Venture capital is a form of intermediation particularly well suited to support the creation and growth of innovative, entrepreneurial companies (Hellmann and Puri (2000, 2002), Kortum and Lerner (2000)). It specializes in financing and nurturing companies at an early stage of development (’start-ups’) that operate in high-tech industries. For these companies the expertise of the venture capitalist, its knowledge of markets and of the entrepreneurial process, and its network of contacts are most useful to help unfold their growth potential (Bottazzi, Da Rin and Hellmann (2004), Gompers (1995), Hellmann and Puri (2002), Lerner (1994, 1995), and Lindsey (2003)). By contrast, when venture capital is applied to companies at a later stage of their growth, or in companies which operate in technologically mature industries, it has less of an opportunity to ’make a difference’ (Michelacci and Suarez (2004)). Economics thus points to the relevance of providing an adequate share of venture investments in high-tech and early stage companies.
The creation of ’active’ venture capital markets, i.e. venture capital markets which provide strong support for early stage and high-tech ventures has received a high priority by economic policy, which appreciates its importance for achieving continued economic growth and job creation ((Bottazzi and Da Rin (2002a), European Commission (2003), OECD (2001)). As economies become ever more dependent on innovation and entrepre neurship for achieving sustained growth (Bottazzi, Da Rin, and Giavazzi (2003), Nelson and Romer (1996), OECD (2001)), governments around the world have been trying to replicate the diffusion and success that venture capital has achieved in the United States (Megginson (2004)). These attempts absorb large sums of public money. Yet, we still know very little about what policies can help create active venture capital markets, and our study contributes a first step towards filling this gap.
We start by providing a simple extension of the seminal model of Holmstrom and Tirole (1997) which helps us focus on the determinants of the distribution of financing between early and late stage, and between high-tech and low-tech, investments. We extend the model by Holmstrom and Tirole by allowing for the possibility of an excess supply of funds. As in the original model, firms are heterogeneous in their ability to pledge collateral against borrowing, but we also assume that this ability is higher for firms that possess more tangible assets, which are more easily accepted as collateral than intangible assets. As firms mature from start-ups to later stage ventures, they can rely more on tangible assets. Likewise, firms in high-tech industries make more use of intangible assets than those in traditional industries. This creates a ’pecking order’ in firms’ ability to pledge collateral against loans.
We use this framework to introduce the notion of ’innovation ratios,’ defined to be the ratio of early stage (or high-tech) investments to total venture investments. These ratios provide a measure of the extent to which venture capital markets are active. They are useful for studying how policy can make venture capital markets not only larger but also focussed on those firms which can most benefit from the support of a venture capitalist.
While simple, this framework is rich enough to point to several potential drivers of active venture capital markets which are under the influence of policy-makers. The first is the sheer supply of funds available for investment, which could ease binding credit constraints. Second, there are factors which directly affect project’s expected returns: the existence of exit markets for venture investments and the level of capital gains taxation, both of which affect the capital gains that investors and entrepreneurs can reap when they succeed; and the existence of good technology, which can be turned into commercially valuable ventures. Third, there are factors which affect expected returns by determining the cost of creating a new venture, namely barriers to entrepreneurship.
Starting with Gompers and Lerner (1998) several previous studies have looked at the development of venture capital markets and their determinants. Our approach allows us to advance the literature on several counts. First, we use a simple model to guide our empirical approach, which is based on ratios instead of levels of demand and supply unlike previous analyses.
Our panel approach, based on data for 14 European countries over 1988 to 2001, is another innovation. Previous studies either lacked panel data or focussed on ’between’ country estimators. Jeng and Wells (2000), for example, use a panel of OECD country level data for 1986-95, and focus on cross-country analysis; their approach raises issues of unobserved heterogeneity, that our panel approach can address adequately. Based on our comprehensive set of policy variables, we choose to provide an answer to the question of what can governments do to increase the innovation ratios rather than understanding cross-country variation in the levels of the ratios; we thus focus on estimating ’within’ country effects. Our choice is motivated also by the nature of our data, which come from a rather homogeneous set of developed economies.
One notable exception to pure cross-country comparisons is Gompers and Lerner (1998), who use a panel of U.S. state-level variables for examining the effect of variations in taxation during the 1976-94 period.1 However, ours is the first study to analyze all policy variables with a panel dimension. Our results provide several novel insights.
First, we look at the opening of ’New’ stock markets target at entrepreneurial companies, and find that it considerably increases both the early stage and high-tech innovation ratios. Our panel setting thus provides strong support for the importance of an exit option for venture capital, as suggested by Black and Gilson (1998) and Michelacci and Suarez (2004).
Second, we examine the taxation of capital gains, which has recently received considerable attention in theoretical studies of venture capital (e.g., Keuschingg and Nielsen (2003, 2004)). Previous empirical analyses looked at cross-sectional comparisons of taxation across OECD countries (e.g., Armour and Cummings (2003)). We extend them by collecting a panel of country-level data which goes back to 1988. We find that a reduction in capital gains taxation increases both the high-tech and early stage ratios, albeit the economic effect is not very large.
Third, we consider the effects of a reduction in barriers to entrepreneurship. Like in the case of taxation, we believe we are the first to introduce a panel dimension into the analysis. We consider several alternative measures of such barriers, and find that those related to labor regulations have a significantly negative effect on the share of high-tech investments.
These results are consistent with our model, which suggests that a higher expected rate of return makes more venture funds available for companies with lower collateral, such as those in high-tech and early stage, and thus results in an increase of the innovation ratios.
Finally, our data do not provide any evidence of a shortage of venture capital funds for European companies, contrary to what assumed by the prevailing policy approach. Nor we find evidence that public expenditure in research and development (R&D) favors the innovation ratios through the creation of better entrepreneurial opportunities.
Overall, the European experience we analyze thus suggests that the creation of active venture capital markets might depend crucially on providing investors and entrepreneurs with the possibility to reap the benefits of their efforts, rather than on providing them with more funds.
The rest of the paper is organized as follows. Section 2 briefly describes recent policy programmes for venture capital markets. Section 3 presents our model. Section 4 describes our data and empirical strategy. Section 5 reports our results, and is followed by a brief conclusion.
CONTENTS
Abstract
Non-technical summary
1 Introduction
2 Public policy for active venture capital markets 9
3 A model of venture capital markets
3.1 Direct finance
3.2 Venture capital finance
3.3 Equilibrium
3.4 The innovation ratios
3.5 Equilibrium with excess supply of venture capital
3.6 Taking the model to the data
4 Data and empirical strategy
4.1 Data sources and description
- 4.1.1 Dependent variables
4.1.2 Independent variables
4.2 Empirical strategy
4.3 Descriptive statistics
5 Regression results
5.1 Main results
5.2 Extentions and robustness checks
6 Conclusion
References
Appendix
Figures and tables
European Central Bank working paper series
Download
PDF Ebook Public Policy And The Creation Of Active Venture Capital Markets
