The Development of Geographical Specialization of Venture Capital

Abstract

Many regions have realized that access to capital is an important prerequisite for establishment and growth of businesses, and have therefore focused policies to ensure an adequate supply of risk capital. The growth of the venture capital industry in the 1990s put pressure on venture capital firms (VCFs) to act more strategically. Many VCFs have thus specialized along one or more dimensions: certain industries, stages of development of the firm, or geographical areas. A theoretical dichotomy is developed in this paper to explain regionally focused venture capital. A competence-based theoretical view sees increased competition in the industry as promoting the growth of geographical specialization, while, according to financial theory, it would lead to diversification in order to spread risk. The empirical analysis illustrates the development in the average distance between VCFs and their Danish portfolio firms. All venture capital investments are included. Findings suggest that the process of geographical specialization follows an inverted V-shaped curve. This is interpreted in light of the developments in competition and in the competencies in the market. VCFs search broadly for investment opportunities in the first phase of the emergence of the venture capital industry, but when competition increases they tend to confine themselves to investments within a closer geographical distance. The implications of these findings are important both for funds-of-funds, regional governments, and VCFs.

Acknowledgements

The author is grateful to former student, now senior analyst at the Danish Growth Fund, Jacob Borup for help with recoding and processing some of the data used in this paper. Also thanks for comments on earlier drafts of the paper to Professor Phil Cooke, participants at the Regional Studies Association conference in Angers, France, April 2004, and to participants at an internal IKE seminar at Aalborg University in March 2005. Finally, comments and suggestions from two anonymous referees are gratefully acknowledged. Responsibility for the content is solely the author's.

Notes

1. However, the demand for venture capital may differ from, for example, the share of the national population of firms in a region. A possible uneven distribution may thus be explained by differences in start-up rates or high-tech industries.

2. This period can be characterized as a second take-off, in as much as the venture capital industry initially started and rapidly expanded in the mid-1980s alongside other venture capital markets in Europe. However, in the late 1980s and beginning of the 1990s there was a serious decline of activities, resulting in only two active VCFs in 1992. A detailed account of the development of the Danish venture capital industry can be found in Christensen (2003).

3. De Clercq et al. (2001) is another study based on realized rather than intended strategies. This study is also similar to the present study in being able to identify all venture-backed firms in the population.

4. See Mason and Harrison (2002) for one of the few articles to refer to the theoretical rationale for the specialization of VCFs (referring to Thompson, 1989). However, apart from this reference to Thompson's view on the development of specialization, this article only takes a distributional approach. Sorenson and Stuart ((2001)2001) develop theoretical arguments for the development of geographical specialization using insights from sociology. Furthermore, they use the pairing of VCF and target firm and the actual distance between the two in their empirical analysis. Thus, this approach resembles that of the present paper.

5. In the proceeding period, 1994–1997, our data also show a remarkable concentration, ranging from 94.4% to 98% of Danish venture capital in the Copenhagen area.

6. De Clercq et al. (2001) discuss specialization strategies versus risk reduction using a more differentiated concept of risk, where risk relates to both agency risk, market risk, business risk and systematic risk. The authors hypothesize that VCFs will specialize geographically in order to control for agency risk and business risk.

7. Manigart et al. (2002) selected a sample of 565 venture-backed firms in Belgium for the period 1987–1997. They estimate that their sample covers 57% of venture investments in Belgium. Since we focus on a relatively narrow definition of venture capital (e.g. excluding MBO funds), and for a smaller country in a shorter period, it seems fair to conclude that the number of venture capital-backed firms we found in Denmark is a realistic reflection of the total.

8. Again, it can be argued that follow-up investments are also part of the diversification-specialization decisions of VCFs. This would be difficult to handle in the data, since several of the VCFs in the dataset use milestone payments, which make it difficult to distinguish lump-sum payments from follow-on investments. For example, one of the VCFs lists 20 separate investments in one firm. Theoretically, using new investments makes a stronger case for decisions on specialization.

9. Gorman and Sahlman (1989) find that VCFs use 10 times as much time on an investment when the VCF is the lead investor compared with a syndicated, late-stage investment.

10. Even if the number of cases is small, it is fair to say that there has been an increase in these strategic investments in the Danish venture capital market. This may be seen as an outsourcing of competencies in areas where the costs of building up such competencies are substantial.

11. It can be argued that a more relevant measure would be the travel time between the parties. Using the physical distance does not greatly affect our argument, however, because we are primarily interested in the development in proximity. However, the geography of Denmark means that travel distances and travel times are not the same throughout the country.

12. Even within Denmark there are huge regional differences in how people think about distance and travel time.

13. A bridge was completed in June 1998, but this was partly after the period studied.

14. The Danish Growth Fund (2002) estimates that the number of venture capital and private equity firms combined totalled 37 in 1999.

15. This may be supported by the fact that, from 1996, average distances between the old VCFs and their portfolio firms are persistently far below that of younger VCFs.

16. In a survey of VCFs in Germany and the UK, Martin et al. (2003) find that geographical proximity to invested firms was regarded as more important than proximity to other factors like financial service companies, investors, research institutions, other VCFs.

17. The organization of the fund could also have an effect. Funds, which only exist for a limited time, may tend to invest in more remote firms where capital gains in a near future are more likely. However, in Denmark, the majority of funds invest through evergreen funds, rather then fixed-end funds, which make this effect less important.

18. Similarly, Kannienen and Keuschnigg (2003) find that there is a trade-off between how much advice VCFs can give and the number of firms in the portfolio.

19. In Harding's ((2002)) study of the Danish venture capital market, she notes that, as regards the regional gaps in venture finance “Investment managers of any fund have to be regionally based since they have the knowledge of the regional market. This is particularly important in a country like Denmark where the regional cultures are keenly protected” (p. 18).

20. In Denmark, the Growth Fund co-finances direct investments and back-funding venture funds. In addition, they operate a guarantee scheme for losses in venture funds. The choice of which funds to finance and the criteria for coverage under the guarantee scheme was previously affected by the strategy of the fund regarding the industry and geographical specialization. Thus, legislation required a minimum of €8 million as the capital base for approving a fund. However, funds with a capital base of €3 million also qualified for a guarantee if they were either geographically specialized or focused on seed/early-stage investments. The general opinion is now that larger funds have a better chance of good performance, which may fall back on the possibilities for establishing regional funds.

21. Mason and Harrison support this argument: “Efforts by governments to promote classic venture capital in countries that lack this form of finance may be constrained by the industry's human resource base” (1999, p. 20).

Additional information

Notes on contributors

Jesper Lindgaard Christensen

Fax: [phone]