Factors influencing Venture Capital Firms preferences regarding
industry and geographic diversification of their investments with the
focus on US at the year of 2015
Amsterdam Business School
Name Ting Lin
Student number 11089679
Program Economics & Business Specialization Finance
Number of ECTS
Supervisor dr. I.J. (Ilko) Naaborg Target completion 07/2016
Statement of Originality
This document is written by Student Ting Lin who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Based on Gupta’s paper Determinants of Venture Capital Firms’ Preferences Regarding The Industry Diversity And Geographic Scope Of Their Investments in 1992, this paper examines the questions of why venture capital firms prefer varying degrees of industry diversify and geographic scope in their investments. The article hypothesis the variations in VCFs’ preferences are a function of the preferred financing stage of ventures, the ownership structure of the firms, the size of the firm, general partners’ background & experience levels and the degree of syndication of the firm’s investments. We use data of 200 US venture capital firms’ profile in five states (California, Massachusetts, New York, Pennsylvania and Texas---the five most concentrated areas of venture capital activity) of 2015. By adding more theories into the literature review part and regressing the original data rather than the second hand ones, it tests and complete Gupta’s model. The findings can be summarized as follows: (1)It supports Gupta’s paper basically besides two points. Firstly, we found that financial or independent VCFs enjoy a broader industry diversity than those are subsidiaries of non-financial institutions while there is weak evidence showing this relevance in their research. Secondly, we found no correlation between VCFs’ size and their preferences regarding industry diversify. (2) Venture capital firms with higher general partners’ background and experience levels prefer to invest in a more diverse industry than is the case with other VCFs; However, there are no
differences in preferences regarding geographic scope. (3) Venture capital firms which has higher degree of syndication in its investments will invest in a
narrower diverse set of industries and geographic scope relative to other venture capital firms.
The paper contributes to the literature in two ways. First, we extend the literature by completing Gupta’s model and adding two more new factors, General partners’ background & experience levels and the degree of syndication of VCF to the model. Second, we test the predictions of the model using an original dataset.
The paper will be divided into four parts. In the literature review, I will show you in detail about what Gupta did and found in 1992 and what is my contribution to this field. And then explain the theoretical background of the model. In
hypothesis, methodology & data part, I will explain the methodology I use and list data sources. After that, the empirical results will be analyzed and a
robustness check will be given as well. Lastly, a conclusion and a discussion of the results can be made.
Keywords:
Venture capital firms, investment preference, industry diversification, geographic scope
Table of Contents
1. Introduction ... 5
2. Literature review ... 8
3. Methodology and Data ... 18
3.1. Methodology ... 18
3.2. Data and descriptive statistics ... 19
4. Analysis ... 25
4.1. Empirical Results ... 25
4.2 Robustness check ... 31
5. Conclusion and discussion ... 33
1. Introduction
Diversification is the process by which firms depart from their core competencies to enter new markets and new technologies. (Texier, Francois, 2000)
Diversification is used as a common strategy by investments firms. According to Calori and Harvatopoulos (1988), there are two dimensions of rationale for diversification. The first one relates to the nature of the strategic objective: Diversification may be defensive or offensive. Defensive reasons may be spreading the risk of market contraction, or being forced to diversify when current product or current market orientation seems to provide no further opportunities for growth. Offensive reasons may be conquering new positions, taking opportunities that promise greater profitability than expansion
opportunities, or using retained cash that exceeds total expansion needs. The second rationale involves the expected outcomes of diversification: management may expect great economic value with their current activities.
There are many types of diversification, for example, industry diversification, geographic (international) diversification, product diversification, development stage diversification and time diversification (Mike W. Peng & Andrew Delios, 2006; Margarethe F. Wiersema & Harry P. Bowen, 2007; April Knill,2009). We focus on the first two dimensions in this paper.
In strategic management field, plenty of the literature examines the effect of diversification on the firm’s performance. Stan Xiao Li and Royston Greenwood (2004) examined the effect of diversification upon intra-industry performance
and proposed that intra-industry diversification promised three sets of benefits, which, separately and in combination, provide firms with a competitive
advantage: synergies arising from economies of scope; premiums from mutual forbearance enabled by multi-market competition; and efficiencies derived from market structuration. Hu seyi̇n Tanriverdi̇ and Chi-Hyon Lee(2008) found out that in the presence of network externalities, complementary related diversification strategies in production and consumption can be critical for achieving positive returns to within-industry diversification. By doing research on Japanese firms, Andrew Delios and Paul W. Beamis (1999) found out geographic scope is positively related to the firms profitability.
Venture capital is financing that investors provide to startup companies and small businesses that are believed to have long-term growth potential. For startups without access to capital markets, venture capital is an essential source of money. In venture capital field, diversification strategy of investments had long been studied by many scholars. While diversification usually brings positive effects for venture capital firms, it has a delaying impact on VC exit because diversification could entail considerable time and expense and Time absorbed in involvement is time taken away from remaining clients (i.e., PCs), aswell as new investment due diligence, holding staff constant (April Knill,2009). However, most of the studies focus on the performance of different levels of diversification strategies (Gompers, Kovner and Scharfstein, 2010; Robert, Alessandro and Federico, 2012) and few mentions the factors influence the investment
degree of industry diversity and geographic scope in their investments and what factors influence their preference?
As we know, the venture capital industry developed rapidly since the 1990s. According to the MoneyTree™ Report from PricewaterhouseCoopers LLP (PwC)
1and the National Venture Capital Association (NVCA), the venture capital
industry deployed $58.8 billion across the United States in 2015, marking the second highest full year total in the last 20 years. Compared with the data of $4 million in the time of 1990s, venture capital investments, known as the “money of invention” (Black and Gilson 1998; Kortum and Lerner 2000), boomed a lot for the last 20 years. It attracted a lot of attention and it is predicted to continue increasing in the future. Thus, given this importance, research on how VCFs attempt to create a portfolio represents a significant research meaning. What’s more, the research result will help new VCFs to grow healthier and better. The way how experienced VCFs attempt to create portfolios can help new venture capitalist make better investments decisions.
This research paper is based on Gupta and Sapienza’s paper Determinants of venture capital firms’ preference regarding the industry diversify and geographic scope of their investments (1992). They present three factors influencing
venture capital firm’s investment preferences: the preferred financing stage, the institutional origins of the VCF, the size of the VCF and the primary source of financing used by the VCF. I will present you with two more new factors: i.e. General partners’ background & experience levels and the degree of syndication of VCFs that influence the diversification strategies of VC firms, and also, examine
whether the factors they put forward more than 20 years ago still works for today’s VCFs.
The paper will be divided into four parts. Chapter 2 is the literature review. I will show you in detail about what Gupta did and found in 1992 and what is my contribution to this field. And then explain the theoretical background of all the factors that influence Venture Capital’s investment preference in Chapter 2.1, 2.2, 2.3, 2.4 and 2.5 respectively. In Chapter 3, I will explain the methodology I use and list data sources so that the readers can have some feelings for the data. In Chapter 4, the empirical results will be analysed and a robustness check will be given as well. Lastly in Chapter 5, a conclusion and a discussion of the results can be made.
2. Literature review
Lots of researches has been done on factors determining VCFs’ investment strategies. Mayer et al. (2005) show that bank-backed VC funds invest generally in late stage, domestic activities, whereas corporate- and individual-backed VC firms invest in early stage, high-technology activities globally rather than domestically. Hana Milanov, Dimo Dimov and Dean A. Shepherd (2006) emphasize a configural view of resources (resource allocation), and tested a three-way interaction of VCFs’ legitimacy, network status and structural holes to explain investment diversification across industries. Organizations will be considered cognitively legitimate only when they are able to rationally account for organizational actions and when they become reliable (Hannan and
Freeman,1984). A firm’s network status is inferred by simultaneously observing its position in the network and accounting for the status of its partners as well.
High status is especially valuable in high uncertainty situations when no other signals of organizational quality can be observed (Podolny,2001) and high status companies are able to borrow money at a lower cost (Podolny,1993). Structural hole is the “network space” between 2 firms who are not mutually connected and do not share the same contacts and argued for three forms of informational benefits: access, referrals and timing (Burt,1992). They tested hypotheses on a longitudinal data set of 207 US-based VCFs over a 15-year period, covering 980 firm-year data points and found support for a configural view of investment diversification based on bundles of external resources. Kangmao Wang, Clement K. Wang and Qing Lu(2002) discovered independent and finance-affiliated VCFs have significant differences in industry preference, to be specific, there are more high-technology companies backed by independent VCFs.
The only paper published related to the determinants of venture capital
investment preference regarding to diversification is Gupta’s paper in 1992. By using data of 169 venture capital firms drawn from the 1987 edition of Pratt’s guide to venture capital sources and multiple linear regression, they found out four factors that matter. The first factor is the development stage of target ventures. They found that VCFs that specialize in early stage ventures tend to prefer less industry diversity and a narrower geographic scope relative to those that invest in late stage ventures. The second factor is venture capital firm’s ownership structure. VCFs that are subsidiaries of non-financial corporations tend to prefer less industry diversity but a broader geographic scope relative to
other VCFs. The third factor is size of the venture capital firm. Large VCFs tend to prefer greater industry and a broader geographic scope relative to smaller VCFs. The last factor is private versus public sources of funds. VCFs that provide SBIC (the Small Business Investment Company) financing prefer a narrower
geographic scope that is the case with that rely exclusively on private sector financing.
In this paper, I will test the first three factors (i.e. development stage of target ventures, venture capital firm’s ownership structure and size of the venture capital firm) of Gupta’s paper in 19922 and add two more new factors, that is,
general partners’ background and experience levels and degree of syndication. I will explain the factors in the following one by one.
2.1 Development stage of target ventures
Different companies have different stage of development, At each stage of development, a company has different available resources, as well as varying needs, investor expectations and levels of risk. For venture capital firms, Early stage investments involve commitments of funds to firms with little more than a business plan(seed stage) or an initial prototype and some market studies (first stage). (Edgar Norton, 1994). Some venture capital firms take a specialized approach, focusing on one key phase of the lifecycle of a growing company while some venture capital investors use a diversified approach, providing initial investment to companies at different stages in the financing lifecycle (for
example, they may invest 25% in startups, 50% in growth-stage companies and
25% in later-stage companies). The focus development stage of ventures is an important characteristics of venture capital firms. For convenience we generally divide the stage of ventures as “early” and “later” in our paper.
A significant stream of research has emphasized the “coach” role that VCF performs (Baum & Silverman, 2004). Usually, early ventures need VCFs to provide them with more industry and market knowledge, strategic advice and managerial expertise compared to later stage ventures because such firms are often in need of complementing internal competencies (Innovation and the contributions from venture capital, 2006). It costs more time and energy for monitoring and involvement. If a venture capital firm invest mostly in early stage ventures, given these complicated and extra requirements as being a “coach” rather than a “pure investor”, it will usually focus on fewer industries and
geographic scope than those invest mostly on late stage ventures. The conclusion also based on a theory in human ecology. In human ecology, the law of distance interaction states that the probability of interaction between social elements declines as a multiplicative function of the distance between them (Hawley 1971). Sociologists believe this law arises in large part because the costs of interacting—including finding and screening exchange partners and maintaining relationships—increase with distance (Zipf 1949). So the first hypothesis can be advanced:
H1: VCFs which focus on early stage ventures tend to invest in a less diverse set of industries and geographic scope than those focus on later stages.
2.2 Venture capital firm’s Ownership Structure
In terms of venture capital firms’ ownership structure, besides independently owned companies which accounts for a large amount of VCFs, there exists certain amount of corporate VCs (CVC). The term corporate venture capital is used to describe the investment of corporate funds directly in external start-up
companies (Henry W. Chesbrough, 2002). The defining feature of CVCs is their close affiliation with large established industrial corporations (Vladimir I. Ivanov, Fei Xie, 2008). They can leverage the assets and capabilities of their parents, using the intro-firm information network and so on. Many large corporations go into venture capital arena for some certain strategic needs. It was reported that in 2015, corporate venture participated in about one out of five deals in the United States or Europe, and one out of three deals in Asia. Founders will
increasingly study how to attract and engage these deep pocketed investors. That creates greater competition for traditional financial VCs to differentiate and prove their value to entrepreneurs. These investments are made primarily to increase sales and profits of the corporation’s own businesses. A company making a strategic investments seeks to identify and exploit synergies between itself and a new venture (Henry W. Chesbrough, 2002). Generally, CVCs consider financial return and liquidity to be less important than independently owned companies (Siegel et al.’s, 1988). They get rewarded in many other ways than pure financial returns--including creating stronger suppliers, putting control levers in their industry, testing products, de-risking innovation, and engineering less expensive acquisitions.
It can be expected that for these strategic benefits, corporate PEs will invest in industry segments related to the corporation’s current base of activities. Further, the geographic locations of the new venture for corporate PEs can be more diversified than independent PEs because large corporations themselves tend to be more geographically dispersed. So the second hypothesis can be given:
H2: Venture capital firms which are backed by corporations will invest in a less diverse set of industries and geographic scope than those independently owned VCFs.
2.3 Size of the venture capital firm
As in Pratt and Morris (1987)’s paper, statistics on US indicates considerable variance in the total size of the capital under management by various VCFs (from less than $1 million to over $500 million, but it is even larger today. In our
sample, it ranges from $0.05 million to $57.600 million. VCFs with large amount of capital under management are usually more experienced and assumed to be of great base of capabilities in attracting and evaluating opportunities and
providing assistance to the target ventures (Gupta, Sapienza,1992). The size of VCFs are related to its preference in industry and geographic scope diversity for several reasons. Firstly, the larger the size of the firm is, the more investment opportunities it can encounter due to large and complicated networks. So it is more likely for large size VCFs to invest in more diverse industries and
geographies. Secondly, large size VCFs tend to choose more diversification to diversify risks for bigger pool of capital than small VCFs. Thirdly, the previous experience in different industries and communications with ventures in various
geographic scopes give the firm confidence and capabilities to cope with a wider set of investment opportunities. So the third hypothesis can be stated as:
H3: Venture capital firms with a larger pool of capital under management will prefer venture investment opportunities within a more diverse industry and a broader geographic scope than will other venture capital firms.
2.4 General partners’ background and experience levels
The possession of general VC expertise complements industry-specific
knowledge and is potentially more valuable, especially to entrepreneurs without managerial experience. Venture capitalists have three value-added roles,
screening, monitoring and advising. Venture capital firms with experienced partners are more likely to be actively involved in the start-up firms. General partners’ background and experience levels can be measured by the number of funds the company successfully raised (Douglas Cumming, April Knill, 2011). It can also be examined by counting the number of years being a venture capitalist (Garry D.Bruton, Sophie Manigart, Vance Fried, Harry J.Sapienza, 2002). I am going to use the first method to measure this factor.
Bygrave(1987) provides evidence that the venture capitalist’s expertise provides access to information and deal networks. A venture capitalist’s prior experience in a particular industry should affect the extensiveness of the venture capitalist’s personal contact network among entrepreneurs and other investors in that industry. Having many contacts in turn facilitates the identification of new investment opportunities. In addition to identifying investment opportunities,
venture capitalists with deep contact networks in an industry or a geographic area can often better assess the veracity of the information they receive about the quality of an investment opportunity (Olav Sorenson and Toby E. Stuart, 2001). Venture capitalists are important parts in networks and are furthermore in between, and central to, several different types of networks. In Florida and Kenney(1988) these networks are grouped in four. That means the deeper the general partners’ background and experience levels a VCF has, the wider its network is and the more opportunities it may encounter. So we have the following forth hypothesis:
H4: venture capital firms with higher general partners’ background and
experience levels tend to invest in a more diverse industry and geographic scope.
2.5 Degree of syndication
Prior research has shown that venture capitalists often “coinvest” with others when allocating capital to new ventures (Brander, Amit, & Antweiler, 2002; Bygrave, 1987, 1988; Lerner, 1994; Lockett & Wright, 2003). Syndication involves two or more venture capital firms taking an equity stake in an
investment, either in the same round or, at different points in time. Syndication yields many benefits to VCFs: 1) it enables exchange of information between firms, enhancing the deal flow and discovery prospects for future investments (Bygrave,1987), 2) a syndicate partners’ advice and expertise can help a focal VCF make better judgments in evaluating a deal, enhance its decision making (Lerner,1994) and complement VCFs’ capabilities to add value to portfolio companies; and 3) syndication networks affect transaction patterns of exchange
among economic actors and in turn enhance VCFs’ ability to overcome boundaries in entering new markets (Sorenson and Stuart,2001). So we can forward the hypothesis as following:
H5a: Venture Capital firms which has higher degree of syndication in its
investments will invest in a wider diverse set of industries and geographic scope.
However, as Smith et al. (1995, p. 19) point out, ‘it is unlikely that any single theory can fully explain the complexities of cooperation’. Although on the one hand, when embedded in multi-firm alliances, firms have to consider the actions and preferences of other group members to determine their own decisions (Wageman, 1995). On the other hand, ambiguity in the anticipated outcomes of cooperation may induce firms to rely on their own, rather than their partners’, strengths when deciding where to allocate their resources (Gifford, 1997; Pfeffer and Salancik, 1978). That is to say, although given syndicate partner’s advice and expertise in many other different industries, the focal VCF may still prefer to invest in limited industries and choose in certain geographic scope it most familiar with.
Another possible explaination is, there exists ‘lead’ investors in a syndication group. Lead investors have been defined variously as the VCF that originates the deal (Gorman and Sahlman, 1989), the ‘most important’ investor (Sapienza, 1992), or the investor with the greatest financial stake in the venture (Wright and Lockett, 2003). Regardless of the definition, consensus states that some investors play a more active role in the PFC than other syndicate members (e.g.
Gorman and Sahlman, 1989). De Clercq, Sapienza, and Zaheer(2008) show that VCF involvement is negatively affected by its own reputation and the reputation of the other syndicate members but positively influenced by its financial stake, relative to that of the syndicate group (The summary of their findings appears in Appendix I). Assume that the lead venture capitalist makes the syndication decision, VCF’s final decision regarding industry diversity and geographic scope may follow the preferences of the lead investor and has little to do with the degree of syndication.
H5b: Venture capital firms’ degree of syndication has no relation with their preference regarding industry diversity and geographic scope.
There is also another point of view. Hans Bruining and Ernst Verwaal (2005) found out transaction costs of the syndicate governance are likely to increase complexity with the number of partners in the syndicate. Smaller venture capital firms have smaller amounts available for investment and a smaller portfolio scale and scope. Ex post transaction costs may rise substantially with a more diverse and larger number of syndicate members involved. And high transaction costs may lead to narrower diverse set of industries and geographic scope
investments.
H5c: Venture capital firms which has higher degree of syndication in its investments will invest in a narrower diverse set of industries and geographic scope.
3. Methodology and Data 3.1. Methodology
The research question is whether and how the five factors influence the venture capital’s investments regarding industry diversity and geographic scope.
To answer this question, first, we run a zero-order correlations among all
variables of interest and then we run a cross-sectional regression of the following model. We will also compare the results with and without controls.
Regression formula
𝑮𝑬𝑶𝑺𝑪𝑶𝑷𝑬𝒋= 𝛼0+ 𝛽1STAGE𝑗+ 𝛽2𝑂𝑊𝑁𝐸𝑅𝑆𝐻𝐼𝑃𝑗+ 𝛽3SIZE𝑗+ 𝛽4PARTNER𝑗 + 𝛽5𝑆𝑌𝑁𝐷𝐼𝐶𝐴𝑇𝐼𝑂𝑁𝑗+ 𝛽5𝐶𝑂𝑁𝑇𝑅𝑂𝐿𝑆 + 𝜀𝑗3
𝑰𝑵𝑫𝑼𝑺𝑫𝑰𝑽𝒋 = 𝛼0+ 𝛽1STAGE𝑗+ 𝛽2𝑂𝑊𝑁𝐸𝑅𝑆𝐻𝐼𝑃𝑗+ 𝛽3SIZE𝑗 + 𝛽4PARTNER𝑗 + 𝛽5𝑆𝑌𝑁𝐷𝐼𝐶𝐴𝑇𝐼𝑂𝑁𝑗+ +𝛽5𝐶𝑂𝑁𝑇𝑅𝑂𝐿𝑆 + 𝜀𝑗
For robustness check we will first use logs of the dependent variables to pull outlying data from a positively skewed distribution closer to the bulk of the data in a quest to have the variable be normally distributed and secondly we are going to divide the sample into two groups, one group for larger venture capital firms whose capital under management is over 30 million and another for small
3In Gupta’s paper Determinants of venture capital firm’s preferences regarding the industry diversity and geographic scope of their investments, they didn’t give the regression formula. The formula above is what I thought they may use. So I change the independent variables and use it in my paper.
venture capital firms whose capital under management is under 30 million. We can see whether the coefficient remains the same for small subsample.
3.2. Data and descriptive statistics Data
Data for this study were collected from Thomson One database under the firm profile for the year of 2015. By writing programming and collect it one by one, I merged the data from original firm profiles so that they can be used for
regression. Since there may exist significant interregional differences in the characteristics and investment preferences of VCFs, in order to control for the possible effects of VCF location without loosing too much degree of freedom, I chose from VCFs located in California, Massachusetts, New York, Pennsylvania and Texas---the five most concentrated areas of venture capital activity (Yochanan Schachmurove, 2010).
A total of 3655 venture capital firms are first selected. After excluding data unavailable4 situations and choosing a random sample from the total sample, a
total of 200 VCFs were selected for this study.
Measures
Diversification by industry/geographic scope(INDUSDIV/GEOSCOPE)
The extent to which the VCFs’ investments were concentrated in particular industries and geography scope can be assessed by calculating a
4 VCFs with only one investment were deleted.
VCFs with unavailable capital under management were deleted. VCFs with unclear stage preferences were deleted.
Hirschman index(HHI)(Dimov, 2006). By calculating an HHI measure on the distribution of the VCFs’ investments across industries and geographic scopes, we estimated the degree to which the VCF specialized across different industries and geographic scopes.
HHI = ∑ 𝑝𝑖2
Where pi represents the proportion of investments made in a particular industry/geographic scope in a given year.
A similar measurement can be found in Cressy et al.’s paper in 2012. They measured firm j’s diversification by industry as 1- ∑ (𝑁𝑖𝑗
𝑁𝑖)^2 𝐽
𝑗=1 , where Nij denotes
the number of investments of firm i in industry j and Ni is the number of companies in the fund portfolio. Firm j’s diversification by geography = 1-∑ (𝑁𝑖𝑦
𝑁𝑖)^2 𝑌
𝑦=1 , where Niy denotes the number of investments of firm i in country y
and Ni is the number of companies in the fund portfolio. We can tell that this measurement and the measurement above is the inverse number. In order to make the regression result clear and in a way convenient to see, we choose to use the second measurement.
So the number ranged from 0 to 1 in this case, 0 means is extreme non-diversify and 1 means extreme diversify. Higher value in this index imply that the VCF is open to a wider industry or geographic scope of venture investments.
The stage of financing are presented in the “stage breakdown” category in the firm profile. If VCF’s investment preferences for four early stages(seed, R&D, start-up, and first stage) excess the five late stages(second stage, third stage, bridge, acquisition, and leveraged buy-out), we marked it as “STAGE=0”. If not, we marked it as “STAGE=1”. On this basis, 44% of the VCFs were classified as early stage investors while the remaining were classified as late stage investors.
VCF’s ownership structure (OWNERSHIP)
Under the firm type of the profiles, we can conclude the VCF’s ownership structure. It is assigned the value “0” if the VCF was a subsidiary of a non-financial corporation (i.e., Corporate PE/Venture, Government Affiliated Program, Incubator/Development program, Service Provider, University Program) and the value “1” otherwise (i.e., Private Equity Firm, Angel Group, Bank Affiliated and Investment management firm).
Size of the VCF (SIZE)
The size of VCF was measured as capital under management. SIZE ranged from 0.05 million to 57.600 million.
General partner’s background and experience levels(PARTNER)
The number of funds a VC has successfully raised derives this proxy. This proxy implicitly assumes retention of VC management. This assumption should not be problematic as long as venture capital firms are able to hire similarly talented executives to lead their firms. The number ranges from 1 to 47 in this sample.
Degree of syndication (SYNDICATION)
Degree of syndication can be measured as the number of co-investors of this VCF. But since the number of co-investors is not released in the database, I chose to use the total number of investments the co-investors participated divided by the total number of investment this VCF made to calculate the degree of syndication.
summary statistics
Table 1. Summary statistics for total sample
Variable Obs Mean Std. Dev. Min Max
Geoscope 200 0.3880 0.2369 0.0568 1
Indusdiv 200 0.3686 0.2149 0.0868 1
Stage of Finaning (% later stage finaning) 200 0.5600 0.4976 0 1 Ownership (% Financial Corporations) 200 0.8850 0.3198 0 1
Size (Billion) 200 1.3480 4.6496 0.0001 57.6
Number of Funds raised 200 4.8550 5.6171 1 47
Degree of syndication 200 0.5865 0.5010 0 3
Age (years) 200 19.0550 14.8230 1 140
Notes: This table presents summary statistics of the total sample for all two dependent variables: Geoscope and Indusdiv, five independent variables: Stage of financing, Ownership, Size, Number of funds raised and Degree of syndication and one control varibles: Age. Stage of financing and Ownership are reported in percentage terms; Size is reported in billons of dollars.
Table 2. Summary statistics for five states
Mean values Total CA
(N=105) MA (N=20) NY (N=52) PA (N=7) TX (N=16) Geoscope 0.3880 0.4706 0.2688 0.2733 0.1893 0.4542 Indusdiv 0.3686 0.3853 0.4236 0.3253 0.2753 0.3724 Stage of finaning
(% later stage finaning)
0.5600 0.4286 0.7000 0.7500 0.7143 0.5625
Ownership
(% financial corporations)
0.8850 0.8667 0.9000 0.9423 0.5714 0.9375
Size (Billion) 1.3480 0.7758 1.9871 1.3274 2.3818 3.9184 Number of Funds raised 4.8550 5.1619 6.4500 3.7500 3.7143 4.9375 Degree of synidication 0.5865 0.5806 0.5737 0.6384 0.5263 0.4988 Age (years) 19.0550 17.2476 21.9500 18.5577 25.0000 26.3125
Notes: This table presents summary statistics of five different states for all two dependent variables: Geoscope and Indusdiv, five independent variables: Stage of financing, Ownership, Size, Number of funds raised and Degree of syndication and one control variables: Age. Stage of financing and Ownership are reported in percentage terms; Size is reported in billons of dollars.
Table 3. Cross-correlation among the Variable of interest
Geoscope Indusdiv Stage Ownership Size Funds Syndication Age
Geoscope Indusdiv 0.2676 Stage -0.4417 -0.3003 Ownership -0.0319 -0.1672 0.1052 Logsize -0.4632 -0.3659 0.3746 0.0554 Fund -0.1898 -0.2971 0.0094 0.1019 0.4968 Syndication 0.3358 0.4188 -0.2026 0.0185 -0.4453 -0.3460 age -0.1030 -0.2437 0.0530 0.0062 0.3011 0.3382 -0.1478
Table 3 presents cross-correlation among all variables of interest for the total sample of 200 venture capital firms. In statistics, the correlation coefficient r measures the strength and direction of a linear relationship between two variables on a scatterplot. The correlation attempts to draw a line of best fit through the data of two variables, and the correlation coefficient, r, indicates how far away all these data points are to this line of best fit (i.e., how well the data points fit this new model/line of best fit). Its value can range from -1 for a perfect negative linear relationship to +1 for a perfect positive linear relationship. A value of 0 (zero) indicates no relationship between two variables. To interpret these values, see Appendix II. None of the coefficients above indicates significant multi-collinear problems, but there are still some places that should be noticed.
1. The same as in Gupta’s paper, the two dependent variables (GEOSCOPE and INDUSIDIV) are positively correlated (r=0.2676) because some of the explanatory variables are predicted to have a similar effect on both.
2. There is a negative correlation between AGE and SYNDICATION (r=-0.1478). This is consistent with the observations of Sophie Manigart (2002) that young VC firms syndicate more than older VC firms because syndication with respected partners increases their legitimacy and reputation in the VC and in the entrepreneurial community. Furthermore, through syndication young VC firms may seek to build central network positions.
3. There is a negative correlation between SIZE and SYNDICATION (r=-0.4453). This is consistent with the financial motive to syndicate, that is, small VC firms benefit more from syndication as syndication allows them to achieve higher levels of diversification.
4. There is a positive correlation between SIZE and STAGE (r=0.3746), SIZE and FUND (r=0.4968), SIZE and AGE (r=0.3011), indicating that larger VC firms tend to invest in later stage ventures, have deeper experience levels and are usually older than small VC firms.
5. There is a positive correlation between FUND and AGE (r=0.3382), indicating that VC firms which has deeper general partners’ experience levels tend to be older than those lack of experience levels.
4. Analysis
4.1. Empirical Results Table 4 Regression result
Dependent variable INDUSDIV GEOSCOPE (1)Without control variables (2)With control variables(age) (3)With control variables (age, location) (4) Without control variables (5) With control variables(age) (6)With control variables (age, location) STAGE 0.0830*** (2.63) 0.0838*** (2.67) 0.0843*** (2.78) 0.1498 *** (4.91) 0.1496*** (4.89) 0.1055*** (3.61) OWNERSHIP 0.1560** (2.08) 0.1607** (2.26) 0.1579** (2.27) -0.1010 (-1.49) -0.1020 (-1.50) -0.1007 (-1.78) LOG(SIZE) 0.0085 (0.99) 0.0060 (0.70) 0.0023 (0.28) 0.0046*** (3.42) 0.0312*** (3.38) 0.0279*** (3.48) PARTNER 0.0052** (2.56) 0.0038** (1.94) 0.0052** (2.39) 0.0002 (0.10) 0.0005 (0.21) 0.0018 (0.94) SYNDICATION -0.1272*** (-4.88) -0.1286*** (-4.83) -0.1326*** (-5.14) -0.0674*** (-2.27) -0.0671*** (-2.24) -0.0845*** (-2.82) AGE 0.0020** (2.20) 0.0021** (2.24) -0.0004 (-0.29) -0.0008 (-0.70) CA 0.0346 (0.54) -0.0054 (-0.70) MA -0.0439 (-0.58) 0.1639** (-0.09) NY 0.0708 (1.05) 0.1531*** (2.31) PA 0.1183 (1.67) 0.2214*** (2.56) Adj.R-sq 0.2765 0.2923 0.3191 0.2770 0.3223 0.4318 F 18.13 15.47 9.99 14.86 12.46 16.16
This table looks at determinants of Venture capital firm’s (VCF) investments preference regarding industry diversity (INDUSDIV) and geographic scope (GEOSCOPE). Column(1) and (3) are regressions containing the only five factors we focus, while column(2) and (4) are regressions taking control variables(AGE, LOCATION) into consideration. STAGE is a dummy variable that takes the value 1 if late stage ventures(second stage, third stage, bridge, acquisition, and leveraged buy-out) exceed 50% of the total investments. OWNERSHIP is a dummy variable that takes the value 1 if the VCF is a independent or financial related companies (i.e., Private Equity Firm, Angel Group, Bank Affiliated and Investment
management firm). CA will take 1 if the company locates in California. MA will take 1 if the company locates in Massachusetts. NY will take 1 if the company locates in New York. PA will take 1 if the company locates in Pennsylvania. For definitions of all variables see “Measures” in the above paragraph. The regression use the updated data from Thomson ONE. Constants were included in the
regressions but are not reported. *,**,*** indicates significance at 10% 5% and 1% level, respectively. Robust t-statistics are reported in parentheses.
Table 4 contains results of 3 kinds of regressions: one containing only five factors we focus (STAGE, OWNERSHIP, SIZE, PARTNER and SYNDICATION), another contains one control variable (AGE), and the third one contains two control variables (AGE and LOCATION). We can see in both cases, as control variable increases, the adjusted R-square also increase. In first case, R-square goes from 0.2765 to 0.2923 and reaches 0.3191. In second case, the number goes from 0.2770 to 0.3223 and reaches 0.4318. We should also noticed that the control
variables do have some impact on the VCF’s investment preferences: (1)Older VCFs do prefer a wider industry diversity. This is consistent with Gupta’s
findings. (2)VCFs in MA, NY and PA tend to prefer broader geographic scope than VCFs in TX. This is slightly different with Gupta’s findings of CA-based VCFs tend to prefer a narrower and MA-based VCFs a broader geographic scope relative to TX-based VCFs. This can also due to interregional variations in the extent of new venture start-ups but the situations changed over the past two decades.
Tests of Hypothesis 1
Stage of financing do have a significant impact on VCF’s preferences regarding both industry diversity and geographic scope. The beta coefficients for stage are positive and significant: 0.0843 (p<0.001) for effect on preference regarding industry diversity and 0.1055 (p<0.001) for effect on preference regarding geographic scope. That means VCFs that prefer invests in later stage ventures tend to prefer greater industry diversity and broader geographic scope compared with those prefer early stage ventures.
We should notice that in column (2),(3),(5) and (6), when we add control variables, the coefficient is still significant at the 1% level, not statistically different from 0.0830 and 0.1498 in column(1) and (4) respectively.
The result is same as in Gupta’s paper in 1992. The coefficient for stage is
significant at 5% level for effect on preferences regarding industry diversity and at 10% level for effect regarding geographic scope. So we prove the hypothesis to
be right again and the evidence is even more stronger since the coefficients are significant at the 1% level.
Test of Hypothesis 2
Hypothesis 2 predict that venture capital firms which are backed by
corporations will invest in a less diverse set of industries and geographic scope than those independently owned VCFs. The result of beta coefficient for
INDUSDIV is positive 0.1579 (p<0.05). However, the effect for GEOSCOPE is relatively weak with beta coefficient of -0.1007 (p<0.1). We can conclude that financial or independent VCFs enjoy a broader industry diversity than those are subsidiaries of non-financial institutions but there is weak difference in
geographic scope. And it should also be noted that this variable has very low variability: only 3.5% of the total sample are subsidiaries of non-financial corporations.
The coefficient changes very slightly as control variables are added. The results are similar to results of Gupta’s paper. In their paper, the hypothesis receives weak support in both cases. This difference may due to classification method between this paper and Gupta’s paper. Since nowadays VCFs have more types of ownership structure (e.g. Government Affiliated Program, University program). We do not classify these VCFs as subsidiaries of non-financial corporations because they don’t meet the characteristics of it.
Hypothesis 3 predicts that larger VCFs will prefer greater industry diversity and broader geographic scope than will smaller VCFs. The results in table 4 supports the latter half part of the hypothesis but not the first half part: beta for the effect of SIZE on GEOSCOPE is 0.0279 (p<0.01) while beta for the effect of SIZE on INDUSDIV is 0.0023 (p>0.1). While in Gupta’s paper, results support hypothesis in both cases. The empirical evidence supports the theoretical proposition that there is a trade-off between VC assistance to entrepreneurial firms in the venture capitalist’s portfolio and the size of the portfolio (Kanniainen and Keuschnigg, 2000). The increase in size burdens VC assistance to entrepreneurial firms in different industries. It may also due to the bigger size range of our sample compare to that of Gupta’s sample. Size ranges from $1 million to $500 million in Gupta’s paper while it ranges from $0.05 million to $57600 million in our
sample.
Test of Hypothesis 4
The results shows VCF with strong general partner’s background and experience levels tend to invest in broader industry diversity: beta equals to 0.0052
(p<0.05). However, we made no prediction concerning the effect of general partner’s background and experience levels on preferences regarding geographic scope.
Test of Hypothesis 5
The regression result in table 4 supports H5c, that is, venture capital firms which has higher degree of syndication in its investments will invest in a narrower diverse set of industries and geographic scope: beta for the effect of
SYNDICATION on GEOSCOPE is -0.0845 (p<0.01) while beta for the effect of SYNDICATION on INDUSDIV is -0.1326 (p<0.01).
4.2 Robustness check
As for robustness check, we can check if the parameter of interest changes when including other relevant variables in most cases. In the first case that seems supported, in the second less so.
In table 5 we provide more various robustness checks of the baseline regression formula in column (3) and (6) of Table4.
Table 5 Robustness check
Dependent variable INDUSDIV GEOSCOPE Withcontrols (1) Log(INDUSIDIV) (2) Large VCFs (3) Withcontrols (4) Log(GEOSCOPE) (5) Large VCFs (6) STAGE 0.0843*** (2.78) 0.2391*** (3.20) 0.0952** (2.90) 0.1055*** (3.61) 0.3221*** (4.41) 0.1043*** (3.53) OWNERSHIP 0.1579** (2.27) 0.3804** (2.58) 0.1364** (1.93) -0.1007 (-1.78) -0.1850 (-1.07) -0.1550* (-1.97) LOG(SIZE) 0.0023 (0.28) 0.0060 (0.29) 0.0161 (1.19) 0.0279*** (3.48) 0.0672*** (3.42) 0.0451*** (4.11) PARTNER 0.0052** (2.39) 0.0211** (3.22) 0.0032* (1.510 0.0018 (0.94) 0.0052 (0.84) -0.0019 (-1.01) SYNDICATION -0.1326*** (-5.14) -0.3348*** (5.84) -0.1136*** (-4.43) -0.0845*** (-2.82) -0.2639*** (3.52) -0.1045*** (-3.59) AGE 0.0021** (2.24) 0.0061 (2.16) 0.0029* (1.84) -0.0008 (-0.70) -0.0008 (0.40) 0.0004 (0.31) CA 0.0346 (0.54) 0.0034 (0.02) -0.0430 (-1.01) -0.0054 (-0.70) -0.0443 (-0.31) -0.2079*** (-6.14) MA -0.0439 -0.1394 -0.1039 0.1639** 0.5173** -0.0351
Notes: This tables the robustness of the link between the five factors of interest (STAGE, OWNERSHIP, LOG(SIZE), PARTNER and SYNIDICATION) and venture capital firms preferences regarding industry diversify (first case) and geographic scope (second case). Column 1 and 4 are the results of regression formula taking all controls into consideration (column 3 and 6 of Table4). Column 2 and 5 are using logs of the dependent variables. Column 3 and 6 are a subsample of the total samples which size (capital under management) is above $30 million. T-stats are in parentheses.
***Significant at the 1 percent level. **Significant at the 5 percent level. *Significant at the 10 percent level.
Column 2 and 5 uses log of the dependent variables. A typical use of a
logarithmic transformation variable is to pull outlying data from a positively skewed distribution closer to the bulk of the data in a quest to have the variable be normally distributed. In both cases of our sample, for all factors of interest, it remains the same significant level with that in Column 1 and 4 and the directions are the same.
(-0.58) (0.74) (-1.64) (-0.09) (2.84) (-0.83) NY 0.0708 (1.05) 0.1514 (0.91) -0.0144 (-0.31) 0.1531*** (2.31) 0.4908** (3.09) -0.0701* (-1.80) PA 0.1183 (1.67) 0.2092 (1.07) -0.0767 (-1.08) 0.2214*** (2.56) 0.7184*** (3.73) -0.1941** (-3.20) Adj.R-sq 0.3191 0.3650 0.3077 0.4318 0.5001 0.4582 Obs 200 200 169 200 200 169
Column 3 and 6 estimates regression formula of Column1 on a subsample of large firms. In the United States, once a firm has more than $30 million in assets under management, it must register with the Securities and Exchange
Commission. So we divide the sample into two groups, one with asset under management over $30 million, which we classified as large firms; another with asset under management under $30 million. In our sample, 15.5% of the firms’ asset are under $30 million and 84.5% firms’ asset are over $30 million. We found that in both cases, the parameters of all five dependent variable are only slightly different, but not significant different from the coefficient in Column 1 and 4.
5. Conclusion and discussion
This article examines the questions of why venture capital firms prefer varying degrees of industry diversify and geographic scope in their investments. For this purpose, the article hypothesis the variations in VCFs’ preferences are a function of the preferred financing stage of ventures, the ownership structure of the firms, the size of the firm, general partners’ background & experience levels and the degree of syndication of the firm’s investments. Data to test these hypotheses are obtained from 200 US venture capital firms from ThomsonONE database. Data was converted into numeric values and then analyzed using multiple linear regression.
The results of these paper contribute a number of key findings to the study of venture capital firms’ investment preference. By adding more theories into the literature review part and regressing the original data rather than the second hand ones, it tests and complete the model of Gupta’s paper Determinants of Venture Capital Firms’ Preferences Regarding The Industry Diversity And
Geographic Scope Of Their Investments in 1992 and add two new factors into the model (e.g. general partners’ background and experience levels, degree of syndication). The findings can be summarized as follows: (1)It supports Gupta’s paper basically besides two points. Firstly, we found that financial or
independent VCFs enjoy a broader industry diversity than those are subsidiaries of non-financial institutions while there is weak evidence showing this relevance in their research. Secondly, we found no correlation between VCFs’ size and their preferences regarding industry diversify. (2) Venture capital firms with higher general partners’ background and experience levels prefer to invest in a more diverse industry than is the case with other VCFs; However, there are no
differences in preferences regarding geographic scope. (3) Venture capital firms which has higher degree of syndication in its investments will invest in a
narrower diverse set of industries and geographic scope relative to other venture capital firms.
The second finding regarding VCFs’ ownership structure and industry diversity somewhat supports Siegel et al.’s(1988) theory that corporate PEs consider financial return and liquidity to be less important than independently owned companies so that they will invest in industry segments related to the
partners’ background and experience levels with VCFs’ industry preference suggest Bygrave(1987)’s theory that venture capitalist’s expertise provides access to information and deal networks. Access to information and networks provide them with more investment opportunities. For the last finding about degree of syndication with investment preferences, there are many different theories providing different opinions for this question. Our results suggest the Hans Bruining and Ernst Verwaal (2005)’s theory that transaction costs of the syndicate governance are likely to increase with the number of partners in the syndicate.
The limitations of this paper should be noticed. Firstly, we use the number of funds the VCF raised to measure the general partner’s background and
experience levels. It is a general measurement. It would be better if the average working years of the partner’s experience in venture capital industry can be used. But it is hard to get the data due to limitations. Secondly, in this study, we use cross-sectional data to do the research rather than panel data. In this case, we omitted the lagged variables. Lastly, this study assumes there exist linear relationships between VCF characteristics and their investment preferences. This assumption can be false under certain situations. (e.g. low variability in the case of ownership structure). The model might be wrong in this case.
Future study can examine the impact of some other variables such as venture capital firms’ reputation (Rajarishi Nahata, 2008), the level of involvement (Elango et al., 1995, capital structure, technology and the market for exit available for VCFs. What is more, we can also examine the impact of these
variables on other dimensions of diversification such as development stage diversification and time diversification (Mike W. Peng & Andrew Delios, 2006; Margarethe F. Wiersema & Harry P. Bowen, 2007; April Knill,2009).
References
Andrew Delios, Paul W. Beamish, 1999, Ownership strategy of Japanese firms: transactional, institutional, and experience influences, Strategic Management Journal, Volume 20, Issue 10, pages 915–933.
Anil K. Gupta and Harry J. Sapienza, 1992, Determinants of Venture Capital Firms’ Preferences Regarding The Industry Diversity And Geographic Scope Of Their Investments. Journal of Business Venturing 7. 347-362.
April Knill, 2009, Should Venture Capitalists Put All Their Eggs in One Basket? Diversification versus Pure-Play Strategies in Venture Capital, Financial Management ,pages 441 – 486.
Burt, R.S. (1992). Structural Holes: The Social Structure of Competition. Harvard University Press: Cambridge, MA.
Calori R., Harvatopoulos Y, 1988, Diversification: Les Règles Deconduite. Harvard University Press: Cambridge, MA.
Cumming, D., Haslem, B. And Knill, A.M., 2011. Entrepreneurial Litigation and Venture Capital Finance. Working Paper, International Conference of the French Finance Association (AFFI).
Dimo Dimova, Dean A. Shepherdb and Kathleen M. Sutcliffec, 2006, Requisite expertise, firm reputation, and status in venture capital investment allocation decisions, Journal of Business Venturing, 22(4):481-502.
Dimo Dimov, Dirk De Clercq, 2006, Venture Captial Investment Strategy and Portfolio Failure Rate: A Longitudinal Study, Entrepreneurship Theory and Practice: 1042-2587.
Dirk De Clercq, Harry J. Sapienza, Akbar Zaheer, 2008, Firm and Group Influences on Venture Capital Firms’ Involvement in New Ventures, Journal of Management Studies, Volume 45, Issue 7, pages 1169–1194.
Douglas J. Cumming, April M. Knill, 2012, Disclosure, Venture Capital and Entrepreneurial Spawning, Journal of International Business Studies.
Edgar Norton, 1994, Venture Capitalist Attributions and Investment Vehicles: An Exploratory Analysis, The Journal of Entrepreneurial Finance, Vol.3:Iss.3, pp.181-198.
Garry D. Bruton, Sophie Manigart, Vance Fried, Harry J. Sapienza, 2002, Venture Capitalists In Asia: A Comparison With The U.S. And Europe, Vlerick Working Papers 2002/15.
Gifford S, 1997, Limited Attention and the Role of the Venture Capitalist, Journal of Business Venturing, 12: 459-482.
Gompers, P.A., 1996, Grandstanding in the venture capital industry, Journal of Financial Economics, 43: 133-156.
Hana Milanov, D Dimov, DA Shepherd, 2006, Bundles of Social Network Resources: Examining Venture Capital Firms’ Diversification of Investments. Frontiers of Entrepreneurship Research.
Hannan, M.T., J. Freeman, 1984, Structural inertia and organizational change,
American Sociological Review. 49: 149-164.
Hans Bruining, Ernst Verwaal, Andy Lockett, Mike Wright and Sophie Manigart, 2005, Firm Size Effects on Venture Capital Syndication: The Role of Resources and Transaction Costs. ERIM Report Series Research In Management, ERS-2005-077-STR.
Henry Chen, Paul Gompers, 2009, Buy Local? The Geography of Successful and Unsuccessful Venture Capital Expansion, Harvard Business School working paper 09-143.
Henry W. Chesbrough, 2002. Making Sense of Corporate Venture Capital, Harvard Business Review, R0203G.
Henrik Berglund, Tomas Hellstrom, Soren Sjolander, 2007, Entrepreneurial Learning and the Role of Venture Capitalists, Venture Capital, Vol. 9, No. 3, 165 – 181.
Innovation and the contributions from venture capital. Paper for DRUID Conference 2006.
James A. Brander, Raphael Amit, Werner Antweiler, 2002, Venture-Capital Syndication: Improved Venture Selection vs. The Value-Added Hypothesis. Journal of Economics & Management Strategy, vol. 11, no. 3, pp. 422-451.
Joel M. Podolny, 2001, Networks as the Pipes and Prisms of the Market, American Journal of Sociology, Vol. 107, No. 1 (July 2001), pp. 33-60.
Joel M. Podolny, 1993, A Status-Based Model of Market Competition, American Journal of Sociology, 98(4): 829-872.
Joel A. C. Baum, Brian Silverman, 2004, Picking winners or building them?
Alliance, intellectual, and human capital as selection criteria in venture financing and performance of biotechnology startups, Journal of Business Venturing, vol. 19, issue 3, pages 411-436.
Kangmao Wang, Clement K. Wang and Qing Lu, 2002, Differences In Performance Of Independent and Finance-Affiliated Venture Capital Firms, The Journal of Financial Research, Vol. XXV, No. 1.
Margarethe F. Wiersema, Harry P. Bowen, 2011, The relationship between international diversification and firm performance: Why it remains a puzzle? Global Strategy Journal, Volume 1, Issue 1-2, pages 152–170.
Mayer, C., Schoors, K., & Yafeh, Y., 2005, Sources of funds and investment activities of venture capital funds: evidence from Germany, Israel, Japan and the United Kingdom. Journal of Corporate Finance,11(3), 586–608.
Mike W. Peng, Andrew Delios, 2006, What determines the scope of the firm over time and around the world? An Asia Pacific perspective, Asia Pacific Journal of Management, Volume 23, Issue 4, pp 385-405.
Mike Wright, Andy Lockett, 2003, The Structure and Management of Alliances: Syndication in the Venture Capital Industry, Journal of Management Studies, vol. 40, issue 8, pages 2073-2102.
Olav Sorenson, Toby E. Stuart, 2001, Syndication Networks and the Spatial Distribution of Venture Capital Investments. American Journal of Sociology, Vol. 106, No. 6, pp. 1546-1588.
Paul Gompers, Anna Kovner, Josh Lerner, David Scharfstei, 2010, Performance persistence in entrepreneurship, Journal of Financial Economics, 96 (2010) 18– 32.
Pfeffer J, Salancik GR, 1978, External Control of Organizations: A Resource Dependence Perspective, Harper Row: New York.
Rajarishi Nahata, 2008, Venture capital reputation and investment performance, Journal of Financial Economics, 90 (2008) 127–151.
Robert Cressy, Alessandro Malipiero, Federico Munari, 2012, Does VC fund diversification pay off? An empirical investigation of the effects of VC portfolio diversification on fund performance. International Entrepreneurship and Management Journal, 10,139-163.
Ruth Wageman, 1995, Interdependence and Group Effectiveness, Administrative Science Quarterly, Vol. 40, No. 1, pp. 145-180.
Sahlman, William and Gorman, Michael, 1989, What Do Venture Capitalists Do? Journal of Business Venturing, Vol. 4, Issue 4, p. 231-248.
Smith JA, Harre R, Van Langenhove L, 1995, Ideography and the case study. In Rethinking Psychology, pp. 59–69. London: Sage Publications.
Sophie Manigart, Andy Lockett, Miguel Meuleman, Mike Wright, Hans Landstro m, Hans Bruining, Philippe Desbrières, Ulrich Hommel, 2002, Why Do European
Stan Xiao Li, Royston Greenwood, 2004, The Effect Of Within-Industry
Diversification On Firm Performance: Synergy Creation, Multi-Market Contact And Market Structuration, Strategic Management Journal, 25: 1131–1153.
Tanriverdi, H., & Chi-Hyon, L., 2008, Within-industry diversification and firm performance in the presence of network externalities: Evidence from the software industry, Academy of Management Journal, 51(2), 381-397.
Texier, Francois, 2000, Industrial diversification and innovation: An international study of the aerospace industry, Doctoral thesis, monograph (Other academic) of Linko ping University.
Venture Capital Companies Syndicate? ERIM Report Series Research In Management, ERS-2002-98-ORG.
Vesa Kanniainen, Christian Keuschnigg, 2000, The Optimal Portfolio of Start-Up Firms in Venture Capital Finance, CESifo Working Paper Series 381, CESifo Group Munich.
Vladimir I. Ivanov, Fei Xie, 2010, Do Corporate Venture Capitalists Add Value to Startup Firms? Evidence from IPOs and Acquisitions of VC-Backed Companies, Financial Management, Volume 39, Issue 1, pages 129–152.
William D Bygrave, 1987, Syndicated investments by venture capital firms: A networking perspective, Journal of Business Venturing, 2(2): 139-154.
Yochanan Shachmurove, 2010, Major U.S. States as Centers of Venture Backed Entrepreneurial Activities. International Journal of Business. 15(2).
Zipf, G. K. 1949. Human Behavior and the Principle of Least Effort. Reading, Mass.:Addison-Wesley
Appendix I Predictors of VCF involvement
Predictor VCF
Involvement Focal VCF investment relative to all its investments No effect
Focal VCF reputation Negative effect
Focal VCF investment relative to average syndicate investment Positive effect Total reputation of other syndicate members Negative effect Dispersion of reputation among other syndicate members No effect
Board membership Positive effect
Lead investor No effect
Appendix II. r-value and degree of correlation
Correlation Negative Positive
Little -0.09 to 0.0 0.0 to 0.09
Weak -0.3 to -0.1 0.1 to 0.3
Moderate -0.5 to -0.3 0.3 to 0.5
