Master Thesis
MSc BA Small Business & Entrepreneurship
The impact of Independent Venture Capital, Corporate Venture
Capital and their Syndication on the financial performance of
clean-tech ventures
by
Tommaso Marchetti
S3225658
Word Count: 11529
University of Groningen
Faculty of Economics and Business
January 2018
2
Abstract
This quantitative Thesis investigates the impact of Independent Venture Capital (IVC), Corporate Venture Capital (CVC) and their syndication on the financial performance of clean-tech firms. The goal was to discover which VC investment form has the best financial performance in clean-tech companies. Drawing from the agency theory and the resource-based view theory, this Thesis offers new insights into the financial performance of IVC-backed clean-tech companies. The results, estimated through econometric analyses of a sample of 147 VC deals, show a positive impact of IVC on the financial performance of clean-tech firms. On the other hand, CVC and IVC-CVC syndication were found to be not statistically significant. However, the negative results of CVC can be attributed to the minor attention of this type of VC towards the financial performance of the company, while, regarding the IVC-CVC syndication, the positive effect linked to the resource-based view and the negative effect linked to the agency costs cancel each other out.
Keywords: Independent Venture Capital, Corporate Venture Capital, IVC – CVC syndication,
3
Acknowledgements
4
Table of contents
1. Introduction ... 6
2. Literature review ... 9
2.1 The impact of IVC on firm performance ... 9
2.2 The impact of CVC on firm performance ... 11
2.3 The impact of the syndication between IVC and CVC on firm performance ... 12
3. Methodology ... 16
3.1 Data collection ... 16
3.2 Measurements ... 17
3.2.1 Dependent variables ... 17
3.2.2 Independent variables ... 18
3.2.3 Control variables ... 18
3.3 Analysis ... 19
3.3.1 Descriptive statistics ... 19
3.3.2 Regression models employed to test the hypotheses ... 20
4. Results ... 21
4.1 Correlation analysis ... 21
4.2 Testing the hypotheses ... 21
4.3 Robustness checks ... 27
5. Discussion and conclusion ... 28
5.1 Practical implications ... 29
5.2 Limitations and future research ... 29
6. References ... 30
5 “In order to limit global warming to 2 degrees Celsius and avoid the worst effects of climate change, investments in low-carbon energy technologies will need to at least double, reaching $500 billion
6
1. Introduction
The importance of Venture Capital (VC) in financing young high-tech entrepreneurial firms has been widely analysed in the entrepreneurship, economics and finance literature since its diffusion in the American market and subsequently in the European one (e.g., Denis,2004; Gompers et al., 2009; Gompers and Lerner, 2001, 2004). The literature highlights that the role of venture capital investors (VCs) is not limited to overcoming small-medium companies’ funding problems, but also to providing relevant services such as signalling, monitoring and giving advice. Namely, getting a VC investment may signal the backed venture’s value to less informed investors. Furthermore, venture capitalists often have relevant positions within the management though which they can check the progress of the firm and provide advice about growth strategies (Gompers and Lerner, 1998).
There are different forms of VC but in this Thesis I will focus on the two most popular VC types: independent and corporate VC (IVC and CVC) funds, and their syndicated investments.
IVC can be described as the typical VC investment, where the IVC firms create a fund (generally organised as a limited liability partnership) to raise capital from several outside sources (e.g. public pension funds, insurance companies, high net-worth individuals, foundations, endowments and family offices). This capital is then used to finance young, rapidly growing companies (Alvarez-Garrido and Dushnitsky, 2016; Christofidis and Debande, 2001; LiPuma, 2006; Sahlman, 1990). The main objective of IVC is to obtain the greatest internal rate of return (IRR) in the shortest possible time (Colombo and Murtinu, 2017; Cumming et al., 2005). In fact, its life span is limited and only in some rare cases it can last more than ten years (Chemmanur et al., 2014).
7 tend to be more innovative than IVC-backed firms because CVCs are more failure tolerant than IVCs. This feature is extremely relevant to obtain innovative information from ventures at high risk of bankruptcy (Tian and Wang, 2011).
This Thesis also investigates the effect of the combination of multiple VCs on the performance of clean-tech ventures. This practice, defined “syndication”, is increasingly common in the venture capital context. For example, between 1980 and 2005, about 70% of VC-backed ventures received at least one investment by two or more VC investors (Tian, 2011). VCs constitute syndicates to obtain an additional opinion on the value of their investment, to spread the risk between the investors and to exploit the resources and skills of other VCs (Brander et al., 2002; Lerner, 1994; Wilson, 1968). There are several forms of syndicates, which consist in different combinations of VC types. This research investigates only syndicates between CVC and IVC.
8 attempt to greenwash1 rather than as a real effort to reduce environmental degradation (Cherry and
Sneirson, 2010; Ghosh and Nanda, 2010).
Furthermore, this sector has attracted a lot of attention since a lot of policy-makers recognise clean-tech industry as a “valuable asset in creating jobs, improving environmental performance, and promoting national resource independence” (Burtis et al., 2006. p.3).
Few studies in the field of VC have addressed the relation between clean-tech companies and the effect of different types (and composition) of VC investments. More precisely, it is not completely clear how VCs interact with clean-tech firms. This Thesis tries to fill this literature gap by using two different theories: agency theory and resource-based view theory. Agency theory incorporates the agency problem that arises between two (or more) cooperating parties with different interests, goals and vision of labour (Eisenhardt, 1989). Typically this harmful relationship occurs between the principal (e.g. the owner) and the agent (e.g. the manager). When referring to VC deals, this conflict of interest occurs between the venture capitalist and the entrepreneur. The most common problem between the parties is the opportunistic behaviour of the agent. When the agent behaves in an individualistic and self-serving way, the principal must bear the additional cost of imposing internal controls to check the agent’s behaviour (Lahovnik, 2008). Differently, the resource-based view is a theory used to determine the relevant resources useful for generating a competitive advantage in a firm. In order to achieve sustainable competitive advantage, these resources have to be valuable, rare, costly to imitate and non-substitutable (Barney, 1991).
Based on the literature gap, the agency theory and the resource-based view theory, I have formulated the following research question:
What investment form between IVC, CVC and IVC-CVC syndication has the best financial performance in the clean-tech companies?
This Thesis is organised as follows. Section 2 presents the development of the hypotheses. Section 3 reports the methodology used to empirically test the hypotheses and describes the sample selection procedures and the main descriptive statistics. Section 4 examines the empirical results. Finally, section 5 shows the conclusions.
1 “Greenwash is an attempt to make people believe that your company is doing more to protect the environment
9
2. Literature review
2.1 The impact of IVC on firm performance.
The resource-based view theory, first introduced by the economist Penrose in 1959, can be related to IVC as this form of financing can provide additional and relevant resources. This theory deals primarily with the application of a set of valuable tangible or intangible resources, available within the firms, in order to obtain a competitive advantage (Penrose, 1959; Wernerfelt, 1984). Barney (1991) and Peteraf (1993) added that, in order to transform a short-term competitive advantage into a long-term one, these resources should be heterogeneous in nature, not perfectly mobile, inimitable and not easily substitutable (Barney, 1991; Peteraf, 1993). Moreover, obtaining financing through IVC may draw on the resource-based view theory since venture capital, in addition to monetary funds, contributes to valuable activities such as: “obtaining additional financing, strategic planning, management recruitment, operational planning, introductions to potential customers and suppliers, and resolving compensation issues” (Maula et al, 2005, p.6). This financial and strategic aid, provided by venture capital, can be considered as a resource that positively influences the performance of the firm and subsequently its competitive advantage (Barney, 1991).
In recent years, a large body of literature, related to the impact of IVC on firm performance, has been analysed by scholars (Colombo and Murtinu, 2017; Hellmann and Puri, 2002; Macmillan et al., 2008). Standard venture capital investors, i.e. IVC investors, are specialised in the first stages of financing (e.g. start-up and seed) and are often forced to deal with new entrepreneurial teams who are inexperienced but highly talented. For this reason, the ability of IVCs in sharing their experience and expertise may be more relevant than financing (Gorman and Sahlman, 1989; Hellmann and Puri, 2002; Sapienza et al., 1994). On this important aspect, Smith (2001) claimed that firms consider the possibility of receiving value-adding services more important than receiving financing. The existing literature has identified three different roles of IVCs: strategic, interpersonal and financial (Rosenstein et al., 1993; Sapienza et al., 1994). The first two, strategic and interpersonal roles, occur when the venture capitalist works with the top management, giving suggestions about business and finance, and in some cases he or she can also replace the CEO. The last role concerns all the activities about financial advising such as obtaining additional funds (Maula et al., 2005). In other words, the main result of all services provided by IVCs is the improvement of the venture capital-backed firm’s performance.
10 better understand this relationship (Gompers and Lerner, 2004). Agency relationship can be described as a legal agreement under which at least one person (the principal/s) engages another person (the agent) to execute some services on their behalf which includes assigning some decision-making authority to the agent (Jensen and Meckling, 1976). As for the VC context, Reid (1999) was one of the first researcher that described the entrepreneur as an agent and the venture capitalist as a principal. Since the entrepreneur and the venture capitalist are both characterised by self-interest and bounded rationality, agency contrasts emerge when the two parties have different risk perceptions or goals (Christensen et al., 2009). Bounded rationality can lead to a condition of asymmetric information which, in turn, generates opportunistic behaviours, such as adverse selection and moral hazard. Adverse selection, usually described as “hidden information”, occurs when one party has relevant information that is not known by the other party. The problem arises when the informed party (typically the entrepreneur in the VC context) tries to overstate the business opportunity to the less informed party (the venture capitalist) in order to obtain better financing conditions (Amit et al., 1998). Moral hazard, often referred to as “hidden action”, is a form of post-contractual opportunism, which can lead entrepreneurs to pursue their own interests at the expense of the venture capitalist, relying on the fact that the latter cannot verify the presence of malice or negligence. As a result, venture capitalists have developed several processes to reduce the risks and to align their goals to those of entrepreneurs. Economic literature has focused in particular on the screening and monitoring process. In the screening process, IVCs can select only the firms with high growth rate, through in-depth due diligence (Chan, 1983; Kaplan and Stromberg, 2001). This process can be correlated to the average higher performance of IVC-backed companies since IVCs are able to select high potential firms (Gompers and Lerner, 2001). Through the monitoring process, IVCs not only control the entrepreneur’s behaviour, but also provide relevant advice and intangible assets (e.g. extensive knowledge and networks) to add value to the IVC-backed firms (Gorman and Sahlman, 1989). Therefore, IVCs can simultaneously reduce the agency problems and improve the performance of these firms.
11 As mentioned in the Introduction, the clean-tech industry has unique characteristics that distinguish it from many other sectors. Investors have more difficulties in assessing risks and opportunities of the clean-tech deal. Investments in this kind of industry are riskier than other VC deals because several clean-tech companies do not operate on the consumption side of the economy, but on the production side (“e.g., electricity generation, energy efficiency, composite materials, wastewater treatment”) (Cumming et al., 2016, p.88) where it is more difficult to calculate market growth opportunities and risks (Nanda et al., 2014). Another problem can arise from the fact that it is a relatively young sector. In fact, many clean-tech ventures do not have a developed track record. This shortage leads to a higher uncertainty for investors. Hence, venture capitalists face several issues to evaluate the profitability of the deal and this feature may increase the conflicts with the entrepreneur, especially regarding adverse selection.
However, taking into account the resource-based view theory, the agency theory and the extant literature about IVC and firm performance, my hypothesis is:
H1: IVC investments have a positive impact on the financial performance of clean-tech IVC-backed
firms.
2.2 The impact of CVC on firm performance.
12 profitable but more innovative than IVC-backed firms (Chemmanur et al,. 2014). Since clean-tech ventures are highly innovative, they could get significant benefits from this peculiar characteristic of CVC (Burtis et al., 2006). Clean-tech firms could obtain a higher innovation level through CVC investment respect to IVC financing. Furthermore, CVC-backed ventures are more focused on improving the level of innovation rather than obtaining significant financial performance.
Agency problems do not only exist in the relationship between IVC and entrepreneurs, but also between CVC and entrepreneurs. Especially in the CVC context, the association principal-venture capitalist and agent-entrepreneur has been questioned by many scholars. A large and increasing collection of studies look at this relationship from a different angle. They consider the role agent-principal in the opposite way: the venture capitalist as an agent and the entrepreneur as a agent-principal (Cable and Shane, 1997; Gifford, 1997; Christensen et al., 2009). Therefore in the context of CVC, following this new relation, moral hazard risks may also include the potential misappropriation of “the key intellectual property assets of the entrepreneur, limiting the entrepreneur’s options to co-operate with other corporations or even, in extremis, the corporate partner becoming a direct competitor to its portfolio company” (Maula et al., 2003, p. 121). In addition, Maula et al. (2003) claimed that, in order to reduce these agency problems, the best solution would be to align the interests between the parties as much as possible through increased ownership.
Consequently, the entrepreneur of a clean-tech company not only fears the possible theft of technology by the corporate investor, but in addition he or she has a lower bargaining power due to the characteristics of the clean-tech industry. In fact, the clean-tech sector is characterised by high exit barriers that generate huge risks for entrepreneurs (Cumming et al., 2016).
To conclude, taking into account the agency theory and the extant literature about CVC and firm performance, my second hypothesis is:
H2: CVC investments do not have a significant impact on the financial performance of clean-tech
CVC-backed firms.
13 syndication composed by two or more IVCs) or between a combination of different VCs (for instance an IVC and a CVC). In this Thesis, only syndicates between CVC and IVC will be analysed.
VCs constitute syndicates to get an additional opinion on the value of their investment, to spread the risk between the investors and finally to exploit the resources and skills of other VCs (Brander et al., 2002; Lerner, 1994; Wilson, 1968). However, Casamatta and Haritchabalet (2007) and Meuleman et al. (2010) argued that this kind of investment may lead to additional costs due to conflicts for non-aligned objectives or uncertainty about partners’ knowledge. Consequently, IVC-CVC syndication can have the same complementary and conflictual relationships (Masulis and Nahata, 2009). In the following part, two different views and theories will be analysed to better understand the opposite effects of IVC-CVC syndication on the performance of clean-tech firms.
15 To conclude, taking into account the resource-based view theory, the agency theory and the extant literature about IVC-CVC syndication and firm performance, I formulated two contradictory hypotheses:
H3a: IVC-CVC syndication investments have a positive impact on the financial performance of
clean-tech IVC-backed firms. (RESOURCE BASED VIEW THEORY)
H3b: IVC-CVC syndication investments have a negative impact on the financial performance of the
clean-tech IVC-backed firms. (AGENCY THEORY)
16
3. Methodology
3.1 Data collection
In this Thesis, two available databases (Orbis and Zephyr) were used in order to collect data. Orbis and Zephyr are databases owned by Bureau van Dijk Electronic Publishing, which is a primary source for company data. Bureau van Dijk (BvD) owns information on 275 million companies worldwide organised in several databases which focus on corporate finance, merger and acquisitions (M&A), compliance and third party due diligence. Orbis is a database containing all kind of financial and accounting data about more than 200 million companies in over 200 countries. Zephyr includes 1.2 million deals about M&A, IPO and venture capital deals. Thus, I focused my efforts to find the desired data in the afore-mentioned databases.
17 (IPO) was later used to measure the financial performance. Unfortunately, several companies presented missing financial data, and for this reason, were dropped. Finally, the sample consisted in 147 VC-backed clean-tech firms in 26 countries.
The next step was to classify the different types of VC. It was not so easy to split the VC deals into independent, corporate and syndication. In fact, despite Zephyr shows several features for each deal such as the number and the type of investors (“Independent”, “Corporate Venturing”, “Business Angel”), many times the deals are not classified in detail. Thus, this limitation has led to a manual search for each individual deal.
Furthermore, a control sample was collected. The control group was searched in Orbis using the same criteria of the previous sample, except not all the keywords were used. For this control sample, the following keywords were selected: “tidal power”, “renewable energy”, “biofuel” and “wind power”. These keywords were randomly chosen to prevent selection bias. Once again, the time period starting from 01-01-2010 to 01-01-2017 was selected. As mentioned previously, the ventures with missing financial data were dropped. The final control sample of non-VC-backed clean-tech firms consisted in 1748 firms in 25 countries.
3.2 Measurements
In the following section, dependent, independent and control variables are explained. In order to identify the most appropriate variables for this Thesis, a large amount of empirical papers were investigated. Thus, all the measures are inspired by extant literature.
3.2.1 Dependent variables
18 Gompers, 1996). Specifically, IPO is a dummy variable that equals 1 if the firm has been listed, and 0 otherwise.
3.2.2 Independent variables
In the following table, several independent variables are explained: Table 1. Independent variables definition
Independent Variables Description
IVC Dummy variable that equals 1 if the clean-tech company is financed by only an IVC; 0 otherwise. This variable takes into account the year of investment: it is equal to 1 in the year of investment and in the following years; 0 in the years before the investment.
CVC Dummy variable that equals 1 if the clean-tech company is financed by only a CVC; 0 otherwise. This variable takes into account the year of investment: it is equal to 1 in the year of investment and in the following years; 0 in the years before the investment.
Syn Dummy variable that equals 1 if the clean-tech company is financed by an IVC-CVC syndication; 0 otherwise. This variable takes into account the year of investment: it is equal to 1 in the year of investment and in the following years; 0 in the years before the investment.
IVC_pre Dummy variable that equals 1 only in the year before the IVC investment; 0 in all the other years, before and after the investment.
CVC_pre Dummy variable that equals 1 only in the year before the CVC investment; 0 in all the other years, before and after the investment.
Syn_pre Dummy variable that equals 1 only in the year before the IVC-CVC syndication investment; 0 in all the other years, before and after the investment.
The last three variables were added in order to take into consideration the “screening effect”. In fact, several researchers (Gompers and Lerner, 2001; Croce et al., 2013) have investigated the ability of VCs in the selection of firms with superior performance before the investment.
3.2.3 Control variables
19 company at the moment of the investment was included as a control variable since older companies could perform better than younger ones. In fact, young companies might face several limitations regarding for example financial resources and management experience. The first step, in order to obtain this control variable, was to calculate the variable Age. This was obtained by subtracting the investment year from the incorporation year. After that, the variable Age_log was calculated through the natural logarithm of the Age. With this process, the distribution of the control variable is less skewed and closer to a normal distribution. In addition, country dummy variables were included in order to eliminate the influence of different economies. In fact, different organisational cultures, regulations and laws might have a positive or negative effect on the performance of the companies taken into consideration. The year dummies were further constructed in order to control the macroeconomic effects such as recessions or years of unrestrained economic growth. In fact, these exceptional economic events might impact on VC-backed firms in a different way.
3.3 Analysis
In order to summarise the extensive set of data, a descriptive statistics table of the sample was drawn up. This table shows the most widespread measurements such as the number of observations, mean, standard deviation, minimum, maximum, skewness and kurtosis. After that, a correlation analysis was conducted in order to measure the degree of association between pairs of variables. Subsequently, the hypotheses were empirically tested through a logit and a panel fixed-effects (FE) regression analysis, considering the two distinct dependent variables of this Thesis. A logit model was employed in order to estimate the odds of the binary dependent variable (IPO). Furthermore, according to Lewis (2007), the quality of predictors in a multiple regression analysis can be established with two different methods: stepwise regression or hierarchical regression. In this case, the latter was chosen since the predictors are selected based on theory and past research. Furthermore, hierarchical regression is preferred in social sciences research, due to the strong correlation between predictors.
3.3.1 Descriptive statistics
20 levelness of a distribution compared to a normal distribution. As can be observed in Table 2, only the Age_log has a negative skewness. In addition, almost all the variables present a positive and different from zero skewness, consequently the data is not symmetric and does not follow a normal distribution. Regarding the kurtosis, it is positive for all the variables and extremely large for half of these. This means that the variables with a high kurtosis are characterised by a widely peaked distribution.
Table 3 reports the distribution of the different types of VC. It is easily noticeable that the highest percentage of the sample is represented by IVC investments. It is far greater than the others, but in particular than the CVC. The few data available regarding CVC have been a limitation for this research (cf. limitation section. p. 29).
Table 3. Types of VC
Type of VC No. Percentage
IVC 106 72,11%
CVC 16 10,88%
IVC-CVC syndication 25 17,01%
Total 147
3.3.2 Regression models employed to test the hypotheses
21
4. Results
In this section, the results obtained from the correlation and regression analyses are discussed in detail. The first part concerns all the pairwise correlations between the variables. It is relevant to analyse the correlation analysis because highly correlated variables can alter the regression results. In the second part, the regression results obtained from the various models are analysed.
4.1 Correlation analysis
Table 4. Correlation analysisNote: *Represents statistical significance at 1%
Table 4 displays the results of the correlations analysis of the variables of interest. As it is clear from the table, there are no strong correlations between the variables, hence there are no issues of multicollinearity. In fact, the highest positive correlations exist between Totalassets_log and IPO, and between Totalassets_log and Age_log. The values of these two correlations are, respectively, 0.23 and 0.27. In addition, the relationship between Age_log and IVC is characterised by the highest negative correlation, -0.25. However, these three values are not likely to create problems in the following regressions.
Furthermore, in order to verify multicollinearity, the variance inflation factor (VIF) was tested. This value must be lower than 10, otherwise it means that there is a correlation among independent variables. In this Thesis, the mean VIF (1.01) is much lower than the threshold so there is no reason for concern about multicollinearity.
4.2 Testing the hypotheses
As mentioned above, the first model regresses the dependent variable (IPO) on the independent (IVC, CVC, and Syn) and control variables. Table 5 below displays the results of this logit regression.
22 Table 5. Model 1, IPO as dependent variable (N=11517)
Logit Logit
Variable Coefficients Odds ratio
IVC 1.5096*** 4.5249*** (0.4646) (2.1021) CVC -1.2161 0.2963 (1.1068) (0.3280) Syn 1.1969 3.3097 (1.2111) (4.0085) Age_log 0.6420*** 1.9003*** Year Dummies Country Dummies (0.1545) YES YES (0.2935) YES YES
Note: *** Represents statistical significance at 1% ** Represents statistical significance at 5% * Represents statistical significance at 10% Marginal effects; Standard errors in parentheses
First of all, it is important to highlight that this first model is statistically significant since the chi2 test rejects that all coefficients are jointly null (p<0.000), meaning that the model is able to predict the performance of the VC-backed firm. In addition, the R-squared is 0.2533. This means that the percentage of the variance in the IPO dummy explained by the independent variables is equal to 25.33%. Table 5 displays the coefficients with their marginal effects in parentheses in the first column and the odds ratio with their standard errors in parentheses in the second column. The interpretation of a logit model is quite different from a linear regression model. For this reason, in order to obtain an easier interpretation, odds ratios were calculated in the logit model.
23 CVC was found to be non-significant. However, its negative coefficient (-1.2161) implies that CVC investments negatively influence the probability of becoming a public company. The main reason for this negative effect might be linked to the fact that CVCs are more focused on improving the level of innovation instead of obtaining significant financial performance. However, since this variable was not found to be significant, not much can be concluded.
Also Syn was found to be non-significant. Nevertheless, in this case, its coefficient and odds ratio are positive (respectively, 1.1969 and 3.3097). It is easily noticeable that the standard error of Syn is high and consequently the variance of the effect of syndicate investments on the likelihood of becoming public is high. Therefore, some syndicate investments might be important to reach the IPO while others might not. This may be an area of future research.
The second model regresses the dependent variable (Totalassets_log) on the independent variables (IVC, CVC, and Syn) and control variables. The Table 6 below displays the results of this regression.
Table 6. Model 2, Fixed effects with Total assets as dependent variable (N=11186) Mutiple regression Variables Coefficients IVC 0.6339*** (0.2088) CVC 0.0219 (0.4502) Syn 0.2908 (0.2448) Age_log 0.0000 Year Dummies Country Dummies (.) YES YES
Note: *** Represents statistical significance at 1% ** Represents statistical significance at 5% * Represents statistical significance at 10% Marginal effects; Standard errors in parentheses
24 In fact, in order to test the endogeneity, this model is not a logit model but a fixed-effects panel regression model in which the independent variables (IVC, CVC and Syn) are binary variables and the dependent variable (Totalassets_log) is a logarithmic function. The interpretation of the coefficients is easier than for the first model.
Again, as in the first model, only the coefficient for IVC was found significant at 1%. The coefficient for this variable is equal to 0.6339, meaning that a clean-tech company financed by an IVC increases its total assets by 63.39% compared to a non-IVC-backed clean-tech company. Therefore, the positive effect of IVC on the performance of the IVC-backed firms is again demonstrated.
CVC was found to be not statistically significant. However, its positive coefficient close to zero (0.0219) implies that CVC investments positively influence the total assets of the CVC-backed company. However, since this variable was not found to be significant, not much can be concluded. Also in this model Syn was found to be not statistically significant, but positive. However, the positive effect linked to the resource-based view and the negative effect linked to the agency costs cancel each other out. Consequently, there is neither a positive, nor a negative effect.
Taking the first two models into consideration, H1 and H2 are supported, while H3a and H3b are rejected.
The third and the fourth models were performed adding three more independent variables (IVC_pre, CVC_pre and Syn_pre). These new variables were added in order to disentangle the screening (or selection) and the value-adding (or treatment) effect of the VCs. Table 7 below shows the results of the logit model through the coefficients and the odds ratios.
Table 7. Model 3, IPO as dependent variable (N=11517)
Logit Logit
Variable Coefficients Odds ratios
25 CVC -1.2019 0.3006 (1.1472) (0.3448) Syn 1.2327 3.4304 (1.2229) (4.1951) Age_log 0.6596*** 1.9339*** Year Dummies Country Dummies (0.1554) YES YES (0.3005) YES YES
Note: *** Represents statistical significance at 1% ** Represents statistical significance at 5% * Represents statistical significance at 10% Marginal effects; Standard errors in parentheses
The third model is once again statistically significant since the chi2 test rejects that all coefficients are jointly null (p<0.000). Analysing the coefficients and the odds ratios of the independent variables, it is possible to understand the real effect of VCs. The coefficient and the odds ratio of IVC_pre are significant at 1% and are equal, respectively, to 1.5882 and 4.8950. This variable describes the screening effect and its odds ratio explains that in the year before the investment, IVCs are able to select firms that already have a higher probability to obtain an IPO by a factor of 4.8950. On the other hand, IVC indicates, as already explained in the first two models, the treatment effect of IVC. The odds ratio of this variable is positive (4.7233) and statistically significant at the one percent level. Hence, the likelihood of an IVC-backed clean-tech company of becoming public is, on average, 4.7233 times larger than a non-IVC-backed clean-tech company. However, these results need an additional analysis since the odds ratio of IVC_pre is larger than IVC. For this reason, the linear combination of IVC minus IVC_pre was calculated. The result shows an odds ratio of 0.9649 with a p-value of 0.841, so highly insignificant. The explanation behind this result is that the value addition ability minus the screening capacity was found to be not statistically different from zero. Consequently, the net effect of IVC on IPO is a pure screening effect.
26 In addition, the odds ratios of Syn_pre and Syn were not found to be significant, but highly positive (respectively, 1.5263 and 3.4304). Thus, it seems that also syndications try to select high performance firms before the investment. However, the high standard error leads to a not statistically significant outcome.
The fourth and last model combines the same previous independent variables (IVC_pre, IVC, CVC_pre, CVC, Syn_pre, and Syn) with a new dependent variable (Totalassets_log), using a fixed effect panel regression model, controlling for screening effects by means of the procedure developed by Chemmanur et al. (2011). The following Table 8 displays the results of this regression.
Table 8. Model 4, Fixed effects with Total assets as dependent variable (N=11186) Multiple regression Variables Coefficients IVC_pre 0.2446 (0.1831) CVC_pre -0.4115 (0.4580) Syn_pre 0.3138 (0.2730) IVC 0.7356*** (0.2272) CVC -0.1092 (0.4139) Syn 0.4062 (0.2945) Age_log 0.000 Year Dummies Country Dummies (.) YES YES
Note: *** Represents statistical significance at 1% ** Represents statistical significance at 5% * Represents statistical significance at 10% Marginal effects; Standard errors in parentheses
27 effect and the value adding effect. The main interesting difference from the previous model is that the coefficient for IVC_pre is not statistically significant anymore. Thus, the screening effect, identified in Model 3, does not appear in this model. On the other hand, the coefficient for IVC is positive (0.7356) and statistically significant at 1%. This means that a clean-tech company, financed by an IVC, increases its total assets by 73.56% compared to a non-IVC-backed clean-tech company. The coefficient for the linear combination between IVC and IVC_pre is equal to 0.4910 with a p-value of 0.035, so statistically significant at 5%. Consequently, the net effect of IVC on total assets is a pure value addition effect. Hence, activities such as “obtaining additional financing, strategic planning, management recruitment, operational planning, introductions to potential customers and suppliers, and resolving compensation issues” (Maula et al, 2005, p.6) contribute to an improvement in the performance of the IVC-backed firm. This outcome is in line with the existing literature (Maula et al., 2005; Smith, 2001).
Regarding the remaining independent variables (CVC_pre, CVC, Syn_pre, and Syn), their coefficients were not found to be significant and their values are similar to those found in the third model. CVC has a negative influence both in the process before the investment and after (CVC_pre = -0.4115 and CVC= -0.1092) while the coefficients of syndicate were both found to be positive (Syn_pre = 0.3138 and Syn = 0.4062).
To conclude, the last two models show interesting results. First of all, in the case of IPO, the screening effect is relevant. Hence, IVC-backed clean-tech companies are more likely to go public because IVC are able to select firms with higher performance before the investment. In the case of total assets, there is no evidence of the screening effect, but IVCs increase the total assets through the supply of valuable resources and relevant activities.
Regarding the hypotheses, the same previous results can be confirmed. H1 and H2 are supported, while H3a and H3b are rejected.
4.3 Robustness checks
It is important to perform robustness checks to test the previous regression outcomes. Through this process, it is possible to verify the reliability and the strength of the statistical procedures. Robustness checks were performed by changing Model 2 and Model 4. Before, in order to test the endogeneity, these models were performed with a multiple regression with fixed effects, they were then estimated again with a multiple regression, but with random effects. These new models (Model 5 and Model 6) are displayed in Table 9 and in Table 10 in the Appendix.
28 negative (-0.0915). However the coefficient is still highly insignificant. IVC and Syn are still positive and, respectively, significant and insignificant.
The results of Model 6 are more similar to its fixed effect model than Model 5. In fact, all the coefficients of Model 6 are very similar to Model 4. IVC is still positive and significant at 1%. IVC_pre, Syn_pre, Syn are still positive but insignificant. CVC_pre and CVC are still negative and insignificant. The only slight difference between the two models is that all the coefficients, in the random effect model (Table 10), have decreased by approximately 10%.
To conclude, these outcomes are in line with the previous results, so Models 2 and 4 can be classified as robust. In addition, the hypotheses are confirmed.
5. Discussion and conclusion
The rate of clean-tech VC deals, as a percentage of all VC investments, has increased from approximately 1% in 1996 to over 10% in 2010 (Cumming et al., 2016). However, few studies in the field of VC have addressed the relation between clean-tech companies and the different composition of VC investments. Thus, drawing from the agency theory and the resource-based view theory, this Thesis offers new insights into the financial performance of VC-backed clean-tech companies. The research has tried to answer to the research question, “what investment form between IVC, CVC, and IVC-CVC syndication has the best financial performance in the clean-tech companies?”. Estimated results have shown that IVC has the best financial performance in the clean-tech industry.
In addition, it was possible to disentangle the screening and the value-adding effect of VCs. Two different results were found. In the case of total assets as dependent variable, the net effect of IVC is a pure value addition effect. Hence, strategic and management activities and valuable resources provided by IVC contribute to an improvement in the performance of the IVC-backed firm. On the other hand, the net effect of IVC on IPO is a pure screening effect.
29 On the other hand, the coefficients of IVC-CVC syndication are positive in all the models. However, this type of VC investment does not significantly impact the financed clean-tech company. The positive effect linked to the resource-based view and the negative effect linked to the agency costs cancel each other out. Consequently, there is neither a positive, nor a negative effect.
5.1 Practical implications
Implications for venture capitalists and entrepreneurs emerge from this study. From the perspective of the IVC, he or she has to take into consideration the difference between short-term and long-term performance. Thus, if a IVC consider more important the short-term, he or she has to focus and to supply valuable resources and relevant activities. On the other hand, if IVC is more interested in the long-term performance, he or she has to focus on the selection of the firms with high growth rate before the investment.
Regarding the entrepreneur, this Thesis has shown that if a clean-tech entrepreneur wants to increase the financial performance of the company, he or she has to consider only IVC investments. In fact, other types of VC can lead to several agency conflicts that emerge when the two parties have different risk perceptions or goals.
5.2 Limitations and future research
Within the context of this Thesis, readers need to take into account several limitations to preserve the reliability of the results. These limitations can lead to future research. First of all, data collection presented several issues. The two databases employed, Orbis and Zephyr, had a lot of missing data. In particular, a lot of financial data of the VC-backed clean-tech firms was not available. Total assets was chosen as performance indicator since it was the one with least missing data. However, some clean-tech companies, without this index, were dropped. In addition, this financial indicator was not available for all the years of interest. In some cases, the financed companies had all the data before the investment but not after, and in other cases the other way around. Thus, all these limitations regarding the databases led to an unbalanced panel dataset.
30 databases, like Thomson one, in order to have a more balanced number of IVC, CVC and syndication for the sample.
In addition, another area of future research could be the analysis of the strategic objectives of CVC. Since the results have shown that CVC might negatively impact the financial performance of the CVC-backed companies, it could be useful to research if this type of VC has, at least, a positive effect on some strategic goals, such as knowledge acquisition or innovation improvement.
31
6. References
van Aken, J., Berends, H., & Van der Bij, H. (2012). Problem solving in organizations: A methodological
handbook for business and management students. Cambridge University Press.
Alvarez‐Garrido, E., & Dushnitsky, G. (2016). Are entrepreneurial venture's innovation rates sensitive to investor complementary assets? Comparing biotech ventures backed by corporate and independent VCs. Strategic Management Journal, 37(5), 819-834.
Amit, R., Brander, J., & Zott, C. (1998). Why do venture capital firms exist? Theory and Canadian evidence. Journal of business Venturing, 13(6), 441-466.
Arthurs, J. D., & Busenitz, L. W. (2003). The boundaries and limitations of agency theory and stewardship theory in the venture capitalist/entrepreneur relationship. Entrepreneurship Theory and Practice, 28(2), 145-162.
Bayar, O., & Chemmanur, T. J. (2011). IPOs versus acquisitions and the valuation premium puzzle: a theory of exit choice by entrepreneurs and venture capitalists. Journal of Financial and Quantitative
Analysis, 46(6), 1755-1793.
Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of management, 17(1), 99-120.
Barry, C. B., Muscarella, C. J., Peavy, J. W., & Vetsuypens, M. R. (1990). The role of venture capital in the creation of public companies: Evidence from the going-public process. Journal of Financial
economics, 27(2), 447-471.
Bottazzi, L., Da Rin, M., & Hellmann, T. (2008). Who are the active investors?: Evidence from venture capital. Journal of Financial Economics, 89(3), 488-512.
Brander, J. A., Amit, R., & Antweiler, W. (2002). Venture‐capital syndication: Improved venture selection vs. the value‐added hypothesis. Journal of Economics & Management Strategy, 11(3), 423-452.
Burtis, P. R., Epstein, B., Parker, N., & O’Rourke, A. (2006). California's Clean-tech Industry: Annual Venture Capital Investment Update 2006. Environmental Entrepreneurs and Clean-tech Venture
Network LLC.
32
Casamatta, C., & Haritchabalet, C. (2007). Experience, screening and syndication in venture capital investments. Journal of Financial Intermediation, 16(3), 368-398.
Chan, Y. S. (1983). On the positive role of financial intermediation in allocation of venture capital in a market with imperfect information. The Journal of Finance, 38(5), 1543-1568.
Chemmanur, T. J., Krishnan, K., & Nandy, D. K. (2011). How does venture capital financing improve efficiency in private firms? A look beneath the surface. The Review of Financial Studies, 24(12), 4037-4090.
Chemmanur, T. J., Loutskina, E., & Tian, X. (2014). Corporate venture capital, value creation, and innovation. The Review of Financial Studies, 27(8), 2434-2473.
Cherry, M. A., & Sneirson, J. F. (2010). Beyond profit: Rethinking corporate social responsibility and greenwashing after the BP oil disaster.
Chesbrough, H. W. (2002). Making sense of corporate venture capital. Harvard business review, 80(3), 90-99.
Christensen, E., Wuebker, R., & Wustenhagen, R. (2009). Of acting principals and principal agents: goal incongruence in the venture capitalist-entrepreneur relationship. International Journal of
Entrepreneurship and Small Business, 7(3), 367-388.
Christofidis, C., & Debande, O. (2001). Financing innovative firms through venture capital. EIB Sector
Papers.
Colombo, M. G., & Murtinu, S. (2017). Venture capital investments in Europe and portfolio firms' economic performance: independent versus corporate investors. Journal of Economics &
Management Strategy.
Croce, A., Martí, J., & Murtinu, S. (2013). The impact of venture capital on the productivity growth of European entrepreneurial firms:‘Screening’or ‘value added’effect?. Journal of Business Venturing, 28(4), 489-510.
Cumming, D., Fleming, G., & Suchard, J. A. (2005). Venture capitalist value-added activities, fundraising and drawdowns. Journal of Banking & Finance, 29(2), 295-331.
33
Denis, D. J. (2004). Entrepreneurial finance: an overview of the issues and evidence. Journal of
corporate finance, 10(2), 301-326.
Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management
review, 14(1), 57-74.
Ghosh, S., & Nanda, R. (2010). Venture capital investment in the clean energy sector.
Gifford, S. (1997). Limited attention and the role of the venture capitalist. Journal of business
venturing, 12(6), 459-482.
Gompers, P. A. (1996). Grandstanding in the venture capital industry. Journal of Financial economics, 42(1), 133-156.
Gompers, P. A. (2002). Corporations and the financing of innovation: the corporate venturing experience. Economic Review-Federal Reserve Bank of Atlanta, 87(4), 1.
Gompers, P., Kovner, A., & Lerner, J. (2009). Specialization and success: Evidence from venture capital. Journal of Economics & Management Strategy, 18(3), 817-844.
Gompers, P., & Lerner, J. (1998). Venture capital distributions: Short‐run and long‐run reactions. The
Journal of Finance, 53(6), 2161-2183.
Gompers, P., & Lerner, J. (2000). The determinants of corporate venture capital success: Organizational structure, incentives, and complementarities. In Concentrated corporate ownership (pp. 17-54). University of Chicago Press.
Gompers, P., & Lerner, J. (2001). The venture capital revolution. The journal of economic
perspectives, 15(2), 145-168.
Gompers, P. A., & Lerner, J. (2004). The venture capital cycle. MIT press.
Gorman, M., & Sahlman, W. A. (1989). What do venture capitalists do?. Journal of business
venturing, 4(4), 231-248.
Hahn, S., & Kang, J. (2017). Complementary or conflictory?: the effects of the composition of the syndicate on venture capital-backed IPOs in the US stock market. Economia e Politica Industriale, 44(1), 77-102.
34
Jääskeläinen, M. (2012). Venture capital syndication: Synthesis and future directions. International
Journal of Management Reviews, 14(4), 444-463.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Kaplan, S. N., & Stromberg, P. (2001). Venture capitalists as principals: Contracting, screening, and
monitoring (No. w8202). National Bureau of Economic Research.
Katila, R., Rosenberger, J. D., & Eisenhardt, K. M. (2008). Swimming with sharks: Technology ventures, defense mechanisms and corporate relationships. Administrative Science Quarterly, 53(2), 295-332. Kukalis, S. (2010). Agglomeration economies and firm performance: the case of industry clusters. Journal of Management, 36(2), 453-481.
Lahovnik, M. (2008). Corporate Governance In Relation To The Enhancement of Social Responsibility. Journal of Diversity Management–Second Quarter, 3(2).
Lerner, J. (1994). Venture capitalists and the decision to go public. Journal of financial
Economics, 35(3), 293-316.
Lewis, M. (2007). Stepwise versus Hierarchical Regression: Pros and Cons. Online Submission.
LiPuma, J. A. (2006). Independent venture capital, corporate venture capital, and the internationalisation intensity of technology-based portfolio firms. International Entrepreneurship and
Management Journal, 2(2), 245-260.
Macmillan, I., Roberts, E., Livada, V., & Wang, A. (2008). Corporate venture capital (CVC): Seeking innovation and strategic growth. National Institute of Standards and Technology, US Department of
Commerce.
Masulis, R. W., & Nahata, R. (2009). Financial contracting with strategic investors: Evidence from corporate venture capital backed IPOs. Journal of Financial Intermediation, 18(4), 599-631.
Maula, M. V. (2001). Corporate venture capital and the value-added for technology-based new firms. Helsinki University of Technology.
Maula, M., Autio, E., & Murray, G. (2003). Prerequisites for the creation of social capital and subsequent knowledge acquisition in corporate venture capital. Venture Capital: An International
35
Maula, M., Autio, E., & Murray, G. (2005). Corporate venture capitalists and independent venture capitalists: What do they know, who do they know and should entrepreneurs care?. Venture Capital:
An International Journal of Entrepreneurial Finance, 7(1), 3-21.
McNally, K. (2002). Corporate venture capital: Bridging the equity gap in the small business sector. Routledge.
Meuleman, M., Lockett, A., Manigart, S., & Wright, M. (2010). Partner selection decisions in interfirm collaborations: The paradox of relational embeddedness. Journal of Management Studies, 47(6), 995-1019.
Nanda, R., Younge, K., & Fleming, L. (2014). Innovation and entrepreneurship in renewable energy. In The changing frontier: Rethinking science and innovation policy (pp. 199-232). University of Chicago Press.
Penrose, E. T. (1959). The theory of the growth of the firm. New York: Sharpe.
Pernick, R., & Wilder, C. (2007). The clean tech revolution. The next big growth and investment
opportunity. New York.
Peteraf, M. A. (1993). The cornerstones of competitive advantage: a resource‐based view. Strategic
management journal, 14(3), 179-191.
Puri, M., & Zarutskie, R. (2012). On the life cycle dynamics of venture‐capital‐and non‐venture‐capital‐ financed firms. The Journal of Finance, 67(6), 2247-2293.
Reid, G. C. (1999). The application of principal-agent methods to investor-investee relations in the UK venture capital industry. Venture Capital: An International Journal of Entrepreneurial Finance, 1(4), 285-302.
Rosenstein, J., Bruno, A. V., Bygrave, W. D., & Taylor, N. T. (1993). The CEO, venture capitalists, and the board. Journal of Business Venturing, 8(2), 99-113.
Sahlman, W. A. (1990). The structure and governance of venture-capital organizations. Journal of financial economics, 27(2), 473-521.
Sapienza, H. J., Amason, A. C., & Manigart, S. (1994). The level and nature of venture capitalist involvement in their portfolio companies: a study of three European countries. Managerial
36
Smith, G. (2001). How early stage entrepreneurs evaluate venture capitalists. The Journal of Private
Equity, 4(2), 33-45.
Tian, X. (2011). The role of venture capital syndication in value creation for entrepreneurial firms. Review of Finance, 16(1), 245-283.
Tian, X., & Wang, T. Y. (2011). Tolerance for failure and corporate innovation. The Review of Financial
Studies, 27(1), 211-255.
Wernerfelt, B. (1984). A resource‐based view of the firm. Strategic management journal, 5(2), 171-180.
37
Appendix
Table 2. Descriptive statistics
Table 9. Model 5, Random effects with Total assets as dependent variable (N=11186)
Note: *** Represents statistical significance at 1% ** Represents statistical significance at 5% * Represents statistical significance at 10% Marginal effects; Standard errors in parentheses
Obs. Mean Std. Dev. Min. Max. Skewness Kurtosis
38 Table 10. Model 6, Random effects with Total assets as dependent variable (N=11186)
Multiple regression Variables Coefficients IVC_pre 0.1467 (0.1813) CVC_pre -0.5106 (0.4421) Syn_pre 0.2508 (0.2601) IVC 0.6241*** (0.2121) CVC -0.2477 (0.3563) Syn 0.3249 (0.2805) Age_log 0.7994*** Year Dummies Country Dummies (0.0704) YES YES
