Portfolio strategies of life science venture capital firms in North America and Europe.

Patzelt, Holger ; Knyphausen-Aufsess, Dodo zu ; Habib, Yasmin 等

Introduction

Most young life science ventures essentially depend on the infusion of venture capital (VC) (Traore, 2005), because their product development processes are risky and capital-intensive. Biopharmaceuticals, for example, demand on average more than 800 million $US R&D expenditure and a 12-year development process (DiMasi, Hansen and Grabowski, 2003) with only one out of 5000 initial drug candidates reaching market launch (Evans and Varaiya, 2003). Financing these expensive development processes is only possible if VCs are willing to take the risk and invest large amounts of money in the young ventures (Powell et al., 2002; Prevezer, 2001). Despite the high risk for investors, however, the life science industry is a major area of activity for VCs because the ventures have an enormous revenue potential. The most successful life science firms such as Amgen and Genentech earn billions of $US every year and their founders and investors have become very rich. In 2006, VCs invested 3.5 billion $ US in North America (US and Canada) and 1.9 billion $US in European life science ventures (Ernst & Young, 2007).

The emergence of the life science sector, however, has not been homogeneous across countries. Already in 1986, the North American industry counted more than 800 companies and employed more than 40,000 people, and in 2006 these numbers amounted to more than 1,900 firms and 188,000 employees (Ernst & Young, 2003c, 2007). In contrast, in European countries the development of the sector is about 10 to 20 years behind the North American industry. In the 1990s governmental programmes such as the BioRegio competition in Germany (Dohse, 2000) were major drivers of the rapid growth of the European life science industry, and the number of firms grew from 450 in 1992 to more than 1,600 in 2006 (Ernst & Young, 2003a, 2007). However, relatively few of these companies are established corporations yet, and total employment in the sector equalled only about 75,000 people, corresponding to less than 40% of the US sector. Whereas in North America 418 life science firms were quoted at the stock markets in 2006, this was the case for only 156 European firms. In that year, North American companies generated 62 billion $US in revenues, as compared to 19 billion $US of European firms (Ernst & Young, 2007). Table 1 provides an overview of the life science industries in North America and Europe.

Several differences between North America and European countries may explain the different development paths of the life science industries in both continents. First, with the exception of the UK, European countries were lacking an established VC industry that is essential for stimulating the foundation and growth of life science ventures (Cooke, 2001; Giesecke, 2000). For example, in large European countries like Germany and France the legal system offers low protection for shareholders, which counteracted the establishment of independent VC firms (La Porta et al., 1997). Moreover, the stock markets in European countries outside the UK are under-developed and left little opportunities for VC firms to exit their investments (Black and Gilson, 1998). In the historically bank-based financial environments of Europan countries like France and Germany, VC firms are often owned by banks (Wright et al., 2004), and the VC managers of these firms usually have a finance background (Ooghe, Manigart and Fassin, 1991), which leads to them to be averse to non-controllable financial risks of young technology ventures (Dimov, Shepherd and Sutcliffe, 2007). Second, many European countries had unfavourable legal and regulatory frameworks that blocked the emergence of the life science sector. For instance, in Germany restrictive laws on the usage of genetic material and methodologies, and the genetic manipulation of organisms made it difficult for young firms to engage in life sciences (Engel and Heneric, 2008). Finally, many European countries are known to lack a culture supportive of entrepreneurship. For example, the national cultures in the Scandinavian countries, France, and Germany are characterized by high levels of uncertainty avoidance and risk aversion (Hofstede, 2001), which is detrimental to entrepreneurial action (McMullen and Shepherd, 2006). Moreover, in some European countries the population shows a rather hostile attitude with regard to the new technologies, making the development of radically new technology-based industries such as life sciences difficult (Giesecke, 2000). Thus, the overall entrepreneurial activitiy is very weak in these countries (Reynolds et al., 2000). In contrast, the North American culture is characterized by much lower levels of uncertainty avoidance than most European countries (Hofstede, 2001), leading to a favourable climate for entrepreneurial action (McMullen and Shepherd, 2006; Reynolds et al., 2000).

The different development paths of both the life science industries and the VC sectors in North America and Europe suggest that the investment strategies of VC firms in both continents may differ. Thus, the purpose of our study is to extend "the rather limited research on international comparisons of venture capital firms" (Wright, Lockett and Pruthi, 2002: 14) to the specific context of the life science sector. Whereas existing studies have, for example, investigated the effect of different institutional settings on the valuation methods and information sources used by VCs (Karsai et al., 1998; Wright, Lockett and Pruthi, 2002; Wright et al., 2004) and the development of the VC industry (Bruton and Ahlstrom, 2003; Bruton, Ahlstrom and Singh, 2002; Bruton, Fried and Manigart, 2005), we analyse the portfolio strategies of North American and European life science VCs. We draw on data on 88 life science VCs financing a total of 1,050 ventures, and investigate differences with respect to investee stages, geographic proximity, and life science markets of investees.

Our article is structured as follows. In the next section, we introduce portfolio strategies of life science VCs. We then describe our methodology and data. In the subsequent part we present our results and possible explanations before we discuss our findings and draw conclusions.

Portfolio Strategies of Life Science Venture Capital Firms

VC firms face various risks when investing in young technology ventures. First, young technology companies often operate in new markets and develop new technologies where little knowledge is available so far, making it difficult for the VCs to judge the future development of technologies and markets of their investees (Sanders and Biovie, 2004). Second, VCs can not exit their investments easily because the valuation of these ventures is difficult (Remer, Ang and Baden-Fuller, 2001) and VCs may not achieve agreement with potential buyers. Finally, VCs face considerable difficulties to evaluate the performance of their investees because considerable information asymmetries between entrepreneurs and VCs exist (Lerner, 1995). In order to counteract these risks, VCs invest in a portfolio of new ventures (Norton and Tenenbaum, 1993). Life science VCs have several possibilities to select the composition of these portfolios.

Investment Stage Diversification

First, VCs can focus on a specific development stage of their investees (early or late stage firms) or diversify across stages (Manigart et al., 2002). Following existing literature, we distinguish between early and late stage investees, where early and late stages are defined in terms of the product-market characteristics of the ventures (Dimov, Shepherd and Sutcliffe, 2007). That is, early stage firms have ill-defined product market characteristics, for example pursue only projects in early development stages far from market launch. In contrast, late stage ventures may have entered the market with their first product(s) and may even have reached profitability. We believe that this definition is more adequate in the context of the life science industry than absolute definitions, such as the age of the ventures. For example, a five-year-old drug development firm is typically far away from introducing its first drug to market and is thus an early stage venture, whereas a five-year-old life science firm developing platform technologies and research services may be profitable already and is better described as a later stage firm (Casper, 2000).

For VC firms, specialization on one specific investment stage (early or late) allows for the acquisition of deeper knowledge and understanding of that investment stage that may help them to better select investees at each particular stage and support them in developing their business (Manigart et al., 2002). In the context of the life science industry, VCs specializing in early stage ventures may develop management skills to assist inexperienced academic founders or help them to build a network with research groups at universities and other research organizations, which can facilitate the development of the venture's technology (Zucker, Darby and Armstrong, 2002). In contrast, in cases where life science ventures are more mature, their management is more professionalised and their technology more developed. Thus, late stage specialist VCs must possess different skills to support their ventures, for example providing assistance in establishing production and marketing facilities or building up contacts to incumbent firms, which often serve as alliance partners for further development of the venture's late stage products (Audretsch and Feldman, 2003; Deeds and Hill, 1996; Rothaermel, 2001).

Geographic Proximity of Investees

VCs can focus on investees in close geographic proximity or also invest in more distant ventures. We define investees in close geographic proximity as those investees that are located in the same continent as the focal VC (Europe or North America). Investing in start-ups in geographic proximity helps VCs to actively assist their investees and be involved in daily business (Gupta and Sapienza, 1992), for example by providing them with management expertise, strategic advice, and external contacts (Gorman and Sahlman, 1989; Sapienza, 1992; Timmons and Bygrave, 1986). In the life science industry, the necessity to assist inexperienced academic founders in management may lead VCs to put particular emphasis on geographic proximity of investees. On the other hand, if VCs focus their investment on a particular geographic region, they may have difficulties creating the desired deal flow of promising investment opportunities, whereas VCs that invest in distant markets usually can draw on a larger deal flow (Hall and Tu, 2003). The problem of deal flow generation may be crucial for specialised life science VCs since they a priori have a lower deal flow than those VCs which do not focus their investments on one industry (Patzelt, zu Knyphausen-Aufsess and Arnoldt, 2006).

Market Focus

Finally, the markets in which prospective investees operate are important factors that influence whether or not VCs should invest in new technology ventures or not (Bachher and Guild, 1997). The life science industry is comprised of firms operating in biotechnology, medical technology, and healthcare/information technology (healthcare/IT) markets (Arundale, 2002). Of those, biotechnology can be further subdivided into firms serving the therapeutics, diagnostics, service/supply, and other markets (Ernst & Young, 2003c). Therapeutics companies develop new therapeutics for unmet clinical needs such as cancer or Alzheimer's disease. Diagnostics firms draw on new biotechnological methods to develop diagnostic tests for humans. Service/supply firms offer research services (for example DNA sequencing) to companies or research organizations, or supply laboratories with material for daily use (for example DNA and protein purification kits). The small group of other firms includes, for example, companies that develop gene-manipulated, pest-resistant plants or animal therapeutics. The medical technology market refers to the development of devices and machines which are used in diagnosis and therapy such as cardiographs, endoscopes, and positron emission tomographs. Firms serving the healthcare/IT market draw on information technology to increase the productivity of processes in the healthcare sector. These ventures offer, for example, efficient personnel management services, risk management for hospitals, and marketing and accounting services, but also software such as patent-specific information systems. It is important to note that the therapeutics and diagnostics markets are well established and traditionally served by pharmaceutical incumbent firms (although their underlying product development technologies differ from those of modern life science ventures). In contrast, many modern medical technology and particularly healthcare/IT firms serve very new markets which mainly have emerged over the last decade (Stolis and Goodman, 2004).

Methodology and Data

We define a "life science VC" as a VC firm with at least 10 investees serving the biotechnology, medical technology, or healthcare/IT markets. This cut-off is a compromise to ensure that, on the one hand, the VCs have the possibility to efficiently diversify their life science portfolios, and, on the other hand, to avoid excluding too many VCs from our analysis. For portfolio analysis, we only took into account investees which are privately held and omitted ventures that were publicly traded at the stock markets. The reason why we excluded the latter ventures is that VCs can easily sell shares of these ventures on the stock market in case they do not perform as expected. In contrast, shares of privately held ventures can not easily be sold and the VCs carry the risk to loose their entire investment when the ventures fail. Thus, it appears that VCs will pay particular attention to the appropriate design and diversificaton of their portfolio of privately held firms.

Sampling Procedure

As a starting point for identification of our target population of life science VCs we drew on industry reports of the consulting company Ernst & Young (Ernst & Young, 2001, 2002, 2003a, 2003b), who are among the leading industry observers. Ernst & Young's reports have been published since 1986 and cover the North American, European, and global life science sectors. It is important to note that while there are some commercially available data bases on the life science industry in North America (for example BioScan), there are no data sources of comparable quality and completeness for the European sector. We therefore used snowball-sampling to access further life science VCs and their portfolio companies. Snowball sampling refers to "a technique for finding research subjects. One subject gives the researcher the name of another subject, which in turn provides the name of a third, and so on" and is based on the assumption that "a 'bond' or 'link' exists between the initial subject and others in the same target population, allowing a series of referrals to be made within a circle of acquaintance" (Atkinson and Flint, 2004: 1). Thus, it is a prerequisite for the application of snowball sampling that the subjects of the population are highly connected. This is certainly true in our context because due to their capital intensity, life science ventures usually have a broad base of VC investors and thus provide a link between individual VCs. These VC investors, on the other hand, invest in a portfolio of start-ups. Thus, the VCs and their investees are connected in a network-like manner, and one subject of the population can be used to identify another subject. While this connectedness supports the use of snowball sampling in our case, it is important to note possible limitations of this procedure. Specifically, snowball sampling faces the risk of a systematic error in identifying target populations, that is, the risk that the identified population is not representative for the sample (Van Meter, 1990). This is because the selection of the starting point(s) of the sampling procedure is not random but determined by the subjective choice of the researcher (Griffiths et al., 1993). In order to ensure high representativeness and validity of the sample, it is therefore necessary to cover a substantial part of the overall population (Van Meter, 1990). Our goal was therefore to identify a substantial part of all existing life science VCs in North America as well as Europe. While we do not have numbers for the size of the target population of life science VC firms for comparison, the use of multiple prominent data sources such as the Ernst & Young reports makes it likely that we identified the majority of the major players. Moreover, the fact that we cover considerably moreconsiderably more than half of all existing VC-backed life science ventures in North America and Europe (see below) supports the view that our sample is large enough to ensure representativeness and validity. Indeed, because of the close connectedness of actors in the VC market, snowball sampling has been previously successfully applied in VC research to identify representative target populations (for example Wright, Lockett and Pruthi, 2002).

Data Sources and Variables

Following existing studies (Dimov, Shepherd and Sutcliffe, 2007; Dimov and Shepherd, 2005; Patzelt, zu Knyphausen-Aufsess and Arnoldt, 2006), we used the web pages of VC firms as a main data source. In rare cases, we supplemented these data with information from industry reports and biotech press. Two researchers (one of them holding a PhD in life sciences, the other one completing a diploma in business administration) content-analyzed the web pages separately and then compared their results. When the researchers came to different conclusions and interpretations about the information posted on the web pages, they discussed these differences until agreement was reached. All data were collected from April to June, 2005.

We used binary variables to describe whether the VCs (i) focused on any particular venture stage (early or late) or not, and (ii) invested only in geographic proximity (exclusively in their home continent North America or Europe) or with a broad geographic scope (also overseas). For the VC firms of our sample, this information was provided on their web pages, which typically contained a section named "strategy," "how we work," or "funds," and indicated whether and on which stages the VCs focused their investments. For example, German Techno Venture Management TVM stated on its web page that its life science fund VI "is funding early and late stage companies in the US and Europe" and was thus classified as being diversified across investment stages, and investing with a broad geographic scope. In contrast, North American VCs Sofinnova and Three Arch Partners stated that "our firm invests venture capital in early-stage life-science [...] companies" and "our primary focus is on early stage investments," respectively. These VCs were classified as being non-diversified across investment stages.

Continuous variables describe the shares of firms in North American and European VCs' portfolios which serve (i) therapeutics, (ii) diagnostics, (iii) service/supply, (iv) medical technology, (v) healthcare/IT, or (vi) other life science markets. The web pages of the VCs typically entailed a section "portfolio" or "our investments" where the names of the portfolio firms are listed and a short description of their markets was provided. If this information was insufficient to assign the portfolio firm to one of the categories mentioned above, we visited the web page of the portfolio firm to gather this information and assign it to the appropriate category.

Sample Description

We identified 171 VC firms with investments in the life science sector in North America and Europe. Eighty-eight VCs held at least 10 non-exited life science investees in their portfolios. These VCs financed a total of 1,050 life science ventures, which corresponds to a substantial part of the total industry. In 2004, Ernst & Young listed 1114 privately owned life science firms in the USA, and 1717 in Europe (Ernst & Young, 2005a). However, most of these firms were not financed by VC. In Germany, for example, where the largest number of life science ventures among all European countries is located, only about one third of them were VC-backed (Ernst & Young, 2005b). Although this fraction may be higher in North America due to the more established VC market, it appears a conservative estimation that not more than two thirds of all privately held life science firms in both continents (about 1,800) receive VC. We thus estimate that, in terms of investees, our sample (1,050 firms) covers considerably more than half of the life science industries in North America and Europe. With respect to VC investors, the coverage may even be higher due to the network-like connectedness of investees and VCs.

Our final sample consisted of 65 North American and 23 European life science VCs. Fifty-one VCs exclusively invested in life sciences (39 were North American and 12 European). We included the major players in the life science industry such as Advent International, Atlas Ventures, Polaris Venture Partners, and MPM Capital in North America, and 3i, Apax, TVM, Abingworth, and Global Life Science Ventures in Europe, but also small firms such as POSCO Bioventures, Spray Partners, and Bioventure Investors. On average, the VCs in the sample held 21 life science portfolio firms and had 1.2 billion US dollars worth of total capital under management (including non-life science investments). With regard to investees, 729 (69%) of the 1,050 life science ventures were located in North America. Of the remaining 321 European firms, 73 (22.7%) had their headquarters in Germany, and 63 (19.6%) in the UK. We also included Israeli firms in the European sample for three reasons. First, this classification is consistent with Ernst & Young who treat Isreali firms as part of the European life science industry. Second, the development stage of the Isreali life science sector is comparable to the sector in European countries such as France, Sweden, and Switzerland (Ernst & Young, 2001). Third, multiple interfirm linkages exist between the European and Isreali sector, including alliances between life science firms and investments of European life science VCs in Israel (Ernst & Young, 2001). Of the investee sample as a whole, 466 (44.4%) firms developed therapeutics, 306 (29.1%) medical technology, 137 (13.0%) were service/supply, 86 (8.2%) healthcare/IT, 34 (3.2%) diagnostics, and 21 (2.0%) other companies. Of the 729 North American investees, 283 (38.8%) developed therapeutics, 248 (34.0%) medical technology, 92 (12.6%) were service/supply, 74 (10.2%) healthcare/IT, 20 (2.7%) diagnostics, and 12 (1.6%) other companies. Of the 321 European portfolio ventures, 183 (57.0%) developed therapeutics, 58 (18.1%) medical technology, 45 (14.0%) were service/supply, 12 (3.7%) healthcare/IT, 14 (4.4%) diagnostics, and 9 (2.8%) other companies (see also Table 4).

Table 2 shows characteristics of the North American and European life science VCs in our sample. European VCs have, on average, less capital under management, a larger portfolio, and a larger team of investment managers allocated to their life science investees. However, none of these differences is statistically significant. These characteristics are in line with industry reports which observe only slightly smaller funds of life science VCs in Europe as compared to North America (Ernst & Young, 2002). The lower number of portfolio firms of North American VCs may be due to the higher tendency to liquidate under-performing investees, an observation described before in the life science industry (Howell, Trull and Dibner, 2003). These findings suggest that our sample well represents the VC industries.

Statistical Method

Depending on whether the variable of interest was binary or continuous, we performed different tests for comparisons of portfolios of North American and European life science VCs. For binary variables, we performed Chi-square-tests, whereas for continuous variables, we used Mann-Whitney-tests. In contrast to t-tests, Mann-Whitney-tests do not require a normal distribution of the variables of interest. Since we have no indication that the variables of our analysis are normally distributed, Mann-Whitney-tests are appropriate for our purpose.

Results and Analysis

Table 3 shows differences between North American and European life science VCs with respect to investment stage diversification and the geographic proximity of their investees. We report the number of VCs in the category of interest (stage diversified vs. stage specialist VCs, VCs investing in geographic proximity vs. VCs investing with broad geographic scope) as compared to the total number of VCs in the group, the percentage of VCs in that category, and the Chi-Square test statistics with asterisks indicating the level of statistical significance.

Table 3 demonstrates that North American life science VCs have portfolios that are not more diversified across investment stages than those of their European counterparts.

These findings are somehow in contrast to previous research on non-industry specialised VCs, which suggest that North American VCs more likely focus on early stage ventures than European VCs (Manigart et al., 2002), potentially because the latter are more averse to the risks of early stage investees. One possible explanation for our result is that the early development stage of the European life science sector makes it difficult for European VCs to acquire a sufficient deal flow of late stage investees.

We find statistically significant differences between North American and European life science VCs with respect to the geographic proximity of their investees. Table 3 shows that North American VCs have a significantly higher tendency to invest exclusively in their home markets than their European counterparts. Almost two thirds of the European life science VCs in our sample also have investees in North America, but only one third of North American VCs invest overseas. Two observations may explain this finding.

First, the different development stages of the life science industries in North America and Europe may yield a higher necessity for European VCs to invest globally in order to enlarge their deal flow of attractive investment opportunities. The home market of European VCs is mostly comprised of young, early stage ventures, which are often led by inexperienced founders (Ernst & Young, 2001). In contrast, North American VCs find attractive opportunities in geographic proximity because the North American sector is mature and many start-ups are led by experienced entrepreneurs and managers. Given that the quality of the venture management is a major criterion for VCs to select their investees (MacMillan, Siegel and Narasimha, 1985; Shepherd, 1999; Tyebjee and Bruno, 1984), European life science VCs may thus have a high incentive to seek investees in the North American market.

Second, investing in the North American life science market is also a means for European life science VCs to syndicate with North American VCs. This renders two advantages. First, syndication enhances their deal flow (Manigart et al., 2006), which appears important for European life science VCs (see above). Second, since European VCs have in general less experience and expertise in screening projects and actively supporting and controling investees than VCs in North America (liege, Palomino and Schwienbacher, 2003), they can access the knowledge and expertise of the latter (Bygrave, 1987; Lockett and Wright, 2001). This may be particularly crucial for European life science VCs, because their target sector is young and VC professionals have had few opportunities to collect industry-specific experience. Their North American counterparts, in contrast, have been investing in life science ventures for almost three decades and have thus accumulated substantial knowledge about the sector.

Table 4 displays differences between North American and European life science VCs with respect to the life science markets. We list for both groups of VCs the mean of the shares of portfolio companies serving the above introduced markets, the standard deviations, and the Mann-Whitney test statistics with asterisks indicating the level of statistical significance.

The numbers in Table 4 demonstrate that the portfolios of North American life science VCs comprise about twice as many firms serving the new medical technology and healthcare/IT markets, but a lower share of investees serving the established therapeutics and diagnostics markets, as compared to European VCs. These differences are statistically significant. Perhaps their superior experience and better ability to screen projects and support their portfolio ventures (liege, Palomino and Schwienbacher, 2003) enables North American life science VCs to invest more in the disruptive and highly uncertain medical technologies and healthcare/IT markets than European VCs. Since there are few individuals that have collected significant experience in the young European life science industry, many European life science VC professionals look back on careers in the pharmaceutical industry. Their knowledge and expertise appear more applicable to markets related to therapeutics and diagnostics development than to medical technology and health care/IT markets.

Discussion and Conclusion

The aim of this study was to analyse differences between portfolio strategies of VCs active in the life science industry in North America and Europe. We find that European life science VCs have a significantly lower tendency to invest in geographic proximity than North American VCs. Moreover, our results show that European VCs focus more on ventures serving the established therapeutics and diagnostics markets, whereas North American VCs invest more in medical technology and healthcare/IT firms. Although our work is exploratory, it offers some insights into areas of VC research.

First, we add to the thus far sparse empirical work (Wright, Lockett and Pruthi, 2002) on international comparisons of VC firms. Recent research has, for example, examined the effect of culture and the regulatory environment on VC investment strategies in emerging markets in Asia and Eastern Europe (Bruton and Ahlstrom, 2003; Bruton, Ahlstrom and Singh, 2002; Karsai et al., 1998; Wright, Kissane and Burrows, 2004). Our results suggest that another variable-the specific characteristics and development stage of the VCs' local target industry-impacts deal flow generation and thus the portfolio strategies of VCs in different countries. Deal flow limitations may be particularly important for industry-specialised VCs because these have a priori fewer investment opportunities than their non-specialised counterparts. For example, we find that European VCs invest more in the North American life science market than North American VCs invest in European ventures, which can be explained by the early development stage of the European sector and the difficulties to acquire attractive investment opportunities in this sector. Thus, particularly for industry-specialised VCs, deal flow limitations of local markets may have a considerable impact on VCs' portfolio strategies. We encourage scholars to explore this seldom analysed issue (Wright and Robbie, 1998) in future studies.

We also provide some insights into internationalisation strategies of VCs. Existing studies have focused on the internationalisation of VCs into emerging markets and have analysed, for example, the internationalisation of US VC firms as they enter the Indian market (Wright, Lockett and Pruthi, 2002). Our result that life science VCs in Europe are more likely to pick investments in North America than North American VCs invest in the European sector suggests that the opposite direction of internationalisation-internationalisation from emerging into developed markets-is also important for VCs. European VCs active in the young domestic life science industry appear to have a particular need to internationalise into the more established life science market in North America. Motives for such internationalisation may include the access to experienced syndication partners and the generation of a deal flow of promising investees.

Finally, our focus on industry-specialised VCs allowed us to gain a deeper insight than previous studies into VCs' portfolio strategies with regard to the technologies and markets they focus on. Markets of technology-based investees are an important investment criterion for VCs (Bachher and Guild, 1997), but there has been only limited work on market diversification of VC portfolios. Whereas existing studies measure the aggregate number of industries VCs invest in (Gupta and Sapienza, 1992; Norton and Tenenbaum, 1993), we go one step further and distinguish different markets within the life science industry. To our knowledge, only one recent study has focused on the life science sector so far. Similar to our study, Patzelt et al. (2006) distinguished life science markets therapeutics, diagnostics, medical technology, and service/supply. These authors provided case studies on investment and risk reduction strategies of life science VCs located in Central Europe (mainly Germany), and also find that deal flow limitations inhibit the VCs' possibilities to efficiently design their portfolio. Our results support this argumentation. In addition, our finding that North American and European VCs prefer different life science markets suggests that the experience, knowledge and expertise VC managers have collected in those markets impacts their portfolio strategy. More experience may be related to a higher tendency to invest in risky technologies and markets where little knowledge is available so far. In contrast, if VCs have collected little experience in an industry because the sector is still young, they may prefer to invest in those ventures serving established markets. Future research may test these propositions on a statistical basis.

The findings we present have several implications for life science entrepreneurs and venture capitalists. First, our results suggest that life science entrepreneurs will find different opportunities to acquire VC depending on the markets they target and the home continent of their venture. Medical technology and healthcare/IT ventures appear to be more likely to attract VC in North America, whereas therapeutics and diagnostics companies, on the other hand, may find good opportunities to raise capital in Europe. European ventures developing medical technology and healthcare/IT may thus try to access the North American VC market, possibly by internationalising into the North American sector. This strategy has been followed by European life science ventures in the past to access public capital markets via a listing on the NASDAQ (Ernst & Young, 2001), and our results suggest that it may also pay off for ventures financed through the private capital markets. Second, with respect to VC investors, our study illustrates that investment strategies vary across countries and continents, and VCs may take these differences into account during the internationalisation process. For example, when European life science VC firms extend their geographic scope and choose to invest also in North America, they may consider that, in that life science market, other portfolio strategies, namely higher shares of medical technology and healthcare ventures in the portfolio, are the most dominant strategies of local (North American) VC firms. Similarly, North American VC firms that start investing in Europe may consider picking more therapeutics and diagnostics ventures among their European investees and invest less in European medical technology and healthcare ventures.

Our paper has limitations which suggest avenues for future research. First, it appears important that scholars study in more detail how the development stage and maturation of a target industry is related to the investment strategies of the VCs. Our exploratory com parison of the young life science sector in Europe with the mature North American industry reveals significant differences in the VCs' strategies. However, we do not account for other country-specific variables such as institutional influences, which may also impact the VCs' behaviour (Bruton and Ahlstrom, 2003; Bruton, Ahlstrom and Singh, 2002; Karsai et al., 1998; Wright, Kissane and Burrows, 2004). Therefore, a more rigorous way to perform the analysis would be to follow the portfolio strategies of VCs in one market for a longer period of time, for example the North American market from the 1980s until today. Such a study would certainly provide further insight on the effect of industry development on VC investment strategies. Second, we do not have data on how the differences between portfolio strategies of North American and European life science VCs are related to their success. Although existing research suggests that North American VCs are on average more successful than their European counterparts (liege, Palomino and Schwienbacher, 2003), in the specific context of the life science industry no information is available so far. The specificities and different development stages of the life science sectors in North America and Europe may make different portfolio strategies for European and North American VCs necessary to achieve high returns. Future research ought to analyse this important topic.

References

Arundale, K. 2002. Money for Growth-The European Technology Investment Report 2001. London, UK: PriceWaterhouseCoopers.

Atkinson, R. and J. Flint. 2004. "Snowball Sampling." Pages 1043-44 in M. Lewis-Beck, A. Bryman and T. Fining Liao (eds.), The SAGE Encyclopedia of Social Science Research Methods. London: Sage.

Audretsch, D.B. and M.P. Feldman. 2003. "Small-firm Strategic Research Partnerships: The Case of Biotechnology," Technology Analysis and Strategic Management 15, no. 2: 273-88.

Bachher, S.J. and P. Guild. 1997. "Financing Early Stage Technology Based Companies: Investment Criteria by Investors." In Frontiers of Entrepreneurship Research, 1996 Edition. Babson Park, MA: Babson College.

Black, B.S. and R.J. Gilson. 1998. "Venture Capital and the Structure of Capital Markets: Banks Versus Stock Markets," Journal of Financial Economics 47: 243-77.

Bruton, G. and D. Ahlstrom. 2003. "An Institutional View of China's Venture Capital Industry: Explaining the Differences Between China and the West," Journal of 'Business Venturing 18, no. 2: 233-59.

Bruton, G. D., D. Ahlstrom and K. Singh. 2002. "The Impact of the Institutional Environment on the Venture Capital Industry in Singapore," Venture Capital 4, no. 3: 197-218.

Bruton, G.D., V.H. Fried and S. Manigart. 2005. "Institutional Influences on the Worldwide Expansion of Venture Capital," Entrepreneurship Theory and Practice 29, no. 6: 737-60.

Bygrave, W.D. 1987. "Syndicated Investments by Venture Capital Firms: A Networking Perspective," Journal of 'Business Venturing 2, no. 2: 139-54.

Casper, S. 2000. "Institutional Adaptiveness, Technology Policy, and the Diffusion of New Business Models: The Case of German Biotechnology," Organization Studies 21, no. 5: 887-914.

Cooke, P. 2001. "New Economy Innovation Systems: Biotechnology in Europe and the USA," Industry and Innovation 8, no. 3: 267-89.

Deeds, D.L. and C.W.L. Hill. 1996. "Strategic Alliances and the Rate of New Product Development: An Empirical Study of Entrepreneurial Biotechnology Firms," Journal of Business Venturing 11, no. 1: 41-55.

DiMasi, J.A., R.W. Hansen and H.G. Grabowski. 2003. "The Price of Innovation: New Estimates of Drug Development Costs," Journal of Health Economics 22, no. 2: 151-85.

Dimov, D.P. and D.A. Shepherd. 2005. "Human Capital Theory and Venture Capital Firms: Exploring 'Home Runs' and "Strike Outs'," Journal of Business Venturing 20, no. 1: 1-21.

Dimov, D., D.A. Shepherd and K.M. Sutcliffe. 2007. "Requisite Expertise, Firm Reputation, and Status in Venture Capital Investment Allocation Decisions," Journal of Business Venturing 22, no. 4: 481-502.

Dohse, D. 2000. "Technology Policy and the Regions-The Case of the BioRegio Contest," Research Policy 29, no. 9: 1111-33.

Engel, D. and O. Heneric. 2008. "Legal Frameworks and Public Support in the Biotechnology Industry." In H. Patzelt and T. Brenner (eds.), Handbook of Bioentrepreneurship. Heidelberg: Springer.

Ernst & Young. 2001. Integration-Ernst & Young's Eight Annual European Life Science Report 2001. Cambridge, UK.

--. 2002. Beyond Borders-Global Biotechnology Report 2002. Cambridge, UK.

--.2003a. Endurance-The European Biotechnology Report 2003. Cambridge, UK.

--. 2003b. Resilience Americas Biotechnology Report 2003. New York, NY.

--. 2003c. Zeit der Bewdhrung-Deutscher Biotechnologiereport 2003. Mannheim, Germany.

--.2005a. Competitive Growth-Growing Competition. Cambridge, UK.

--. 2005b. Krdfte der Evolution-Deutscher Biotechnologiereport 2005. Mannheim, Germany.

--.2007. Beyond Borders: The Global Biotechnology Report 2007. Cambridge, UK.

Evans, A.G. and N.P. Varaiya. 2003. "Assessment of a Biotech Market Opportunity," Entrepreneurship Theory and Practice 28, no. 1: 87-105.

Giesecke, S. 2000. "The Contrasting Roles of Government in the Development of Biotechnology Industry in the US and Germany," Research Policy 29, no. 2: 205-23.

Gorman, M. and W.A. Sahlman. 1989. "What Do Venture Capitalists Do?," Journal of Business Venturing 4, no. 4: 231-48.

Griffiths, P., M. Gossop, B. Powis and J. Strang. 1993. "Reaching Hidden Populations of Drug Users by Privileged Access Interviewers: Methodological and Parcital Issues," Addiction 88: 1617-26.

Gupta, A.K. and H.J. Sapienza. 1992. "Determinants of Venture Capital Firms' Preference Regarding the Industry Diversity and Geographic Scope of Their Investments," Journal of Business Venturing 7, no. 5: 347-62.

Hall, G. and C. Tu. 2003. "Venture Capitalists and the Decision to Invest Overseas," Venture Capital 5, no. 2: 181-90.

Hege, U., F. Palomino and A. Schwienbacher. 2003. "Determinants of Venture Capital Performance: Europe and the United States." Working Paper. Paris: HEC School of Management.

Hofstede, G. 2001. Culture's Consequences: Comparing Values, Behaviors, Institutions and Organizations across Nations. Thousand Oaks, CA: Sage.

Howell, M., M. Trull and M.D. Dibner. 2003. "The Rise of European Venture Capital for Biotechnology," Nature Biotechnology 21, no. 11: 1287-91.

Karsai, J., M. Wright, Z. Dudzinski and J. Morovic. 1998. "Screening and Valuing Venture Capital Investments: Evidence from Hungary, Poland and Slovakia," Entrepreneurship and Regional Development 10, no. 3: 203-24.

La Porta, R., F. Lopez-De-Silanes, A. Shlcifer and R. Vishny. 1997. "Legal Determinants of External Finance," Journal of Finance 52, no. 3: 1131-50.

Lerner, J. 1995. "Venture Capitalists and the Oversight of Private Firms," Journal ofFinance 50, no. 1: 301-18.

Lockett, A. and M. Wright. 2001. "The Syndication of Venture Capital Investments," Omega 29, no. 5: 375-90.

MacMillan, I. C., R. Siegel and S.P.N. Narasimha. 1985. "Criteria Used by Venture Capitalists to Evaluate New Venture Proposals," Journal of Business Venturing 1, no. 1: 119-28.

Manigart, S., A. Lockett, M. Meuleman, M. Wright, H. Landstr6m, H. Bruining, P. Desbrieres and U. Hommel. 2006. "Venture Capitalists' Decision to Syndicate," Entrepreneurship Theory and Practice 30, no. 2: 131-53.

Manigart, S., K.D. Waele, M. Wright, K. Robbie, P. Desbrieres, H.J. Sapienza and A. Beekman. 2002. "Determinants of Required Return in Venture Capital Investments: A Five-country Study," Journal of Business Venturing 17, no. 4: 291-12.

McMullen, J.S. and D.A. Shepherd. 2006. "Entrepreneurial Action and the Role of Uncertainty in the Theory of the Entrepreneur," Academy of Management Review 31, no. 1: 132-52.

Norton, E. and B.H. Tenenbaum. 1993. "Specialization Versus Diversification as a Venture Capital Investment Strategy," Journal qf Business Venturing 8, no. 5: 431-42.

Ooghe, H., S. Manigart and Y. Fassin. 1991. "Growth Patterns of the European Venture Capital Industry," Journal of Business Venturing 6, no. 6: 381-404.

Patzelt, If., D. zu Knyphausen-Aufsess and I. Arnoldt. 2006. "How do Venture Capitalists Spread Risk by Diversification Within Specialized Life Science Portfolios?" International Journal of Technology Management 34, nos. 1-2: 105-25.

Powell, W.W., K.W. Koput, J.I. Bowie and L. Smith-Doerr. 2002. "The Spatial Clustering of Science and Capital: Accounting for Biotech Firm-Venture Capital Relationships," Regional Studies 36, no. 3: 291-305.

Prevezer, M. 2001. "Ingredients in the Early Development of the US Biotechnology Industry," Small Business Economics 17, noS. 1-2: 17-29.

Remer, S., S.H. Ang and C. Baden-Fuller. 2001. "Dealing with Uncertainties in the Biotechnology Industry: The Use of Real Options Reasoning," Journal of Commercial Biotechnology 8, no. 2: 95-105.

Reynolds, P.D., M. Hay, W.D. Bygrave, S.M. Camp and E. Antic. 2000. Global Entrepreneurship Monitor. Kansas City.

Rothaennel, F.T. 2001. "Incumbent's Advantage Through Exploiting Complementary Assets Via Inter-firm Cooperation," Strategic Management Journal 22, nos. 6-7: 687-99.

Sanders, W.G. and S. Biovie. 2004. "Sorting Things Out: Valuation of New Finns in Uncertain Markets," Strategic Management Journal 25, no. 2: 167-286.

Sapienza, H.J. 1992. "When do Venture Capitalists Add Value?," Journal of Business Venturing 7, no. 1: 9-28.

Shepherd, D.A. 1999. "Venture Capitalists' Assessment of New Venture Survival," Management Science 45, no. 5: 621-32.

Stolis, A. and D.E. Goodman. 2004. Patient Capital: Improving the Life of Millions-The Vital Role of Venture Capital in the Life Science Innovation. Arlington, Virginia: National Venture Capital Association.

Timmons, J.A. and W.D. Bygrave. 1986. "Venture Capital's Role in Financing Innovation for Economic Growth," Journal of Business Venturing 1, no. 2: 161-76.

Traore, N. 2005. "Access to Financing Capital by Canadian Innovative Biotechnology Finns," Statistics Canada, cat. # 88F0006XIE-No. 010.

Tyebjee, T.T. and A.V. Bruno. 1984. "A Model of Venture Capitalist Investment Activity," Management Science 30, no. 9: 1051-66.

Van Meter, K. 1990. "Methodological and Design Issues: Techniques for Assessing the Representativeness of Snowball Samples." Pp. 31-43 in AIDA Research Monograph. Rockville, MD: National Institute on Drug Abuse.

Wright, M., J. Kissane and A. Burrows. 2004. "Private Equity and the EU Accession Countries of Central and Eastern Europe," Journal of Private Equity 7, no. 3: 32-46.

Wright, M., A. Lockett and S. Pruthi. 2002. "Internationalization of Western Venture Capitalists Into Emerging Markets: Risk Assessment and Information in India," Small Business Economics 19, no. 1: 13-29.

Wright, M., A. Lockett, S. Pruthi, S. Manigart, H. Sapienza, P. Desbrieres and U. Hummel. 2004. "Venture Capital Investors, Capital Markets, Valuation and Information: US, Europe and Asia," Journal of International Entrepreneurship 2, no. 4: 305-26.

Wright, M. and K. Robbie. 1998. "Venture Capital and Private Equity: A Review and Synthesis," Journal of Business Finance and Accounting 25, no. 5: 521-70.

Zucker, L.G., M.R. Darby and J.S. Armstrong. 2002. "Commercializing Knowledge: University Science, Knowledge Capture, and Firm Performance in Biotechnology," Management Science 48, no. 1: 138-53.

Contact Information

For further information on this article, contact:

Holger Patzelt, Max Planck Institute of Economics, Kahlaische Str. 10, 07745 Jena, Germany

Tel.: +49 (0) 3641 686 726, fax: +49 (0) 3641 686 710

E-mail: patzelt@econ.mpg.de

Holger Patzelt, Max Planck Institute of Economics, Jena, Germany

Dodo zu Knyphausen-Aufsess, Chair for Human Resources and Organization Theory, University of Bamberg, Germany

Yasmin Habib, Chair for Human Resources and Organization Theory, University of Bamberg, Germany

Table l. Life sciences in North America (US and Canada) and Europe

                              North America         Europe

First company (foundation)   Genentech (1976)   Celltech (1980)
Total companies in 2006           1,917              1,621
Public companies in 2006           418                156
Revenues in 2006              62 billion WS     19 billion $US
Employees in 2006                188,000            75,810
VC investment in 2006        3.5 billion WS I   1.9 billion WS

Table 2. Characteristics of North American and European life
science VCs

                              VC             Std.    Mann-Whitney
VC characteristics         location   Mean   dev.   test statistics

Capital under management      NA      1349   1479         681
  (million $US)               EU      1181   2282
Number of firms in VC         NA      20.2   10.5         584
  portfolio                   EU      24.5   14.0
Life science team size        NA       8.8    4.8         550
                              EU       0.7    5.2

Asterisks relate to results of the Mann-Whitney-U test for differences
between the two groups of VCs, with * p < 0.1, ** p < 0.05, and
*** p < 0.01 level of significance.

Table 3. Portfolio strategies of North American and European
life science VCs

                          VC       Number
Binary variables       location   (of total)   Percentage

Investment stage          NA          32          49.2
  diversification         EU          10          43.5
Geographic proximity      EU          48          34.8

                       Chi-Square test
Binary variables         statistics

Investment stage            0.225
  diversification
Geographic proximity        12274 ***

Asterisks relate to results of the Chi-square test for differences
between the two groups of VCs, with * p < 0.1, **p < 0.05, and
*** p < 0.01 level of significance.

Table 4. Life science markets of North American and European
life science VCs

                                                     Mann-Whitney
Continuous         VC                                    test
variables        location   Mean share   Std. dev.    statistics

Therapeutics        NA        0.388        0.252         409 ***
                    EU        0.571        0.186
Diagnostics         NA        0.028        0.048         554 **
                    EU        0.043        0.044
Service/Supply      NA        0.125        0.110         654
                    EU        0.140        0.093
Medical             NA        0.340        0.241         454 ***
  technology        EU        0.184        0.147
Healthcare/IT       NA        0.101        0.132         503 **
                    EU        0.038        0.071
Others              NA        0.016        0.032         723
                    EU        0.027        0.056

Asterisks relate to results of the Mann-Whitney-U test for
differences between the two groups of VCs, with * p < 0.1,
** p < 0.05, and *** p < 0.01 level of significance.