Industry-specific specialization in venture capitalists’ internationalization decisions

Previous research examines several drivers of venture capital’s cross-border capital flows. These drivers mainly relate to the characteristics of the institutional environment and the country in which the investment is made. Investors prefer to invest in countries with highly developed institutional environments since, from an investor’s point of view, this implies greater market transparency and regulatory stability (Groh et al. 2010; Guler and Guillen 2010; Aizenman and Kendall 2012). In particular, the literature mentions that strong legal, financial, and political institutions facilitate cross-border venture capital flows (Devigne et al. 2018). Previous literature also reports that venture capitalists prefer investing in countries with cultural similarities to their own countries (Li et al. 2014; Dai and Nahata 2016). These similarities may include a common language, colonial ties (Aizenman and Kendall 2012), or economic integration between the countries (for example, the European Union) (Alhorr et al. 2008). With increasing cultural similarities, communication and trust between the participating parties—in this case, the investor and the portfolio company—increase (Sorenson and Stuart 2001). Hence, the investor’s liability of foreignness decreases when investing in a country with cultural similarities to the investor’s own country.

Furthermore, studies have found that the economic growth of the target market is a driver of international venture capital (Groh et al. 2010; Schertler and Tykvová 2011; Aizenman and Kendall 2012). Economic growth within a country can be represented by a dynamic development of entrepreneurial ecosystems, which may foster the creation of new companies (Neck et al. 2004). An entrepreneurial ecosystem is a system of interactions between individuals and organizations—for example, financial intermediaries, research institutions, suppliers, customers, and the government (Colombo et al. 2019a, b). It thus comprises the area in which start-ups are established. From the perspective of a foreign investor, this leads to attractive investment opportunities, particularly due to the framework created by a strong institutional environment for investors (Mack and Mayer 2016). In addition, constant low interest rates in global markets may have led to an increase in the popularity of venture capital as an asset class. This may have resulted in increased competition for interesting and promising investment opportunities, encouraging investors to explore international markets. Further, the amount of ‘dry powder’, defined as venture capitalists’ committed but unallocated capital on hand, has increased in the last years, which may have forced investors to seek investment opportunities beyond their national borders.

Despite the aforementioned drivers of international venture capital investments, comprising strong institutions and commonalities between countries, it is nevertheless argued that compared to domestic deals, international venture capital deals are characterized by higher information asymmetries and additional risks. These risks arise from the geographical, cultural, and institutional distance between the investor and the portfolio company (Lockett and Wright 2002; Pruthi et al. 2003; Wright et al. 2005; Devigne et al. 2016; Devigne et al. 2018). These risks are described as the liability of foreignness, which captures all additional costs arising for the investor that would not occur in a domestic deal (Zaheer 1995). This perspective, however, tends to focus on the institution-based view (Peng 2002). Likewise, the drivers of international venture capital flows mainly relate to institutional differences. At this point, the supplementary view focusing on the characteristics and capabilities of investors, following the resource-based view by Wernerfelt (1984), is missing. Under the circumstances of an international investment with high information asymmetries, an investor’s firm-specific knowledge can represent a competitive advantage relative to other investors who do not possess these characteristics.

Some previous findings examine the probability of investing abroad according to investor type. For example, some studies show that corporate venture capitalists exhibit a broader geographic scope, as they invest in companies from which their corporation can derive a strategic advantage. Such target companies require a certain fit with the parent company’s product and service offerings so the corporate investor can access the target’s technologies and knowledge. From a strategic point of view, limiting investment opportunities to the domestic market would reduce the number of target companies that fit the investor’s product (Gupta and Sapienza 1992; Bertoni et al. 2015).

Furthermore, existing literature describes the decision on whether to internationalize from the perspective of an optimal portfolio allocation. Investing in different countries leads to geographic portfolio diversification and therefore may reduce an investor’s overall investment risk (Knill 2009; Humphery-Jenner and Suchard 2013).

In addition, several studies consider the characteristics of an investor’s human capital (De Prijcker et al. 2012; Devigne et al. 2018). Regarding the international investment expertise of venture capitalists’ investment managers, there is evidence that more experienced managers are more likely to invest abroad (Schertler and Tykvová 2011). This is substantiated by their superior knowledge of the institutional environment and better access to networks (Devigne et al. 2018). In terms of an investor’s social capital (e.g., networks), research shows that venture capitalists’ social networks impact their international investment behaviours. To share the risks accompanying international investments, venture capitalists often decide not to invest in stand-alone deals rather than syndicates (Tykvová and Schertler 2014; Chemmanur et al. 2016). This practice enables venture capitalists to pool resources for monitoring, supporting and exiting their portfolio companies in international markets. In this context, local investors play a special role, as they are not affected by the liability of foreignness, and they know the local markets and behaviours. Hence, venture capitalists syndicating with local investors can solve problems arising abroad with the help of local partners (Sorenson and Stuart 2001; Cumming and Dai 2010; Vedula and Matusik 2017).

Rather than addressing the capabilities of external partners, this paper presents the firm-specific characteristics that influence internationalization decisions. This focus is meaningful because firms’ characteristics determine their success in enhancing their portfolio companies’ value and internal efficiency (Mahoney and Pandian 1992; Hart 1995; Abell and Nisar 2007). These characteristics include the investors’ level of industry-specific knowledge on technological product development and access to key resources like personnel, raw materials, and distribution channels—what could be considered their level of industry specialization (Gupta and Sapienza 1992). Industry specialization is not static; rather, organizations acquire it through experience and practice in the form of tacit knowledge (Itami and Roehl 1991; Hart 1995). Firms can learn directly from their investment experience and thus develop specialized knowledge (Gupta and Sapienza 1992; Liu and Maula 2016). From venture capitalists’ point of view, the provision of industry- and product-specific support and advice to portfolio companies is likely to be facilitated if firms have already developed their industry specialization in the past (Zhang and Pezeshkan 2016). Specialization development may help investors learn about the characteristics and behaviours of markets and their participants in a specific field of business, which is not limited to the domestic market but can also be applied globally. One the one hand, a higher level of specialization likely attenuates an investor’s liability of foreignness as the benefits of specialization in the market exceed the risks of cross-border investment. Through a high degree of industry specialization, the insecurity linked to investment abroad may be reduced by industry-specific expertise accumulated through similar previous investments. Therefore, it is assumed that industry specialization facilitates investments abroad. On the other hand, one could also argue that investors with increasing levels of industry specialization face fewer investment opportunities in their home markets if they want to continue investing in similar business models. Investment opportunities matching an investor’s industry specialization may not be available in the domestic market, resulting in a push effect to invest abroad. In summary, industry specialization may be a vehicle for reducing the liability of foreignness and facilitating investments abroad. In addition, international investments may result from an investor’s industry specialization. Consequently, the first hypothesis is as follows:

Hypothesis 1: Venture capitalists’ degree of specialization is positively related to the probability of internationalization.

Beyond venture capitalists’ industry specialization, a considerable body of research examines the mechanisms for overcoming information asymmetries within foreign markets. One strategy is observing and interpreting the determinants of the market and the portfolio company (Valliere 2012).

Regarding the determinants of the target market, the concept of the entrepreneurial ecosystem is widely used to represent increased entrepreneurship and growth in a certain region (Cohen 2006; Stam 2015; Spigel 2017). Entrepreneurial ecosystems are networks of social, cultural, political and economic elements in a region that contribute to the competitiveness and success of new ventures within the system (Dubini 1989; Spigel 2017). This is because favourable institutional conditions for growth are created in a developed entrepreneurial ecosystem: ease of firm communication and cooperation (Gertler 2003), wide social networks that create pathways for knowledge spillover (Powell et al. 2005), and strong connections between founders and founding sources (Powell et al. 2002). Evolved entrepreneurial ecosystems are present in start-up hubs, where the density of innovative companies, networks, and venture funding sources is high. From a venture capital investor’s point of view, the risk of a cross-border deal may be evaluated as lower if the investment occurs in an evolved entrepreneurial ecosystem. Thus, an evolved entrepreneurial ecosystem may act as a quality signal for the target market. Thereby reducing information asymmetries. In such regions, the value and success stories of other investors already exist, which venture capital firms can draw on in their decision-making. This ecosystem is missing in regions where only a few deals have been executed.

In sum, I assume that due to entrepreneurial ecosystems’ supporting and assisting effects, an investor’s industry specialization is less essential because it is locally available. Therefore, investors are not compelled to rely only on their own expertise. Under the assumption that venture capitalists assess an entrepreneurial ecosystem as highly developed, it may be that less specialized investors are more inclined to invest in start-up hubs than specialized investors, as less specialized investors are necessarily dependent on the assisting effect of start-up hubs. Thus, the second hypothesis is as follows:

Hypothesis 2a: If the portfolio company is located in an evolved entrepreneurial ecosystem, this negatively moderates the relationship between specialization and internationalization.

Another determinant influencing venture capital investors’ international investment decisions may be whether the portfolio company under consideration has a previous financing history. New venture financing usually proceeds in several rounds in which portfolio companies must reach milestones and demonstrate progress in their business activities (Gompers 1995). Each additional round carries certification effects from previous rounds, and an investment decision made in a later round is thus made with less uncertainty than decisions made in the first round (Ruhnka and Young 1987; Gompers 1995; Wang and Zhou 2004). Portfolio companies with longer financing histories are also likely to enter a more advanced stage of the life cycle; accordingly, investors may be faced with fewer investment risks, as the portfolio company’s first operational challenges have been solved by other investors who were already involved. Consequently, the information asymmetries for investors decrease with each additional round from which information about the portfolio company’s past performance is available. Therefore, previous financing rounds could mitigate new investors’ insecurities due to the skill sets, experiences, and networks of the investors who are already involved (Schertler and Tykvová 2012; Devigne et al. 2013). Hence, one can argue that a firm’s industry specialization is less important when an investment is made in a later round. Accordingly, the third hypothesis is as follows:

Hypothesis 2b: If the firm initially invests in a later round, this negatively moderates the relationship between specialization and internationalization.

The literature also relates differences in the amount of entrepreneurial funding to the characteristics of start-ups (Hsu 2007). If a foreign portfolio company has already received a higher amount of funding, a potential investor may consider this a sign of security. The more equity a company has already received from other parties, the higher potential investors rate the quality of the company. Furthermore, investors already involved in the portfolio company have an incentive not to lose their invested capital, as they would incur financial and reputational damage (Walter 2013). It can be assumed that to prevent losses, previous investors have already contributed to improving the portfolio company’s products and processes. Consequently, the company’s development has likely progressed further with higher financing. The existing information asymmetries are thus reduced for new investors. Regarding the necessity of specialized knowledge, additional technical support from a specialized investor may then be less necessary, as this support has already been provided. Therefore, I suspect that the funding a company has already received negatively moderates the effect of industry specialization on the probability of a cross-border deal. Thus, the fourth hypothesis is as follows:

Hypothesis 2c: The higher the portfolio company’s funding, the lower the influence of industry specialization on the probability of a cross-border deal.

Another proxy reducing information asymmetries between the portfolio company and the venture capital investor might be the portfolio company’s age. Particularly young ventures imply a higher risk associated with investment than older companies, and venture capitalists may be affected by this risk in their investment decisions (Zhang and Pezeshkan 2016). The age of the portfolio company indicates how long a company has been established on the market so far. Hence, the older a company is, the more trust it conveys towards investors. Furthermore, the availability of information about the company and its operations is higher for older companies, since they have a track record that investors can use to assess the performance of the company. Information asymmetries are thus lower for older companies.

Older companies are also more likely to have established business processes, a developed product and a customer base generating revenues. New investors can draw on these existing assets and do not have to build up the business from scratch. Regarding industry specialization, investors can access an already existing pool of knowledge, skills and resources. Their specialization is therefore unnecessary to the same extent as is the case with younger companies that still have little investment in their product development and the establishment of processes and customer generation. Thus, the fifth hypothesis states as follows:

Hypothesis 2d: The older the portfolio company, the lower the influence of industry specialization on the probability of a cross-border deal.

The main data set is taken from the Refinitiv database Eikon (formerly Thomson Financial), powered by VentureXpert. The data contain detailed information about independent venture capitalists’ internationalization decisions at the deal level. Information is included on the corresponding venture capital firms and referring portfolio companies for the last 2 decades (2001–2019). This period includes several global economic events such as the global financial crisis, which may have affected internationality and investment decisions in the venture capital industry. Some variables and further information on the institutional characteristics of the target markets were taken from the websites of Rafael La Porta and the Entrepreneurship and Development Institute, as well as from the International Monetary Fund. The data are analyzed in terms of each internationalization decision to initially invest in a company. A detailed overview and explanations of the variables are provided in Table 1.

Every observation contains information about the investing venture capital firm’s characteristics and the corresponding portfolio company into which the firm invested. Data at the institutional and cultural levels were collected separately from the investment deal information and merged into the main data set. Table 2 shows the descriptive statistics of the sample, which consists of 46,525 investments made by 2125 venture capital firms into 21,804 portfolio companies. An overview of the geographical distribution of venture capital firms and portfolio companies is presented in Table 3.

Only venture capitalists’ initial decisions to invest in a portfolio company are analyzed. These can occur in the first round or later rounds. Concerning follow-on investments in the same portfolio company, these choices are no longer decisions to internationalize in the sense that the portfolio company and its environment are no longer afflicted with uncertainty. As firms already gain information about the processes, stakeholders and business environment of the company in the first investment round, initial investment decisions and follow-on investments may differ substantially. There is likely a different intention behind the first investment in a company compared to that of a follow-on investment. For example, the dilution of the investor’s share may need to be protected if additional investors enter, and further capital may be required to achieve company-specific goals. Therefore, in a follow-on investment, the decision to internationalize is not new, nor is the addition of a new company to the portfolio. However, since firms realize industry-specific learning effects by investing in the same company multiple times, these investments are counted as further industry specialization.

The dependent variable cross-border deal is a binary variable indicating the internationalization decision of the venture capital firm under consideration. Cross-border deal equals zero if the deal is domestic and one if the deal is international.

The main independent variable, industry specialization, is a ratio variable reflecting the proportional deal experience of a venture capital firm within an industry measured against its previous investments before the deal under consideration. The measure is calculated using a 5-year rolling window as the timespan and, therefore, the number of deals increases over time. Since new investment firms entering the sample will have no experience at all due to their missing history in the sample. Therefore, the first 5 years of the data set and the first 5 years of a venture capital firm’s lifetime are not included in the analyses. Industry specialization equals zero if a firm has no experience in the industry of the deal being investigated. Industry specialization can take a maximal value of one if all deals a firm has made within the timespan of the rolling window can be assigned to the industry of the deal under consideration to date. The measurement of Industry specialization incorporates learning effects by assuming that with each additional deal within an industry, a firm’s experience increases through successfully supporting companies in that industry in the market. Therefore, for the first deal in a portfolio, the variable industry specialization always equals zero, as no learning effects can occur until a firm’s first investment.

Industry classification within the sample was conducted by Refinitiv Eikon Industry Minor Classification, and this classification has been adopted by this study. The specialization categories include biotechnology, communication and media, computer hardware, computer software and services, consumer-related, industrial/energy, internet-specific, medical/health, other products, and semiconductors.

Four different moderators referring to the portfolio company and the corresponding target market are considered. All moderators should illustrate the strength of the information asymmetries between the venture capital firm and the portfolio company. The first moderator is a Start-up hub dummy, which measures whether the region of the portfolio company’s headquarters is known for supporting the creation and growth of new ventures. Start-up hub dummy equals one if the portfolio company is located in a start-up hub and zero otherwise. Start-up hotspots were identified based on a data set of all tracked venture capital deals in the Refinitiv Eikon platform from 2001 to 2019. Cities with the most start-up companies in the corresponding year were labelled as start-up hotspots. For this purpose, the companies per city and year were counted, and those cities with values above the 75 percent quartile were marked as hotspots.

In addition to examining institutional factors, the study also examines known indicators of portfolio companies’ financing histories that may moderate the influence of industry specialization on decisions to invest abroad. The round in which the investment takes place (funding round) takes the value of the respective investment round. Log (company funding to date) is the logarithm of the financing in US dollars received by a portfolio company from all investors up to the respective investment date. Lastly, the age of the portfolio company Log (Age of portfolio company), is included in the models. The age of the portfolio company is measured as the logarithm of the difference between the founding year and the year of the investment decision.

Several control variables are included in the multivariate analyses. It is likely that in addition to industry specialization, knowledge of other institutional environments and cultures is also necessary; thus, the international investment experience is controlled for (Gupta and Sapienza 1992; Cumming and Dai 2010). International investment experience influences investment decisions because experienced investors are more familiar with foreign institutional and legal environments and have better access to international networks (Schertler and Tykvová 2011; De Prijcker et al. 2012; Devigne et al. 2018). In addition to knowledge of the particularities of international markets, the investors’ experience in domestic markets also influences their investment decisions. For example, through experience in domestic markets, investors gain skills in structuring deals and monitoring investments (Gompers et al. 2009; Wang and Wang 2011). Both international and domestic experience are measured by the logarithm of the number of deals a firm made prior to the year of the deal under consideration. Furthermore, the age of the firm (age of firm) as well as the size of the firm (firm size), based on the venture capital firm’s assets under management, are included as control variables. Age of firm is measured as the logarithmic value of the difference between the date of the establishment of the firm and the date of the investment. Particularly young ventures imply a higher risk associated with investment than older companies, and venture capitalists may be affected by this risk in their investment decisions (Zhang and Pezeshkan 2016).

In addition to the characteristics of investment firms and portfolio companies, the study also includes other structural and institutional characteristics of the target market.

In addition to the company’s industry specialization, the availability of promising investment alternatives in the investor’s home country could also play a role in an investor’s decision to internationalize. Thus, I add a control variable which considers this factor. The variable Local demand measures the number of venture capital deals in the industry of the deal under consideration in the home country of the venture capital firm prior to the year before the respective deal.

The efficiency of the legal system (efficiency legal system) impacts venture capital investments through more assertive mechanisms for solving agency and control problems (Cumming et al. 2010). The analyses in this study use the index created by Porta et al. (1998) to illustrate legal efficiency. Efficiency legal system can take values from zero to ten, with low scores indicating low efficiency.

Beyond the aforementioned formal institutions, investment behaviours are also likely to be influenced by informal institutions such as cultural heritage (Bruton et al. 2005; Gantenbein et al. 2019). People speaking the same language or sharing the same religion often share similar values (Gantenbein et al. 2019); therefore, some authors have used a country’s language and religion as proxies for culture (Grinblatt and Keloharju 2000; Guiso et al. 2006). To control for cultural differences between the investment nation and the nation of the portfolio company, two dummy variables that equal one if the two countries share the same language or religion are included in the models. The variables cover cultural differences among the main languages and religions of the data set. The main official languages of firm nations and target nations are Arab, English, French, German, Spanish, and Portuguese. Data about countries’ official languages are taken from the CIA World Factbook. The religion dummy variable covers Buddhism, Christianity, Hinduism, and Islam. Data on countries’ denominations are also taken from the CIA World Factbook. Since languages or beliefs that are only represented in one country act as a fixed effect for that country, these could not be integrated. To further control for country-specific cultural peculiarities, in particular, those related to entrepreneurship and financing, the global entrepreneurship score developed by Acs et al. (2017) is included in the models. The global entrepreneurship score measures the health of the entrepreneurship ecosystem annually and is provided by the global entrepreneurship development institute (GEDI). GEDI collects data on the entrepreneurial attitudes, abilities, and aspirations of local populations and weights these against the prevailing social and economic infrastructure of a country. The score consists of various components, for example, the average of a population’s willingness to take risks, the quality of a country’s human capital, and a country’s financial openness.

Year and legal effects are also included in the analyses. Legal classification fixed effects capture the legal classification of the target nation based on the ‘Legal Classification of Investment Nation’ by Porta et al. (1998). Countries are classified as French, English, German, Scandinavian or Socialist.

In the sample, 8855 deals are labelled cross-border deals, and 37,670 are labelled domestic deals. Therefore, almost 20% of all deals in the sample are characterized as cross-border deals.

On average, the firms’ industry specialization is 0.286, meaning that on average, 28.6% of the firms’ portfolio investments at the time of the investment decision can be assigned to the industry of the company in which the investment is made. Therefore, on average, firms do not invest in a company without any prior industry experience. The maximum industry specialization in the sample is 1, and the minimum value is no industry specialization at all. Most deals can be classified into the categories of computer software and services (33.99%), internet-specific (26.7%), biotechnology (8.91%), and medical/health (8.16%). The communications and media category has the fewest deals (2.67%). A detailed overview of the firms’ industry specializations is displayed in Table 4.

Slightly less than 30% of the deals included in the sample are located in a start-up hotspot (mean = 0.293). The average funding a portfolio company received (measured in logarithmic values) at the date of the investment of the firm under consideration is 16.746 US dollars (SD = 1.76). Firms are on average 2.403 years old and at most 5.13 years old at the time of the investments (also measured in logarithmic values). The average logarithmic age of the portfolio companies is 1.26 years and the maximum age is 5.537 years. The earliest investments into portfolio companies occur in the year of establishment. On average, investments into portfolio companies occurred after two rounds (SD = 1.701). The mean of the number of investment rounds seems relatively small. However, this may be because the data set only includes initial investments in companies and does not include follow-on investments in later rounds. Before the investments, venture capital firms had executed an average of 2.69 deals in their home countries and 1.68 deals abroad. The local demand for venture capital financing in the specific industry under consideration in the investor’s home market is on average 0.208. This means, that in the investors’ home country, on average 20.8% of the deals in the previous year were made in the sector of the deal under observation.

Most deals occur in countries with legal systems that are considered efficient (mean = 9.64). The lowest efficiency score for a legal system is 2.5 points and the highest is 10 points, the maximum efficiency. Similar patterns can also be observed regarding the financial openness of the target markets. The average financial openness of the target markets is 0.954 points, with 1 as the maximum score and 0 as the lowest score. Investment deals primarily occur in countries with the same official language (mean = 0.834) and the same religious orientation (mean = 0.924). Deals also mainly occur in countries with high global entrepreneurship scores (mean = 74.681). The countries with the lowest global entrepreneurship are assigned a score of 11.8, while some countries reach maximum scores of 90.23.

Within the framework of multivariate analysis, this paper examines if and to what extent a firm’s industry specialization impacts its international investment decisions. Furthermore, this work analyzes the extent to which characteristics of the portfolio company and the corresponding target market influence the relationship between a firm’s industry specialization and the decision to invest abroad. Binomial logistic regression models are conducted to investigate the relationship between a firm’s industry specialization and internationalization decisions. Therefore, the dependent variable P (Cross-border deal = 1) indicates the probability of a cross-border deal. The base model specification, investigating the isolated effect of a firm’s industry specialization on the probability of investing abroad, is:where investment decisions are indexed by i.

To avoid biased regression estimates, all regression models control for possible cross-correlation effects arising from unobserved individual firm effects among all deals made by a respective firm. Therefore, clustered standard errors are used. Pearson correlations of all variables included in the models are displayed in Table 5. As variance inflation factors (VIFs) do not exceed values of 6.11 (see also Table 5), there is no evidence of multicollinearity.

The models analyzing the interaction effects have the same model specification, including the respective additional interaction effect. All interaction effects relate to characteristics reflecting informational asymmetries between the venture capital firm and the portfolio company, which I expect to moderate the impact that industry specialization has on the investor’s internationalization decision.

Model 2 includes the moderating effect of the portfolio company being located in a start-up hub on the impact that industry specialization has on the probability a firm invests abroad. The model specification is the following:

Model 3 analyzes the moderating effect of the funding round on the impact that industry specialization has on the probability a firm invests abroad. The model specification is as follows:

Model 4 analyzes the moderating effect of the portfolio company’s funding on the impact that industry specialization has on the probability that a firm invests abroad. The model specification is as follows:

Model 5 analyzes the moderating effect of the portfolio company’s age on the impact that industry specialization has on the probability a firm invests abroad. The model specification is as follows:

The venture capital industry is mainly located in the North American market, including the United States and Canada (Tykvová 2018). This is reflected in the sample: 72.83% of all deals refer to the North American market. Therefore, it may be questionable whether the results from the entire global data set can be generalized as they are dominated by the North American market. For this reason, a robustness check excluding the North American market was conducted. The models for the reduced data set were calculated in the same way as they were for the entire sample.

As shown in Table 7, the results are largely robust to changes in the data set. Again, industry specialization has a positive impact on the probability of a cross-border deal. However, the moderating effect of the financing round and the portfolio company’s age show no statistical significance. One possible reason for this could be that the certification effects associated with later rounds are less meaningful to investors outside the North American market. Harrison and Mason (2019) describe the European market as being inferior to the American market in terms of the conditions for entrepreneurial growth. Portfolio companies’ employees are less trained in growth and innovation and the corresponding leadership capacity is also less present in companies (Harrison and Mason 2019). Investors, therefore, must increasingly engage themselves in companies to drive growth. Industry specialization might therefore be equally important in different financing stages.

Syndication is widely discussed in prior venture capital literature as a strategy to reduce information asymmetries and pooling resources for monitoring and supporting portfolio companies (Dai and Nahata 2016; Liu and Maula 2016) in international deals. It is assumed that syndication thus reduces the investment risk, as the risk can be shared among different investors (Schertler and Tykvová 2012; Devigne et al. 2013). To account for the impact of syndication on the firm’s internationalization, I ran all models with and without a syndication-dummy control variable. The results remain robust in both samples with and without the North American market. In that the role of local investors as syndicate partners in internationalization decisions is emphasized in the literature, I also ran the models including a control variable accounting for the number of local partners participating in the deal. Likewise, the results are robust.