 10 million euro were excluded, because they are on average highly speculative (Beatty and Ritter, 1986). After looking up all the variables needed for this study and excluding observations where one or more variables were missing, a final dataset of 1,751 IPOs was left, including 168 venture capital backed IPOs. From this final dataset, outliers in the level of underpricing were excluded, with 1% of the minimum and maximum values for underpricing, resulting in 1,715 IPOs including 164 venture capital backed. Data found in other currencies than the euro were converted, using the exchange rate with the euro at the day of the offer. Exchange rates were collected from the Bloomberg financial database.
4.2 Descriptive statistics
21
4.2.1 Full data sample
The descriptive statistics of the full sample of IPOs can be found in table 2. A firm going public is underpriced if the underpricing variable is positive.
Table 2. Descriptive statistics full sample (N=1,715)
Variable Minimum Maximum Mean St. Dev
Underpricing -22.00% 165.00% 10.04% 20.22%
Deal size (EUR m) 10.00 9,059.16 191.65 504.73
Retention 0.00% 100.00% 57.49% 24.09%
IPO year 1.00 11.00 5.75 2.89
Technology 0.00 1.00 0.17 0.38
Hot market 0.00 1.00 0.75 0.43
VC backed 0.00 1.00 0.10 0.29
Table 2 shows an average underpricing for the full sample of 10.04%. This result confirms the existence of underpricing in European IPOs. A wide range of underpricing is found, with a minimum of -22.00% and a maximum of 165.00%, found for an IPO in the technological sector in 2000, when the dot-com bubble was at its peak. The average deal value of the full sample is € 191.65 million, with a minimum of € 10.0 million (the minimum deal size chosen) and a maximum of € 9,059 million. The average amount of retention is 57.49%.
Table 3 presents a correlation matrix for the control variables and the venture capital dummy. When using the OLS estimation method, it is important to check if the independent variables correlate with one another. Problems can occur when independent variables are highly correlated with each other, also known as multicollinearity (Brooks, 2008). It follows from table 3 that multicollinearity is not a problem for variables included in the correlation matrix, with the highest correlation of -0.311 between IPO year and the technology dummy.
Table 3. Correlation matrix control variables and venture capital backed dummy
Deal size (log) Retention IPO year Technology Hot market VC backed
Deal size (log) 1.000 -0.200 0.060 -0.151 -0.007 0.019
Retention -0.200 1.000 0.201 0.134 0.004 -0.017
IPO year 0.060 -0.201 1.000 -0.311 -0.106 0.020
Technology -0.151 0.134 -0.311 1.000 0.053 0.024
Hot market -0.007 0.004 -0.106 0.053 1.000 -0.053
VC backed 0.019 -0.017 0.020 0.024 -0.053 1.000
22
Table 4. Distribution of all IPOs per country (N=1,715)
Country Number of IPOs % of sample Underpricing
Austria 22 1.28% 4.49% Belgium 25 1.46% 8.83% Denmark 26 1.52% 12.47% Finland 13 0.76% 8.98% France 156 9.10% 7.48% Germany 223 13.00% 17.92% Italy 139 8.10% 5.92% Netherlands 29 1.69% 9.41% Norway 68 3.97% 3.32% Poland 108 6.30% 15.28% Portugal 7 0.41% 12.42% Spain 36 2.10% 6.88% Sweden 43 2.51% 4.25% Switzerland 46 2.68% 11.51% United Kingdom 774 45.13% 9.39% Total 1,715 100.00% 10.04%
The United Kingdom accounts for 45 percent of all European IPOs in the chosen period, followed by Germany with 13.00% and France with 9.10%. From table 3, the first conclusions can be drawn with respect to the level of underpricing per country. In Germany, the level of underpricing is much higher than the European average, with an average underpricing of 17.92%. This high level of underpricing can be explained by the large number of IPOs in Germany in 2000 in the technology sector. France is below the European average with an average underpricing of 7.48%, while the United Kingdom is close to the average with an average underpricing of 9.39%. The minimum average underpricing is found for Norway with an average of 3.32%, followed by another Scandinavian country Sweden with 4.25%. These low average levels of underpricing can be a result of the higher institutional environment quality in the Scandinavian countries, measured by the six indicators presented in part 4.3.
23
Table 5. Distribution of all IPOs per year (N=1,715)
Year Number of IPOs % of sample Underpricing
2000 319 18.60% 21.49% 2001 67 3.91% 6.32% 2002 38 2.22% 4.57% 2003 27 1.57% 6.85% 2004 137 7.99% 8.37% 2005 264 15.39% 7.96% 2006 359 20.93% 7.39% 2007 341 19.88% 8.09% 2008 52 3.03% 6.21% 2009 20 1.17% 3.90% 2010 91 5.31% 5.67% Total 1,715 100.00% 10.04%
4.2.2 VC backed and non-VC backed data sample
Table 6 compares the descriptive statistics for the VC backed data sample with the non-VC backed sample.
Table 6. Descriptive statistics VC (N=164) and non-VC backed sample (N=1,551)
Variable Minimum Maximum Mean St. Dev
VC backed sample
Underpricing -20.00% 82.00% 8.21% 13.82%
Deal size (EUR m) 10.72 1,796.07 144.70 231.53
Retention 0.00% 86.00% 56.26% 20.04%
IPO year 1.00 11.00 5.92 1.82
Technology 0.00 1.00 0.20 0.40
Hot market 0.00 1.00 0.68 0.40
Non-VC backed sample
Underpricing -22.00% 165.00% 10.23% 20.78%
Deal size (EUR m) 10.00 9,059.16 196.69 525.17
Retention 0.00% 100.00% 57.62% 24.48%
IPO year 1.00 11.00 5.73 2.98
Technology 0.00 1.00 0.17 0.38
Hot market 0.00 1.00 0.76 0.43
24
minimum of -20.00%. The average deal value of VC backed IPOs is € 144.70 million, compared with an average deal size of the non-VC backed sample of € 196.69 million. This difference in average deal size is not significant. The amount of shares in retention for VC backed IPOs is lower compared with the non-VC backed sample, with an average of 56.26% compared to 57.62%, although this difference is not significant.Table 7 presents an overview of the distribution of VC backed IPOs per country. Similar to the full data sample, United Kingdom accounts for almost half of the IPOs with 43.29%, followed by Germany with 14.63% and France with 8.54%. The most important facts from this table are the differences in average underpricing per country with the full data sample. For example, Germany has an average underpricing for the full data sample of 17.92% but only an average underpricing of 9.44% for the VC backed data sample. This result can be explained by the small number of VC backed IPOs in 2000, when the level of underpricing in Germany was at its peak. The contrary effect is found for the Netherlands, which has a much higher average underpricing for the VC backed data sample, namely 22.92%, although based on only one observation. The average underpricing for the United Kingdom is almost similar for both samples (9.39% and 9.98%).
Table 7. Distribution of VC backed IPOs per country (N=164)
Country Number of IPOs % of sample Underpricing
Austria 2 1.22% 3.68% Belgium 2 1.22% 3.56% Denmark 4 2.44% 6.00% Finland 1 0.61% 0.21% France 14 8.54% 7.81% Germany 24 14.63% 9.44% Italy 6 3.66% 5.81% Netherlands 1 0.61% 22.92% Norway 11 6.71% -0.58% Poland 4 2.44% 9.56% Portugal 1 0.61% -1.38% Spain 3 1.83% 17.43% Sweden 10 6.10% 4.43% Switzerland 10 6.10% 7.79% United Kingdom 71 43.29% 9.98% Total 164 100.00% 8.21%
25
Table 8. Distribution of VC backed IPOs per year (N=164)
Year Number of IPOs % of sample Underpricing
2000 5 3.05% 38.00% 2001 5 3.05% 1.14% 2002 10 6.10% 7.09% 2003 4 2.44% 5.94% 2004 31 18.90% 6.40% 2005 41 25.00% 9.45% 2006 43 26.22% 6.41% 2007 22 13.41% 8.51% 2008 0 0.00% 0.00% 2009 1 0.61% 8.37% 2010 2 1.22% 3.69% Total 164 100.00% 8.21%
For this sample more than half of the IPOs are found for the years 2005, 2006 and 2007. For the VC backed data sample, 2000 accounts only for 3.05% of the IPOs, compared to 18.60% for the full data sample. Again underpricing is at its peak in 2000, with an average underpricing of 38.00%.
4.3 Indicators of institutional environment quality
The scores per country for each of the six indicators of the governance index of Kaufman, Kraay and Mastruzzi (2010) can be found in table 9. For each country, the minimum and maximum score per indicator for the 10 years included in this study are presented. Higher scores represent higher institutional environment quality for each indicator.
26
Table 9. Summary governance indicator scores per country
VA PV GE
Country Min. Max. Min. Max. Min. Max.
Austria 1.301 1.494 1.062 1.279 1.617 1.952 Belgium 1.359 1.476 0.660 1.166 1.233 1.921 Denmark 1.489 1.827 1.035 1.360 1.933 2.237 Finland 1.489 1.814 1.357 1.577 1.907 2.191 France 1.080 1.471 0.525 0.946 1.433 1.821 Germany 1.371 1.519 0.697 1.221 1.409 1.927 Italy 0.915 1.148 0.500 0.899 0.333 0.945 Netherlands 1.483 1.730 0.822 1.428 1.656 2.043 Norway 1.504 1.726 1.151 1.455 1.726 2.084 Poland 0.804 1.093 0.281 0.907 0.415 0.650 Portugal 1.211 1.477 0.783 1.279 0.846 1.227 Spain 1.055 1.304 -0.180 0.681 0.913 1.881 Sweden 1.491 1.758 1.100 1.353 1.837 2.059 Switzerland 1.425 1.714 1.142 1.393 1.820 2.139 United Kingdom 1.275 1.601 0.304 0.979 1.476 1.890 RQ RL CC
Country Min. Max. Min. Max. Min. Max.
Austria 1.475 1.649 1.758 1.928 1.062 1.952 Belgium 1.180 1.365 1.199 1.371 0.660 1.921 Denmark 1.644 1.855 1.773 1.964 1.035 2.237 Finland 1.511 1.866 1.830 1.937 1.357 2.191 France 0.979 1.255 1.274 1.444 0.525 1.821 Germany 1.431 1.590 1.574 1.698 0.697 1.927 Italy 0.869 1.088 0.313 0.874 0.333 1.148 Netherlands 1.650 2.012 1.672 1.781 0.822 2.043 Norway 1.042 1.451 1.783 1.948 1.151 2.084 Poland 0.646 0.934 0.391 0.732 0.281 1.093 Portugal 0.979 1.295 0.999 1.300 0.783 1.477 Spain 1.147 1.352 1.059 1.374 -0.180 1.881 Sweden 1.452 1.661 1.726 1.927 1.100 2.059 Switzerland 1.492 1.746 1.751 1.878 1.142 2.139 United Kingdom 1.537 1.820 1.517 1.706 0.304 1.890
Table 10. Correlation matrix governance indicators
27
With respect to the first indicator measuring the capacity of the government to effectively formulate and implement sound policies, Government Effectiveness, high scores are found for the Scandinavian countries, the Netherlands and Switzerland. Less effective governments are found for Italy, Poland and Portugal, with the lowest minimum and maximum scores. With respect to the second indicator, Regulatory Quality, the highest maximum score is found for the Netherlands, indicating a well developed regulatory environment. Again, the Scandinavian countries score well, although Norway and Sweden score relatively low on this indicator compared to the Government Effectiveness indicator. In line with Barth et al. (2004) high scores on these two indicators are associated with better government regulation and effectiveness, helping to correct market failures such as information asymmetries.The last two indicators measure the respect of citizens and the state for the institutions that govern economic and social interactions among them, Rule of Law and Control of Corruption. Scandinavian countries show the highest scores for the Rule of Law indicator, while the South European countries and Poland show the lowest scores, meaning that citizens have less confidence in the rules of the society. No surprising results are found for the Control of Corruption indicator, where Italy and Poland show the lowest maximum scores, indicating less control of corruption in these countries. Following the conclusion of Engelen and Van Essen (2010), high scores on these indicators are associated with less ex ante uncertainty about the current and future value of the firm.
28
5. Results
In the first part, the results on the relationship between venture capital influence and underpricing are presented. In the second part, the results on the relationship between institutional environment and underpricing are presented and in the last part the interaction results between venture capital influence and institutional environment are presented.
5.1 Venture capital influence on underpricing
To test the relationship between venture capital influence and underpricing, regression 1 was tested. The expected signs and results are found in table 11.
Table 11. Results on regression 1: venture capital influence and underpricing
Variable Expected sign Coefficient prob.
Deal size (log) - -0.023 0.009 ***
Retention + 0.042 0.046 ** IPO year + -0.013 0.000 *** Technology + 0.049 0.000 *** Hot market + 0.034 0.002 *** VC backed - -0.013 0.410 Belgium 0.039 0.492 Denmark 0.097 0.082 * Finland 0.017 0.803 France -0.001 0.988 Germany 0.100 0.021 ** Italy -0.003 0.937 Netherlands 0.066 0.225 Norway 0.004 0.933 Poland 0.126 0.006 *** Portugal 0.097 0.249 Spain 0.049 0.347 Sweden -0.002 0.963 Switzerland 0.063 0.208 UK 0.054 0.196 Constant 0.106 0.032 ** Adj. R Square 0.097
29
Table 11 shows that the relationship between venture capital influence and underpricing meets the expectation with a negative sign. Although the descriptive statistics confirm the certification hypothesis (8.21% average underpricing for VC backed IPOs compared to 10.23% for non VC backed IPOs), the relationship between venture capital influence and underpricing is not significant.The coefficients of the control variables deal size, retention, technology dummy and hot market meet the expected relationship with underpricing. Deal size shows a significant negative relationship on a 1% level. Retention shows a significant positive relationship on a 5% level and the technology and hot market variable both show a significant positive relationship on a 1% level. The variable concerning the year of the IPO (IPO year) shows a result contrary to the expectations, with a negative significant relationship on a 1% level. This result can be explained by the average underpricing in 2000, with 319 IPOs with an average underpricing of 21.49%. The country dummies show three significant relationships. Denmark shows a significant positive relationship on a 10% level, Germany a significant positive relationship on a 5% level and Poland shows a significant positive relationship on a 1% level.
5.2 Institutional environment and underpricing
The results for the relationship of the different indicators of institutional environment and underpricing are presented in table 12. The results are mixed, since three indicators show an expected negative relationship and three indicators show a positive relationship. A negative relationship is found for Voice and Accountability (VA), Rule of Law (RL) and Control of Corruption (CC). The result for the RL variable is the only significant results. This result supports the theory of Engelen and Van Essen (2010), who state that higher legal framework quality reduces ex ante uncertainty and the level of underpricing. A positive relationship is found for the indicators Political Stability and Absence of Violence/Terrorism (PV), Government Effectiveness (GE) and Regulatory Quality (RQ), where the results for PV and RQ are close to significant on a 10% level. The result of RQ contradicts the helping-hand approach of Barth et al. (2004), who predict a negative relationship between regulation and underpricing.
30
Table 12. Results on regression 2: institutional environment and underpricingModel 1 Model 2 Model 3
Variable Coeff. prob. Coeff. prob. Coeff. prob.
Deal size (log) -0.023 0.009 *** -0.024 0.007 *** -0.024 0.008 ***
Retention 0.042 0.046 ** 0.040 0.059 * 0.042 0.046 ** IPO year -0.013 0.000 *** -0.010 0.000 *** -0.011 0.000 *** Technology 0.049 0.000 *** 0.048 0.000 *** 0.049 0.000 *** Hot market 0.032 0.003 *** 0.033 0.002 *** 0.037 0.001 *** VA -0.066 0.244 - - - -PV - - 0.057 0.104 - -GE - - - - 0.034 0.524 RQ - - - -RL - - - -CC - - - -Belgium 0.041 0.466 0.054 0.343 0.044 0.443 Denmark 0.110 0.054 * 0.095 0.087 * 0.082 0.177 Finland 0.034 0.621 0.002 0.978 0.008 0.913 France -0.008 0.856 0.028 0.559 0.007 0.882 Germany 0.102 0.018 ** 0.105 0.015 ** 0.102 0.019 ** Italy -0.025 0.603 0.024 0.606 0.036 0.633 Netherlands 0.081 0.146 0.073 0.180 0.065 0.233 Norway 0.018 0.715 -0.004 0.929 -0.003 0.954 Poland 0.096 0.067 * 0.152 0.002 *** 0.167 0.031 ** Portugal 0.092 0.274 0.107 0.202 0.122 0.191 Spain 0.035 0.515 0.102 0.097 * 0.070 0.256 Sweden 0.008 0.872 -0.011 0.829 -0.009 0.857 Switzerland 0.073 0.153 0.055 0.269 0.055 0.279 UK 0.056 0.181 0.084 0.065 * 0.055 0.187 Constant 0.197 0.035 ** 0.033 0.622 0.035 0.770 Adj. R Square 0.097 0.098 0.096
31
Table 12. Results on regression 2: institutional environment and underpricing (continued)Model 4 Model 5 Model 6
Variable Coeff. prob. Coeff. prob. Coeff. prob.
Deal size (log) -0.024 0.008 *** -0.022 0.015 ** -0.023 0.009 ***
Retention 0.043 0.043 ** 0.043 0.045 ** 0.043 0.042 ** IPO year -0.013 0.000 *** -0.013 0.000 *** -0.014 0.000 *** Technology 0.050 0.000 *** 0.049 0.000 *** 0.049 0.000 *** Hot market 0.034 0.002 *** 0.032 0.004 *** 0.033 0.002 *** VA - - - -PV - - - -GE - - - -RQ 0.081 0.119 - - - -RL - - -0.105 0.046 ** - -CC - - - - -0.050 0.273 Belgium 0.065 0.268 -0.023 0.722 0.008 0.896 Denmark 0.083 0.144 0.098 0.077 * 0.119 0.046 ** Finland 0.010 0.882 0.022 0.749 0.037 0.600 France 0.038 0.454 -0.050 0.321 -0.031 0.552 Germany 0.104 0.016 ** 0.075 0.094 * 0.090 0.040 ** Italy 0.053 0.353 -0.134 0.090 * -0.075 0.345 Netherlands 0.055 0.322 0.053 0.336 0.074 0.176 Norway 0.025 0.613 0.006 0.904 0.005 0.916 Poland 0.191 0.002 *** -0.016 0.848 0.048 0.578 Portugal 0.140 0.113 0.014 0.882 0.051 0.586 Spain 0.082 0.147 -0.027 0.673 0.062 0.251 Sweden 0.002 0.965 -0.010 0.845 0.005 0.918 Switzerland 0.060 0.226 0.058 0.246 0.064 0.199 UK 0.045 0.285 0.030 0.490 0.047 0.262 Constant -0.024 0.806 0.301 0.006 *** 0.214 0.054 * Adj. R Square 0.097 0.098 0.097
32
5.3 Interaction model
To test for the interaction effect between venture capital influence and institutional environment and the relationship with underpricing, regression 3 is tested. For each of the six indicators, the indicator and the interaction term between the venture capital backed dummy and the indicator were added to regression 1. The results can be found in table 13.
The results found in table 13 are mixed. Two indicators, VA and RQ, show the same sign as found in regression 2, including the same sign for the interaction term with venture capital. Since these results are insignificant, no conclusions can be drawn from them. The PV variable shows a switch in sign from positive to negative when interacted with the venture capital dummy, with a prob. of 11.3%. This results indicates that the negative relationship between venture capital influence and underpricing is stronger is countries that are politically more stable and the threat of violence and terrorism is lower. From this result it can be concluded that the indicator PV can be seen as a complement of venture capital influence to reduce IPO underpricing. The same relationship if found for the indicator GE, although not significant. The opposite relationship is found for the indicators RL and CC, although not significant. Therefore, no conclusions can be drawn about the interaction between these indicators and venture capital and their relationship with underpricing.
5.3.1 Robustness check: VC vs. non VC backed sample
To test if the relationship between the indicators of institutional environment quality and underpricing differs between venture capital backed and non venture capital backed IPOs, regression 2 is tested again, but now for the venture capital backed and non venture capital backed sample separately. The same six models are tested as in part 5.2. A short summary of the results for the six indicators for the six models can be found in table 12, excluding the results on the control variables, which are in line with the results in the previous parts.
33
Table 13. Results on regression 3: interaction effect and underpricing
Model 1 Model 2 Model 3
Variable Coeff. prob. Coeff. prob. Coeff. prob.
Deal size (log) -0.023 0.009 *** -0.025 0.006 *** -0.023 0.009 ***
Retention 0.042 0.047 ** 0.040 0.061 * 0.042 0.046 ** IPO year -0.013 0.000 *** -0.010 0.000 *** -0.011 0.000 *** Technology 0.049 0.000 *** 0.048 0.000 *** 0.049 0.000 *** Hot market 0.032 0.004 *** 0.033 0.002 *** 0.036 0.001 *** VC backed 0.087 0.523 0.049 0.233 0.014 0.855 VA -0.053 0.366 - - - -VC*VA -0.069 0.470 - - - -PV - - 0.062 0.080 * - -VC*PV - - -0.079 0.113 - -GE - - - - 0.037 0.493 VC*GE - - - - -0.016 0.716 RQ - - - -VC*RQ - - - -RL - - - -VC*RL - - - -CC - - - -VC*CC - - - -Belgium 0.041 0.470 0.053 0.352 0.044 0.442 Denmark 0.110 0.054 * 0.097 0.083 * 0.082 0.173 Finland 0.032 0.638 0.002 0.978 0.007 0.920 France -0.007 0.880 0.026 0.584 0.007 0.876 Germany 0.102 0.018 ** 0.105 0.015 ** 0.102 0.018 ** Italy -0.022 0.651 0.023 0.627 0.038 0.618 Netherlands 0.079 0.161 0.071 0.195 0.064 0.241 Norway 0.018 0.711 -0.001 0.982 -0.002 0.971 Poland 0.101 0.056 * 0.149 0.002 *** 0.168 0.030 ** Portugal 0.094 0.263 0.109 0.195 0.123 0.186 Spain 0.036 0.502 0.101 0.103 0.070 0.252 Sweden 0.010 0.849 -0.003 0.952 -0.007 0.892 Switzerland 0.075 0.143 0.061 0.221 0.057 0.262 UK 0.056 0.181 0.082 0.071 * 0.055 0.187 Constant 0.180 0.059 * 0.031 0.645 0.031 0.798 Adj. R Square 0.096 0.098 0.096
34
Table 13. Results on regression 3: interaction effect and underpricing (continued)
Model 4 Model 5 Model 6
Variable Coeff. prob. Coeff. prob. Coeff. prob.
Deal size (log) -0.024 0.008 *** -0.022 0.015 ** -0.023 0.009 ***
Retention 0.043 0.045 ** 0.043 0.043 ** 0.043 0.043 ** IPO year -0.013 0.000 *** -0.013 0.000 *** -0.014 0.000 *** Technology 0.050 0.000 *** 0.050 0.000 *** 0.049 0.000 *** Hot market 0.033 0.003 *** 0.031 0.005 *** 0.033 0.003 *** VC backed -0.032 0.744 0.031 0.688 0.006 0.928 VA - - - -VC*VA - - - -PV - - - -VC*PV - - - -GE - - - -VC*GE - - - -RQ 0.081 0.121 - - - -VC*RQ 0.012 0.847 - - - -RL - - -0.104 0.050 ** - -VC*RL - - 0.029 0.548 - -CC - - - - -0.048 0.296 VC*CC - - - - 0.011 0.775 Belgium 0.065 0.267 -0.024 0.713 0.009 0.888 Denmark 0.083 0.142 0.100 0.072 * 0.119 0.045 ** Finland 0.010 0.884 0.021 0.751 0.036 0.607 France 0.038 0.450 -0.050 0.318 -0.030 0.563 Germany 0.104 0.016 ** 0.075 0.093 * 0.091 0.039 ** Italy 0.052 0.357 -0.134 0.089 * -0.073 0.356 Netherlands 0.054 0.329 0.052 0.346 0.073 0.182 Norway 0.026 0.596 0.008 0.874 0.006 0.899 Poland 0.191 0.002 *** -0.017 0.843 0.049 0.563 Portugal 0.141 0.111 0.014 0.883 0.052 0.574 Spain 0.082 0.146 -0.028 0.663 0.061 0.255 Sweden 0.004 0.936 -0.007 0.891 0.007 0.885 Switzerland 0.062 0.215 0.061 0.224 0.066 0.186 UK 0.045 0.287 0.030 0.492 0.047 0.260 Constant -0.022 0.819 0.300 0.007 *** 0.211 0.058 * Adj. R Square 0.097 0.098 0.096
35
Table 14. Results on regression 2: VC backed sample vs. Non VC backed sample
VC backed (N=164) Non VC backed (N=1,551)
Variable Coeff. prob. Coeff. prob.
Model 1: VA 0.011 0.931 -0.062 0.325 Model 2: PV -0.062 0.523 0.061 0.105 Model 3: GE -0.048 0.746 0.031 0.593 Model 4: RQ 0.214 0.130 0.065 0.248 Model 5: RL -0.087 0.586 -0.102 0.070 * Model 6: CC -0.345 0.012 ** -0.040 0.414
Note: *,** and *** show the level of significance at a 10%, 5% and 1% level
confirms the result found in table 13 on the relationship between PV, venture capital influence and underpricing. Model 3 shows a change in sign from negative to positive in the absence of venture capital influence, although not significant. The first significant relationship is found in model 5, where RL is negatively related to underpricing for both samples, although only significantly for the non venture capital sample. This result highlights the importance of a high quality legal framework to reduce the level of underpricing, especially in the absence of venture capital influence. A contrary result is found for model 6, with a negative relationship between CC and underpricing for both samples, although only significantly for the venture capital backed sample. This result indicates that the negative relationship between venture capital influence and underpricing is stronger in countries with a higher level of control of corruption.
5.3.2 Venture capital influence and law system origin
36
IPOs per law system origin. Relatively most venture capital backed IPOs are found in the Scandinavian civil law countries, followed by German civil law countries. The data in this table is in line with the conclusion of Beck et al. (2003), who state that institutional environment quality is important to reduce the financing obstacles of firms. For this study, a higher percentage of venture capital backed IPOs is found for countries in high quality institutional environments.Table 15. Distribution of IPOs per law system origin
Law system # IPOs # VC IPOs % VC IPOs Quality *
Common law 774 71 9.17% 1.36
German civil law 291 36 12.37% 1.54
French civil law 392 27 6.89% 1.17
Scandinavian civil law 150 26 17.33% 1.67
Note: * institutional environment quality is measured as the average scores of the six indicators for all countries
37
Table 16. Results on regression 3. Interaction between VC and law system originModel 1 Model 2
Variable Coeff. prob. Coeff. prob.
Deal size (log) -0.022 0.012 ** -0.027 0.002 ***
Retention 0.028 0.193 0.014 0.481 IPO year -0.013 0.000 *** -0.012 0.000 *** Technology 0.058 0.000 *** 0.054 0.000 *** Hot market 0.037 0.001 *** 0.032 0.005 *** VC backed -0.017 0.434 0.001 0.936 Common law 0.013 0.245 - -VC*Common law 0.017 0.604 -
-German civil law - - 0.059 0.000 ***
VC*German civil law - - -0.067 0.081 *
French civil law - - -
-VC*French civil law - - -
-Scandinavian civil law - - -
-VC*Scandinavian civil law - -
-Constant 0.152 0.000 *** 0.162 0.000
Model 3 Model 4
Variable Coeff. prob. Coeff. prob.
Deal size (log) -0.019 0.031 ** -0.024 0.005 ***
Retention 0.039 0.062 * 0.021 0.296 IPO year -0.014 0.000 *** -0.013 0.000 *** Technology 0.058 0.000 *** 0.056 0.000 *** Hot market 0.036 0.001 *** 0.035 0.002 *** VC backed -0.026 0.132 -0.002 0.899 Common law - - - -VC*Common law - - -
-German civil law - - -
-VC*German civil law - - -
-French civil law -0.052 0.000 *** -
-VC*French civil law 0.071 0.088 * -
-Scandinavian civil law - - -0.023 0.208
VC*Scandinavian civil law - - -0.038 0.388
Constant 0.162 0.000 *** 0.168 0.000 ***
38
6. Discussion
No significant results are found for the first hypothesis, that venture capital influence reduces underpricing. Although the relationship found between venture capital influence and underpricing is negative and the descriptive statistics show a significant lower average underpricing for venture capital backed IPOs, the result of regression 1 is not significant. This result is in line with the latest studies on the role of venture capital on underpricing, where no significant differences were found (Jelic et al., 2005; Georgen et al., 2006; Klaassen and Von Eije, 2007; Tykova and Walz, 2007) for IPOs in Europe. The results can not significantly confirm the certification hypothesis explained by Megginson and Weiss (1991), that venture capital firms act as a form of third party certification to reduce information asymmetry and the level of underpricing.
The results on the relationship between institutional environment quality and underpricing are mixed. Three of the six indicators, Voice and Accountability, Rule of Law and Control of Corruption, meet the expectations that institutional environment quality reduces the level of information asymmetry and thus the level of underpricing. The Rule of Law variable is the only significant result, with an expected negative relationship on the level of underpricing. This result is in line with Engelen and Van Essen (2010), who studied two ways in which a country’s legal framework can influence the ex ante uncertainty about the value of a firm. The results of this study confirmed their conclusion that higher legal framework quality reduces the ex ante uncertainty and the level of underpricing. The positive, although not significant, relationship between Regulation Quality and underpricing contradict the helping-hand approach discussed by Barth et al. (2004) that government regulation can help to reduce information asymmetry. The positive relationship found is more in line with the grabbing-hand approach discussed by Barth et al. (2004), which states government failure is a bigger problem than market failure, leading to higher levels of information asymmetry. Overall, the results are only limitary in line with the results of Daouk et al. (2005) and Hopp and Dreher (2007), who state that institutional environment quality and capital market governance reduce the problem of information asymmetry and the level of underpricing.
39
influence to reduce underpricing. This result is confirmed when testing the relationship between the six indicators and underpricing for the venture capital backed and non venture capital backed sample separately, although not significant. A significant result is found for Control of Corruption, which shows a significant negative relationship to underpricing for the venture capital backed sample. This result indicates that Control of Corruption works as a complement to venture capital influence to reduce IPO underpricing. The results on the role of law system origin confirms the conclusion of Bruton et al. (2005), that when legal protection of shareholders is better, venture capital influence has a stronger negative effect on the level of underpricing.7. Conclusion and recommendations
In this study the relationship between venture capital influence, institutional environment quality and underpricing is investigated. The key reason relevant in this study to explain underpricing is information asymmetry. The dataset used consists of 1.715 IPOs, including 164 venture capital backed IPOs. The geographical area chosen is Europe, in order to investigate differences between countries due to the quality of the institutional environment. The time period chosen is 2000-2010, covering the last ten full calendar years.
40
These hypotheses are tested using three regression models, introduced and explained in the methodology section. In this section the independent variables, including the venture capital dummy and the six indicators to measure institutional environment quality are introduced. The data description section shows an average underpricing for the full sample of 10,04%, an average of 8,21% for the venture capital backed sample and an average of 10,23% for the non venture capital backed sample. These descriptive statistics give some first inferences about the influence of venture capital. In the last part of the data description section, the data on the six indicators is introduced by showing a summary of the data with the minimum and maximum scores for each indicator for each country.The results of this study are mixed. The first hypothesis, that venture capital influence reduces underpricing, cannot be significantly confirmed, although the sign is negative and the descriptive statistics show a significant lower average underpricing for venture capital backed IPOs. This result is in line with the latest studies mentioned in the literature review but it cannot confirm the certification hypothesis developed by Megginson and Weiss (1991). The results on the second hypothesis, which relates institutional environment quality to lower levels of underpricing, are mixed. Although three indicators show the expected negative relationship with underpricing, three indicators show a positive relationship with underpricing. The Rule of Law indicator is this only significant result, confirming the theory of Engelen and Van Essen (2010) that legal framework quality reduces underpricing. The results for the third hypothesis are limited. From these results it can only be concluded that the negative relationship between venture capital influence and underpricing is stronger in countries that are politically more stable. This result is confirmed by the robustness check, where the six indicators are related to underpricing for the venture capital and non venture capital backed sample separately.
41
8. References
Barnes E., Cahill A., McCarthy Y., 2002. Grandstanding in the UK venture capital industry,
EFMA 2002 London Meeting Papers.
Barry C.B., 1989. Initial Public Offering underpricing: the issuer’s view, Journal of Finance 44, 1099-1103.
Barry, C., Muscarella, C., Peavy, J., Vetsuypens, M., 1990. The role of venture capital in the creation of public companies: Evidence from the going public process, Journal of Financial
Economics 27, 447- 471.
Barth J., Caprio G., Levine R., 2004. Bank regulation and supervision: what works best?,
Journal of Financial Intermediation 13, 205-248.
Bascha, A., Walz, U., 2001. Convertible securities and optimal exit decisions in venture capital finance, Journal of Corporate Finance 7, 285-306.
Beatty R.P., Ritter J.R., 1986. Investment banking, reputation, and the underpricing of Initial Public Offerings, Journal of Financial Economics 15, 213-232.
Beck T., Demirgüç-Kunt A., Laeven L., Maksimovic V., 2003. The determinants of financing obstacles, Journal of Internation Money and Finance 25, 932-952.
Benveniste L.M., Spindt P.A., 1989. How investment bankers determine the offer price and allocation of new issue, Journal of Financial Economics 24, 343-361.
Bessler W., Kurth A., 2004. The performance of venture-capital-backed IPOs in Germany: exit strategies, lock-up period and bank ownership, EFMA 2004 Basel Meetings paper.
Bradley D.J., Jordan B.D., 2002. Partial adjustments to public information and IPO underpricing, Journal of Finance and Quantitative Analysis 37, 595-616.
42
Brooks C., 2008. Introductory econometrics for finance, Cambridge University Press, Cambridge, UK.Bruton G.D., Fried V.H., Manigart S., 2005. Institutional influence of the worldwide expansion of venture capital, Entrepreneurship Theory and Practice 29, 737-760.
Buckley S., Duer K.M., Mendel T., 2008. Broadcasting, voice and accountability: a public
interest approach to policy, law and regulation, University of Michigan press.
Cumming D., MacIntosh J., 2003. A cross-country comparison of full and partial exits,
Journal of Banking and Finance 27, 511-548.
Daily, S., Dalton D., Roengpitya R., 2003. IPO underpricing: A meta-analysis and research synthesis, Working paper, ET&P Baylor University, Texas.
Daouk H., Lee C.M.C., Ng D.T., 2005. Capital market governance: how do security laws affect market performance? Working paper, Cornell University, Ithaca New York.
Demirgüç-Kunt A., Maksimovic V., 1998. Law, finance, and firm growth, Journal of Finance 53, 2107-2137.
Dissanaike G., Amel-Zadeh A., 2007. Discussion of venture capitalists, business angels, and performance of entrepreneurial IPOs in the UK and France, Journal of Business Finance &
Accounting 34, 529-540.
Dolvin S., 2005. Venture capital certification of IPOs, Venture Capital 17, 131-148.
Engelen P.J., Van Essen M., 2007. Underpricing of IPOs and investor protection around the world: country-level evidence. Working paper, Utrecht University.
Engelen P.J., Van Essen M., 2010. Underpricing of IPOs: firms-, issue- and country-specific characteristics, Journal of Banking and Finance 34, 1958-1969.
43
Franzke, S.A., 2003. Underpricing of venture-backed and non venture-backed IPOs: Germany’s Neuer Market, CFS Working Paper.Francis B.B., Hasan I., 2001. The underpricing of venture and non venture capital IPOs; an empirical investigation, Journal of Financial Services Research 19, 99-113.
Gladstone D., Gladstone L., 2002. Venture capital handbook: an entrepreneur’s guide to
raising venture capital, FT press, New Jersey, 26-33.
Goergen M., Khurshed A., Renneboog L., 2009. Why are the French so different from the Germans? Underpricing of IPOs on the new Euro market. International Review of Law and
Economics 29, 260-271.
Hamao Y., Packer F., Ritter J., 2000. Institutional affiliation and the role of venture capital: evidence from Initial Public Offerings in Japan, Pasific-Basin Finance Journal 8, 529-558. Helwege J., Liang N., 2004. Initial Public Offerings in hot and cold markets, The Journal of
Financial and Quantitative Analysis 39, 541-569.
Hopp C., Deher A., 2007. Do differences in institutional and legal environments explain cross-country variations in IPO underpricing?, CESifo Working paper 2082.
Jelic R., Saadouni B., Wright M., 2005. Performance of private to public MBOs: the role of venture capital, Journal of Business Finance & Accounting 32, 643-681.
Kaufman D., Kraay A., Mastruzzi M., 2010. The worldwide governance indicators: methodology and analytical issues, Policy Research Working Paper, The World Bank.
Klaassen, E., Von Eije, H., 2007. Earnings growth and underpricing with venture capitalists backed IPOs, Working Paper, University of Groningen, The Netherlands.
La Porta R., Lopez-de Silanes F., Shleifer A., Vishny R.W., 1998. Law and finance, Journal
of political economy 106, 1113-1155.
44
Ljungqvist A., 2007. IPO underpricing, Handbook of Corporate Finance: EmpiricalCorporate Finance 1, 253-304.
Loughran, T., Ritter J.R., 2004. Why has IPO underpricing changed over time?, Financial
Management 33, 5-37.
Megginson, W.L., Weiss, K.A., 1991. Venture capitalist certification in Initial Public Offerings, Journal of Finance 46, 879-903.
Tykova T., Walz U., 2007. How important is participation of different venture capitalists in German IPOs?, Global Finance Journal 17, 350-378.
