BUSINESS( NO PLAGARISM A+ WORK, ON TIME)

Pelicans!!322

wk4Cross-border_venture_capital_i.pdf

Home >Business & Finance homework help >Management homework help >BUSINESS( NO PLAGARISM A+ WORK, ON TIME)

Cross-border venture capital investments:

The impact of foreignness on returns

Axel Buchner1, Susanne Espenlaub2, Arif Khurshed2 and Abdulkadir Mohamed3

1Department of Business and Economics,

University of Passau, Passau, Germany; 2Alliance Manchester Business School, University of

Manchester, Manchester, UK; 3Cranfield School of

Management, Cranfield MK43 0AL, UK

Correspondence: A Mohamed, Cranfield School of Management, Cranfield MK43 0AL, UK. Tel: +44 1234 750111; e-mail: [email protected]

Abstract Against the background of the growing internationalization of venture capital

(VC) investing, this is the first global comparison of the returns generated by individual domestic and cross-border deals. We examine investments

worldwide during 1971–2009 and find that cross-border investments

significantly underperform compared with equivalent domestic investments. Returns are negatively affected by geographic distances, cultural disparities,

and institutional differences between the home and host countries. Returns on

cross-border and domestic deals also decline after the late 1990s. International portfolio diversification and the saturation of domestic markets may explain

why VC investors make cross-border investments despite poor expected

returns.

Journal of International Business Studies (2018) 49, 575–604. https://doi.org/10.1057/s41267-017-0116-6

Keywords: venture capital; cross-border; return; IRR; PME; foreignness; distance

INTRODUCTION This study examines the returns generated by venture capital (VC) investments in domestic and cross-border deals. Venture capital (VC) firms are specialized financial intermediaries which raise funds from investors. The VC firms then invest the funds in innovative new businesses, so-called portfolio companies, with a view to realizing their investments after approximately 5–7 years (e.g. Sahlman 1990; Black and Gilson 1998; Gompers and Lerner 2004). In a domestic deal, a VC firm invests in its home country; in a cross-border deal, it invests outside its home country. VC firms are experts at investing in inherently risky and informationally opaque start-up ventures (e.g. Gorman & Sahlman, 1989; Gompers, 1995; Amit, Brander, & Zott, 1998). The high information asymmetries involved in such investments give rise to adverse selection prior to investment and agency conflicts post-investment. In order to limit these problems, VC firms closely screen potential investee compa- nies; conduct careful due diligence; and align entrepreneurs’ incentives with firm value through monitoring, governance, con- tracts, and other mechanisms, including staged financing (e.g. Sahlman, 1990; Admati & Pfleiderer, 1994; Wright & Robbie, 1998; Manigart & Wright, 2013). By resolving information problems and

Received: 16 March 2016 Revised: 7 September 2017 Accepted: 9 September 2017 Online publication date: 23 October 2017

www.jibs.net

http://www.jibs.net/

incentive conflicts, and by providing portfolio companies with advice, expertise, and access to networks, VC investors are able to add value to their investments (e.g. Gorman & Sahlman, 1989; Sapienza, 1992; Sapienza, Manigart, & Vermeir, 1996; Devigne, Vanacker, Manigart, & Paeleman, 2013).

In apparent defiance of this view, the two decades prior to the financial crisis of 2007 saw a large increase in the number and size of cross- border VC investments. Aizenman & Kendall (2012) report an increase in worldwide cross-border VC investment deals from 15% of global deals in the early 1990s to over 40% in 2007. More recently, cross-border investing by US VC firms has risen sharply, with early-stage VC investments increasing from under 10% to more than 30% of VC deals in 2013 (Wuebker et al. 2016). In this context, more than 70% of VC deals in Asia are funded by foreign VC firms (Dai et al. 2012).

Against this background of increasing VC inter- nationalization, we examine the returns perfor- mance of cross-border and domestic VC investments. VC investors may require higher or lower returns from cross-border investments than from domestic investments. On the one hand, VC firms investing abroad are likely to encounter ‘liabilities of foreignness’ due to geographic dis- tances, cultural disparities, and institutional differ- ences between VC investors and their portfolio companies (Zaheer 1995; Wright, Pruthi, & Lockett, 2005; Sojli & Tham, 2017; Taussig, 2017; Wu & Salomon, 2017). As a result, cross-border investing gives rise to higher transaction costs (e.g. Portes & Rey, 2005) and greater costs due to more severe information asymmetries and agency conflicts (Wright & Robbie, 1998; Wuebker et al., 2016). In this case, VC investors require higher returns from

cross-border investments to compensate for the additional costs. On the other hand, cross-border investing facilitates portfolio diversification; thus, VC investors with portfolios predominantly invested in domestic ventures may accept lower returns from cross-border investments (e.g. Poterba & French, 1991). High levels of VC funds chasing limited numbers of promising investment oppor- tunities may also drive VC investors to resort to cross-border investing, even though they expect these investments to generate relatively low returns (Gompers & Lerner, 2000). In a previous study of cross-border returns, Cumming, Fleming, and Sch- wienbacher (2009) find that Asian-Pacific VC firms investing in US-based portfolio companies experi- ence returns that are lower than their domestic investments. Cumming et al. (2009) use hand- collected and largely proprietary data to assess the performance of 468 individual investments by VC firms in 12 Asian countries during 1989–2001 based on each investment’s internal rate of return (IRR). While Cumming et al. (2009) are able to calculate

investment returns due to their access to propri- etary data, most other previous VC studies lack sufficiently detailed deal-level data to compute direct measures of performance such as IRR or public market equivalent (PME) for individual investments. Because of data limitations, most studies measure performance in terms of the like- lihood of successful VC exit (e.g. Hochberg, Ljungqvist, & Lu 2007) rather than IRR or PME. Bengtsson and Hsu (2015) explicitly note the use of exit success as a limitation of their analysis.1

Devigne, Manigart, and Wright (2016) highlight that the existing evidence on VC returns at deal level is limited and call for further research in order to understand the variation of returns. We are able to overcome this limitation by using detailed cash flow data on individual VC investments obtained from the Centre of Private Equity Research (CEPRES) to calculate actual returns on individual VC investments.2 Our study contributes to the literature by comparing the returns of individual domestic and cross-border VC investments using a sample of 6529 domestic and cross-border VC deals made worldwide during 1971–2009. To the best of our knowledge, this is the first study to find that the underperformance of cross-border investments rel- ative to equivalent domestic investments is a global and persistent phenomenon. We base this conclu- sion on our analysis of the returns and other performance measures of a broad global sample of VC investments, comprising large numbers of

Cross-border venture-capital investments Axel Buchner et al

576

Journal of International Business Studies

home and destination countries and spanning more than three decades.

Our results show that cross-border deals generate lower returns than equivalent domestic deals in terms of return-based performance. We address differences in the targeted selection and risk of VC investors and conclude that the observed return differentials amount to cross-border underperfor- mance. We find that geographic distances, cultural disparities, and institutional differences between the home countries of VC investors and portfolio companies negatively affect cross-border returns. Additional tests show that VC firms benefit from cross-border investing by achieving portfolio diver- sification and overcoming shortages of domestic investment opportunities in saturated markets. Diversification, and the saturation of domestic markets, may explain why VC investors are attracted to cross-border investing despite the poor returns performance of cross-border deals.

Our study helps to resolve conflicting evidence in prior studies which examine the effect of geo- graphic distances between US VC investors and portfolio companies on VC exit performance (Chen et al., 2010; Cumming & Dai, 2010; Bengtsson & Hsu, 2015). For example, Chen et al. (2010) report that non-local deals outperform local deals in terms of initial public offering (IPO) exit probabilities, while Cumming and Dai (2010) find that local exit deals outperform non-local deals. Bengtsson and Hsu (2015) focus on the ethnicity of VC investors and founders of entrepreneurial companies and find that while shared ethnicity increases the likelihood of investment, it reduces exit perfor- mance. We explore what happens when VC invest- ments cross borders and examine the impact of geographic distances, cultural disparities, and insti- tutional differences in a similar way to the studies of Nahata, Hazarika, and Tandon (2014) and Dai and Nahata (2016). Like other cross-border perfor- mance studies, Nahata et al. (2014) and Dai and Nahata (2016) measure performance purely in terms of IPO exit probabilities

Our study lies at the intersection of the literature of several academicfields includingeconomics, finance, entrepreneurship, management, and international business, as outlined in ‘‘Literature, Conceptual Framework, and Hypotheses’’ section. Focusing lar- gely on entrepreneurial finance, our study con- tributes to the growing literature on the internationalization of VC and private equity (PE) investment and on cross-border VC/PE activity and flow (e.g. Schertler & Tykvová, 2011; Tykvová &

Schertler, 2011; Schertler & Tykvová, 2012; Dai et al., 2012; Li & Zahra, 2012; Cao, Cumming, Qian, & Wang, 2015). It also contributes to the literature on the performance of cross-border VC investments in terms of the ability of VC investors to achieve successful exits (e.g. Wang & Wang, 2012; Humph- ery-Jenner & Suchard, 2013; Bertoni & Groh 2014; Nahata et al., 2014; Cumming, Knill, & Syvrud, 2016; Dai & Nahata, 2016). In the context of this literature, our study builds on the existing research about the ways in which exit performance is affected by geographic distances, cultural/ethnic disparities, and institutional differences between the locations of US VC providers and those of their portfolio companies (Chen et al., 2010;Cumming&Dai, 2010; Bengtsson&Hsu, 2015) and between the locations of international VC providers and their portfolio com- panies (Nahata et al., 2014; Dai &Nahata, 2016). Our study is also related to the literature on syndication and networks in cross-border investments (Hursti & Maula, 2007; Guler & Guillén, 2010; Meuleman & Wright, 2011; Jääskeläinen & Maula, 2014; Reuer & Ragozzino, 2014; Hain et al., 2015; Chemmanur, Hull, & Krishnan, 2016; Meuleman, Jääskeläinen, Maula, &Wright, 2017). The remainder of this article is organized as

follows. ‘‘Literature, Conceptual Framework, and Hypotheses’’ section reviews the literature and develops our hypotheses. ‘‘Data and Methodology’’ section describes the data and methodology, and ‘‘Results’’ section presents our analysis and results. ‘‘Conclusion’’ section concludes the article.

LITERATURE, CONCEPTUAL FRAMEWORK, AND HYPOTHESES

VC firms are experts at bearing risk and dealing with the information and agency problems (adverse selection and moral hazard) which complicate investments in promising, young entrepreneurial businesses characterized by high information asym- metries, high risk, and high potential (e.g. Gorman & Sahlman, 1989; Gompers, 1995; Amit et al., 1998). Such ventures typically have little or no trading records or other information, such as balance sheets and past cash flow data, which are used in traditional valuation methods. Moreover, these ventures frequently operate in innovative industries with no established benchmark compa- nies. VC firms have developed methods to limit the information and agency problems arising from such investments. Such methods include screening,

Cross-border venture-capital investments Axel Buchner et al

577

Journal of International Business Studies

due diligence, contracting, monitoring, gover- nance, and staged financing (e.g. Sahlman, 1990; Admati & Pfleiderer, 1994; Wright & Robbie, 1998; Manigart & Wright, 2013).

The effectiveness of these specialized methods, mechanisms, and practices is often believed to depend crucially on VC investors’ familiarity with local markets; their access to local information, knowledge, and networks; and the proximity between VC investors and their investee (portfolio) companies in order to maintain close links, fre- quent interaction, and valuable reputational capital (e.g. Cumming & Johan, 2007; Chen et al., 2010; Dai et al., 2012; Hain et al., 2015; Wuebker et al., 2016). Significant distances between VC firms and portfolio companies are likely to increase informa- tion asymmetries between investors and investees, causing more pronounced adverse selection and moral hazard (e.g. Dai et al. 2012; De Prijcker, Manigart, Wright, & De Maeseneire, 2012; Hain et al., 2015; Dai & Nahata, 2016). Distance may thus increase the costs to VC firms of identifying and screening suitable investment opportunities (Cumming & Dai, 2010; Wuebker et al., 2016). VC firms typically conduct due diligence which is more rigorous, and hence, costlier for remote ventures (Nahata et al., 2014). In this regard, Wright, Lockett, and Pruthi (2005) report significant differ- ences in the risk assessments and information sources which are used in targeted selection between foreign and domestic investors.

Non-local and cross-border investors are at a disadvantage relative to local investors in terms of access to portfolio companies, local information, networks, reputational capital, and resources, and typically incur higher information and transaction costs (Nahata et al., 2014). The higher costs of contracting and monitoring cross-border portfolio companies result in lower firm value and lower value added (Sapienza et al., 1996; Sorensen & Stuart, 2001; Dai et al., 2012; Wuebker et al., 2016). Mäkelä and Maula (2008) confirm that local VC firms are more knowledgeable about portfolio companies’ markets than foreign VC firms. The general partners of VC firms provide advice and monitoring to portfolio companies during meet- ings held at the companies’ offices; in this context, geographic distance increases the costs and the amount of time involved for the partners (Hain et al., 2015). Chemmanur et al. (2016) find that the absence of (geographic) proximity makes it more difficult for VC firms to move scarce human capital, such as skilled general partners of VC firms, to a

portfolio company’s location. As a result, it is costlier to screen, monitor, advise, and support more distant portfolio companies. This advantage of local investors and invest-

ments is the focus of several studies from a range of academic disciplines including entrepreneurial and corporate finance, economics, and international business. Studies in international business and management, based on the seminal study by Zaheer (1995), refer to the disadvantage of not being local as the ‘liability of foreignness’. Studies in asset pricing and corporate finance refer to investors’ preferences for more familiar local rather than non-local investments such as ‘home bias’ (Poterba & French, 1991; Coval & Moskowitz, 1999). Other studies show that VC firms investing abroad encounter the ‘liability of foreignness’ problem (Wright et al., 2005; Nahata et al., 2014; Dai & Nahata, 2016). In order to compensate for the higher costs of

cross-border investing as a result of the liability of foreignness, VC firms are likely to require higher returns from cross-border investments than from equivalent domestic investments. However, cross- border investing facilitates portfolio diversification; thus, VC investors with portfolios predominantly invested in domestic ventures may accept lower returns from cross-border investments (e.g. Poterba & French, 1991). A lack of promising investment opportunities in their home countries, or rigorous competition for attractive deals because of excess funds available to VC investors, may also motivate VC firms to embark on cross-border ventures even if they expect them to generate relatively low returns (Gompers & Lerner, 2000). With regard to 53 VC funds based in 12 countries in the Asia-Pacific region, Cumming et al. (2009) find that these funds achieve lower internal rates of return (IRR) for investments in US portfolio companies than for domestic investments. At a global level, whether VC investors require higher or lower returns from cross-border investments than from domestic investments remains a question to be resolved empirically. Comparisons of the performance of VC firms

show that the firms’ selection criteria and invest- ment behaviour may also differ between home and cross-border investments. Empirical evidence con- firms that VC investors select different types of venture at home and abroad. The results of Dai et al. (2012) suggest that VC investors mitigate the higher information and monitoring costs of invest- ing abroad by investing in later financing rounds

Cross-border venture-capital investments Axel Buchner et al

578

Journal of International Business Studies

and in larger, more mature companies which are more transparent and less costly to screen and monitor. VC investors also self-select cross-border investments. Cumming and Dai (2010) find that investments in more distant firms tend to be undertaken by more reputable and experienced VC investors acting in syndicates in order to spread the risk. Thus when comparing the returns gener- ated by domestic and cross-border investments, it is essential to control for targeted selection and VC self-selection in order to compare equivalent domestic and cross-border investments.

Based on the foregoing reasoning, we formulate our first testable hypothesis as follows.

Hypothesis 1: All else being equal, there is no difference in performance between domestic and cross-border VC investments.

We measure absolute returns in terms of IRR and also measure the returns relative to a market benchmark in the form of PME. In addition to returns, we also examine exit performance. To the best of our knowledge, all prior studies of non-local and cross-border VC investments measure perfor- mance in terms of exit success. Following Giot and Schwienbacher (2007), they evaluate performance on the basis of whether and how quickly VC investors successfully exit their investments. Exist- ing evidence on the exit performance of invest- ments by US VC firms in local and non-local US portfolio companies is mixed (Chen et al., 2010; Cumming & Dai, 2010; Bengtsson & Hsu, 2015). While Chen et al. (2010) find that non-local deals of US VC firms outperform local deals in terms of their IPO exit probabilities, Cumming and Dai (2010) discover that local exit deals outperform non-local deals. Bengtsson and Hsu (2015) focus on the ethnicities of VC investors and founders of entrepreneurial companies and find that shared ethnicity increases the likelihood of investment but reduces exit performance. Our analysis of VC investments across borders is closely related to Dai et al. (2012), Li, Vertinsky, and Li (2014), Nahata et al. (2014), and Dai and Nahata (2016).

It is possible that foreign VC firms are less committed to their portfolio companies than local VC firms and may withdraw more quickly when a portfolio company’s performance is disappointing. In this regard,VCfirms’ premature exitsmaydamage portfolio companies (Mäkelä & Maula, 2008). Alter- natively, their lower levels of commitment and local embeddednessmay enable foreign VC firms tomake

more efficient exit decisions (Devigne et al., 2016). Whatever the circumstances, exit decisions may have implications for returns: foreign investors’ premature exits may result in reduced returns, while more efficient exit decisions may lead to increased returns for cross-border investments. Longer periods of investment in portfolio companies not only cause higher monitoring costs but also liquidity problems for VC backers. If cross-border investments require greater effort and money spent on advising and monitoring portfolio companies, the higher costs of carrying cross-border investments relative to domes- tic investments may tip the balance in favour of earlier exits from cross-border investments (Cum- ming & Johan, 2007; Espenlaub, Khurshed, & Mohamed, 2015). In the final part of our analysis, we explore

possible reasons why the returns of domestic and cross-border investments differ. In Hypothesis 1, we focus on a binary definition of foreignness. Now, we examine the impacts of geographic distances, cultural disparities, and institutional differences. Studies show that geographic distances, cultural disparities, and institutional differences between home and host countries affect the exit perfor- mance of VC cross-border investments (Mäkelä & Maula, 2008; Guler & Guillén, 2010; Li et al., 2014; Nahata et al., 2014; Dai & Nahata, 2016). Evidence from the literature suggests that geographic dis- tances and institutional differences have a negative impact on exit performance while the direction of the impact of cultural disparities is mixed. Our study contributes to this literature by examining the impact of geographic distances, cultural dispar- ities, and institutional differences on returns rather than on exit success. First, we explore whether there is a difference

between investments across a single, shared land border and investments across multiple borders. Investments across a single land border may involve less costly travel. Chemmanur et al. (2016) find that greater travel time adversely affects the exit success of cross-border VC investments. However, travel costs may depend to a greater extent on geographic distance; thus, in the next step, we examine the impact of geographic dis- tance. Prior studies show mixed results with some reporting that greater geographic distance reduces exit performance (Cumming & Dai, 2010) while others report a positive effect (Chen et al., 2010) or no significant effect after controlling for other measures of distance (Li et al., 2014; Dai & Nahata,

Cross-border venture-capital investments Axel Buchner et al

579

Journal of International Business Studies

2016). Consequently, we formulate the following hypothesis.

Hypothesis 2a: A VC firm’s performance in cross-border deals is unrelated to the geographic distance between the home countries of the VC firm and its portfolio company (after controlling for other measures of distance).

Because of colonial linkages, geographically dis- tant countriesmay share similar languages, cultures, and institutions. However, because of historical accidents and conflicts, nearby or neighbouring countries may differ greatly not just in terms of their languages but also their cultures and institutional frameworks. Consequently, because geographic dis- tances may not adequately capture the liability of foreignness, we also examine cultural disparities and institutional differences, following Li et al. (2014), Nahata et al. (2014), and Dai and Nahata (2016). VC investors’ lack of awareness of local cultural and social practices in unfamiliar cross-border environ- ments can be a source of conflict between a VC firm and its portfolio company, thereby increasing agency costs and reducing VC performance (Nahata et al., 2014). Cultural disparities and institutional differences can adversely affect levels of trust, repu- tation, financial contracting, and company perfor- mance (Li et al., 2014; Nahata et al., 2014). Cultural disparity is commonly measured using Kogut and Singh’s (1988) approach, based on the cultural measures (power distance, individualism, masculin- ity, and uncertainty avoidance) developed by Hofst- ede (1980). This approach is also used by Li et al. (2014), Nahata et al. (2014), Hain et al. (2015), and Dai andNahata (2016), amongothers. Li et al. (2014) find that cultural disparity reduces exit success, while Nahata et al. (2014) report that greater cultural disparity increases exit success. Nahata et al. (2014) argue that greater cultural disparity motivates VC investors to engage in closer pre-investment due diligence and screening; this in turn increases exit success. Focusing on the impact of cultural disparity on returns performance (as opposed to exit success), we test the following hypothesis.

Hypothesis 2b: A VC firm’s performance in cross-border deals is negatively related to the greater cultural disparity between the home coun- tries of the VC firm and its portfolio company.

VC investors encounter greater unfamiliarity and liability of foreignness in countries with institutional frameworks which differ from those in their home

country. In a different institutional environment, a VC firm’s familiar practices are likely to be at odds with local institutionalized practices with regard to deal selection, contracting, monitoring, and advising (Li et al., 2014). For example, VC firms from countries with strict and well-enforced legal rules and regula- tions rely on financial and accounting information to evaluate proposals and assess investment risk. How- ever, in countries with weak institutional environ- ments, VC firms must depend instead on personal contacts in order to access relevant information and enforce agreements (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 1998; Cumming, Fleming, & Schwienbacher, 2006; Cumming, Schmidt, & Walz, 2010; Li et al. 2014). VC firms which are used to the effective legal protection of investors and contract enforcement in their home countries find they can no longer employ complex, state-contingent con- tracts in host countries with weak legal institutions (Guler and Guillén 2010). In this regard, Chemma- nur et al. (2016) examine the impact of legal systems on exit success but find no significant impact. Measuring institutional differences using the World Governance Index, Li et al. (2014) find that such differences significantly reduce VC exit success. We examine the impact on VC cross-border returns of three dimensions of institutional difference: the difference in the legal systems of home and host countries (based on La Porta et al. 1998, similar to Chemmanur et al. 2016), and differences in regula- tory quality and political stability (similar to Li et al., 2014). Thus we formulate the following hypothesis.

Hypothesis 2c: A VC firm’s performance in cross-border deals is negatively related to pro- nounced institutional differences between the home countries of the VC firm and its portfolio company.

DATA AND METHODOLOGY

Data Sources and Sample Our data on individual VC investments are obtained from the Centre of Private Equity Research (CEPRES).3 CEPRES and its data are described in detail in Franzoni, Nowak, and Phalippou (2012). CEPRES data are used in a number of studies, including those of Krohmer, Lauterbach, and Calanog (2009), Cumming et al. (2010), Cumming and Walz (2010), and Franzoni et al. (2012).

Cross-border venture-capital investments Axel Buchner et al

580

Journal of International Business Studies

Through their special data-collection method (based on the so-called ‘Private Equity Analyser’), CEPRES effectively anonymizes all information relating to investments in order to meet the confidentiality requirements of the VC and PE firms which provide data to CEPRES. This tech- nique means that no third parties are able to identify the performance of individual firms, funds, or managers. The importance of such anonymity is that it eliminates the incentives for VC and PE firms to overstate the results which they report to CEPRES. The lack of anonymity in other databases may result in overstating and backfilling informa- tion, a situation which amounts to positive self- reporting bias.

Another important advantage of the CEPRES database is the availability of detailed information on cash flow at the level of individual VC invest- ments. Other databases either lack this information or provide cash flow or IRR data only at the fund level.

We start with all 14,224 observations in CEPRES for VC investments made from January 1971 to December 2009. We exclude 2484 partial exits and non-exits and 5057 buyout investments.4 Of the remaining 6683 observations, we have insufficient cash flow data for 154 deals. This leaves us with 6529 observations on fully realized VC investments exited through IPOs, mergers and acquisitions (M&A), or liquidations (write-offs). We split our sample into four geographic regions in accordance with the locations of the VC investors: North America, Europe (excluding the UK), the UK, and the rest of the world (ROW). Our sample comprises 4334 observations for North America, 839 for Europe, 363 for the UK, and 993 for the ROW.5

We classify investments as domestic (cross-border) if a VC firm and its portfolio company are located in the same country (different countries).6

Methodology In this section, we discuss the methodology used in our analysis. In order to measure the financial returns of VC investments, we calculate IRR based on all cash flow to VC investors (both outflow and inflow). This cash flow is reported in the CEPRES database. Except for a few studies which use proprietary data (e.g. Cumming et al., 2009), most prior research is limited to observing IRR at the fund level. Because we have access to cash flow data, we are able to calculate IRR based on actual (not proxied) cash flow at the level of individual VC investments.7 The cash flow is converted into US

dollars, following the approach adopted by Fran- zoni et al. (2012). Cash flow is not adjusted for management fees, interest, or carried interest. VC firms commonly use IRR to evaluate their invest- ments in-house and are often reluctant to disclose IRR figures. Moreover, if the firms do disclose their IRR figures, they have incentives to overstate them. As a result, reliable IRR data at the level of individual VC investments were not previously available to researchers. Some studies are able to calculate IRR but only at the level of an overall VC fund (rather than at the level of individual invest- ments held within a fund, as we do here). However, an understanding of investment level returns is crucial for VC firms so that they can allocate capital efficiently between domestic and cross-border investments and for VC fund investors so that they can select appropriate funds. We observe the stream of cash flow between the

start date of each investment and the final liquida- tion (exit) date. We then calculate the IRR as the discount rate which equates the present value of the net cash flow to zero. The cash flow consists of investments in portfolio companies, dividend repayments, and proceeds from exiting the investments.8

In addition to IRR, our analysis uses PME. IRR is an absolute measure of performance in the sense that it is not measured relative to a benchmark. In contrast, PME is a relative performance measure which compares a VC investment to an equiva- lently timed investment in the relevant public market. PME has been interpreted as a market- adjusted multiple of invested capital in that a PME greater than one means that investors in a given VC deal gain more wealth than they would have achieved if they had invested in the public markets. We calculate PME as the ratio of discounted cash inflow to discounted cash outflow, where the discount rate is the total return in the correspond- ing stock market. For investments in US portfolio companies, we use the S&P 500 index to act as a proxy for the public market, as in Kaplan and Schoar (2005). For investments outside the US, we use the corresponding local stock market index. Sorensen and Jagannathan (2015) present rigorous economic underpinnings for PME and show that PME is equivalent to measuring performance using Rubinstein’s (1976) dynamic version of the capital asset pricing model (CAPM). The authors show that under reasonable assumptions about investor util- ity, PME is robust and valid regardless of the beta of an investment, even when the beta is time varying.

Cross-border venture-capital investments Axel Buchner et al

581

Journal of International Business Studies

They conclude that with ‘PME, investors can eval- uate risk-adjusted performance without explicitly calculating any betas or even knowing the risk of the underlying investments’ (Sorensen & Jagan- nathan, 2015: 44). Hence we interpret PME as a risk-adjusted measure of performance.

In our initial multivariate analysis, we regress VC performance on whether or not a portfolio com- pany is domestic or cross-border. Subsequent anal- yses relate VC performance to measures of distance between a VC firm and a portfolio company. VC performance (the dependent variable) is measured as either IRR or PME. In order to account for the endogeneity of the cross-border indicator arising from an (un)observable difference in VC backers’ selection criteria and investment behaviour at home and abroad, we estimate a two-stage Heck- man model. At the first stage, we estimate a probit model of the probability of an investment being cross-border with the cross-border indicator (coded one for a cross-border investment and zero other- wise) as the dependent variable. The instrument used at the first stage is the capital inflow into the VC industry of the VC provider’s home country in the year of the investment.9 At the second stage, we estimate VC performance by including the inverse Mills ratio based on the estimates of the first stage among the control (explanatory) variables. The explanatory variable of interest is the cross-border indicator. As control variables, we include deal and VC characteristics such as VC experience, investment size, fund age, and indicators of syndication, the financing stage, the industry of the portfolio com- pany, and the year of investment (the deal year).10

We also control for country-specific stock market liquidity (based on the portfolio company’s coun- try of origin) in the year prior to VC exit. Further, we use bootstrapped standard errors. In addition to the two-stage model, we estimate mixed-effects models based on Hesketh, Skrondal, and Pickles (2005) which control for observable and unobserv- able heterogeneity, including differences between domestic and cross-border investments, and for the impact of outliers.

In our initial analysis, we use a binary indicator which captures whether a portfolio company is domestic or cross-border. In a subsequent analysis, we examine an additional binary indicator (cross- border not sharing border) to differentiate between neighbouring countries and foreign countries with- out shared borders. In the final part of our analysis, we investigate whether distance between a VC firm and a portfolio company affects performance. We

use three different measures of distance relating to geographic distances, cultural disparities, and insti- tutional differences between countries of VC firms and portfolio companies. Geographic distances between VC firms and portfolio companies are measured as the physical distances between the capitals of the respective home countries. As in Dai and Nahata (2016), we quantify cultural disparities between the countries of VC firms and portfolio companies using the four cultural dimensions of Hofstede, Hofstede, and Minkov (2010), who follow the approach of Kogut and Singh (1988).11 The four dimensions relate to power distance, individualism, masculinity, and uncertainty avoidance. Hofstede et al. (2010) framework is the most widely used and recognized framework for measuring cultural dis- parities in different disciplines, including interna- tional business and management research (Sivakumar & Nakata, 2001; Kirkman, Lowe, & Gibson, 2006). We obtain data from Geert Hofst- ede’s website (www.geerthofstede.nl) and use the Cartesian distance measure to calculate culture disparity (see Appendix Table 9 for details of this measure). We use three measures of institutional differences between the home countries of VC firms and portfolio companies: differences in the regu- latory quality, political stability, and the legal sys- tems. Appendix Table 9 provides details of data sources and definitions of the variables. All vari- ables are from CEPRES except for those representing market liquidity, differences in regulatory quality, and differences in political stability, which are collected from the World Bank online database, and the variable representing differences in legal systems, which is based on data collected from Rafael La Porta’s website. In order to examine the effect of explanatory

variables on the time from a VC investment to the VC exit or, more accurately, on the exit hazard rate defined as the inverse of the time to exit, we estimate the Cox proportional hazard model. The hazard function measures the likelihood of a VC firm to exit its investment within a small time interval, conditional on VC and market character- istics. The interesting feature of the Cox propor- tional hazard model is that it does not require any distributional assumptions about the exit rate. The coefficients of the Cox proportional hazard model are estimated through maximum likelihood esti- mation. A positive coefficient suggests that a unit increase in the covariate accelerates the exit, while a negative coefficient decelerates the exit. Specifi- cally, we estimate a frailty Cox model which is

Cross-border venture-capital investments Axel Buchner et al

582

Journal of International Business Studies

http://www.geerthofstede.nl

similar to a fixed-effects model in a linear regres- sion. Our model controls for ‘fixed effects’ in terms of heterogeneity across VC firms. The model esti- mates an additional parameter theta, which indi- cates the presence of such heterogeneity.

RESULTS

Univariate Analysis Devigne et al. (2016) highlight the need for further research on the variation of VC deal-level returns across different exit routes. Panel A of Table 1 reports the annual rates of return earned by VC firms from fully exited investments as measured by the IRR. The figures are presented by regions (North America, the UK, Continental Europe, and the ROW) and exit routes (IPOs and M&A). In almost all regions and for all exit routes, cross-border investments generate lower IRR than domestic

deals (except for investments exited through IPOs by VC firms in the ROW).12

Next, we examine the PME, which we interpret as a measure of relative and risk-adjusted performance as outlined in ‘‘Data and Methodology’’ sec- tion. Panel B of Table 1 shows that the mean PMEs for domestic investments range from 2.02 to 2.86 depending on region. PMEs above 2 suggest that the wealth generated by VC investments in domes- tic portfolio companies is more than twice the wealth generated by investments in public markets. In contrast, the mean PMEs for cross-border invest- ments range from just 1.4 to 1.7. We find statisti- cally significantly higher PMEs for domestic investments in almost all regions except for the ROW. In North America, the mean domestic PME is twice that of cross-border investments. The differ- ence between domestic and cross-border PMEs is particularly pronounced for investments which

Table 1 Returns and PMEs at exit

North America Europe (Ex. UK) UK ROW

Dom Cross-border Dom Cross-border Dom Cross-border Dom Cross-border

Panel A: IRR

Full sample

Mean 46.69 28.73** 43.02 24.47*** 34.16 18.90** 27.54 41.69**

Median 24.76 21.45* 17.52 23.68* 19.99 14.99* 27.15 24.32*

No of obs 3945 389 553 286 258 105 794 199

IPO

Mean 125.33 114.62* 112.86 101.91* 117.69 103.76* 85.88 118.09***

Median 42.63 39.32 30.42 40.72* 27.52 18.013* 45.69 41.67

No of obs 861 25 82 41 69 28 95 23

M&A

Mean 77.58 49.99** 78.11 40.73** 54.47 37.94** 98.87 66.83**

Median 11.30 12.30 12.18 11.27 13.79 10.57 12.19 11.47

No of obs 1575 195 342 139 136 58 542 118

Panel B: PME

Full sample

Mean 2.86 1.40** 2.02 1.49* 2.04 1.74* 1.30 5.10**

Median 1.20 0.89* 1.05 1.03 1.40 1.12 1.02 2.05*

No of obs 3945 389 553 286 258 105 794 199

IPO

Mean 7.96 2.79*** 2.52 1.86** 2.58 1.95* 2.05 6.66

Median 3.92 1.12** 1.54 1.00 1.06 0.88 1.98 1.08*

No of obs 861 25 82 41 69 28 95 23

M&A

Mean 4.12 2.19** 2.20 1.71* 2.21 1.48* 1.92 4.48***

Median 2.45 0.98* 1.47 1.06 1.33 1.11 1.06 2.08**

No of obs 1575 195 342 139 136 58 542 118

This table shows the mean and median IRR and PMEs by exit routes and regions of VC firms. We show the results for the full sample and for IPOs and M&A exits separately. Panel A shows IRR and panel B shows PMEs. We use statistical tests to assess whether the performance of cross-border investments differs significantly from domestic deals. The t test for each mean is based on an unequal sample and unequal variance. The test for differences in medians is the Wilcoxon test.

***, **, and * indicate significance at the 1, 5, and 10%, levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

583

Journal of International Business Studies

were exited through IPOs, with the domestic PMEs in North America being almost three times the cross-border PMEs. We find a similar pattern for the median PMEs of North American VC firms. With such firms, median PMEs are significantly higher for domestic investments than cross-border invest- ments. However, differences in medians are not statistically significant for the other regions. The magnitude of our overall median PME for domestic North American investments is comparable to that reported by Harris, Jenkinson, and Kaplan (2014).

We examine whether VC investors select deals abroad which are systematically different from their domestic deals and whether certain types of VC backers self-select cross-border investments. Our univariate analysis in Table 2 examines the characteristics of cross-border and domestic invest- ments. Table 2 presents the figures by regions. With regard to VC firms from all four regions, we find that those firms engaging in cross-border invest- ments are on average significantly older (more experienced) than those involved in domestic investments. The average (mean) difference in age between firms backing cross-border portfolio com- panies compared with those investing in domestic companies is broadly similar at approximately 2.5–2.7 years for VC firms in all regions except North America.13 Among North American VC firms, the age difference is much more pronounced and is between seven and nine years (based on medians and means, respectively). This clearly shows that cross-border investments are under- taken by seasoned VC firms, a finding which is consistent with those for the US reported by Cumming and Dai (2010).

We find a similar but weaker effect in terms of the age of VC funds (rather than VC firms). Across all regions, it appears that cross-border deals are carried out by older funds. However, this difference between domestic and cross-border deals is margin- ally significant (at 10%) based on the means of fund age; based on medians, the difference is insignifi- cant. Perhaps the fund age difference reflects a tendency of funds to invest first in domestic deals perceived to be less risky and to delay ‘gambling’ on potentially more risky cross-border investments until the funds have matured. Thus once earlier domestic deals show signs of success, VC firms are safe in the knowledge that they can offset the potential risk of cross-border deals against their existing domestic successes.

On average, cross-border deals are larger in size than domestic deals except for those undertaken by

Continental European VC firms, whose cross-bor- der deals are smaller than domestic deals. This may be because European VC firms invest in cross- border regions with relatively underdeveloped institutions and capital markets. In each of the four regions, cross-border invest-

ments are exited more quickly than domestic investments. This finding is consistent with the higher costs of screening, monitoring, and foster- ing cross-border investments which cause VC back- ers to exit cross-border investments more quickly than domestic investments. This result may also be driven by the propensity of VC firms to invest in cross-border ventures only later in the life of VC funds, leaving less time to realize investments before funds are wound up. Liquidity is defined as the level of stock market

activity which VC firms face in a portfolio com- pany’s country of origin. Table 2 shows average liquidity across the four regions. Liquidity is higher for North American VC firms investing in domestic companies than the liquidity these North American VC firms face when investing abroad. These results are clearly unsurprising given the highly developed North American capital markets. Focusing on VC firms in other regions, we find that European and UK VC investors face more liquidity in their domestic markets than in their cross-border desti- nations. In contrast, VC firms in the ROW seem to come from countries with markets which are on average less liquid than the markets in their cross- border destinations. In sum, most VC firms find lower liquidity in their cross-border destinations than in their domestic markets. In terms of syndication, our results show signif-

icant differences between domestic and cross-bor- der VC deals, with cross-border deals being more likely to be syndicated in all regions except in the ROW. In particular, North American and UK VC firms syndicate cross-border deals more frequently than domestic deals. UK VC firms are more than twice as likely to syndicate cross-border deals than domestic deals, and among North American VC firms, the frequency of syndicating cross-border deals is 54% higher than for domestic deals. We observe the opposite among VC firms in the ROW: these firms more frequently syndicate domestic deals, with approximately 70% of their domestic deals being syndicated compared with just 40% of their cross-border deals. Among Continental Euro- pean VC firms, the proportion of syndicated deals is very similar for all (domestic and cross-border) deals. Because a high proportion of syndication

Cross-border venture-capital investments Axel Buchner et al

584

Journal of International Business Studies

may reflect VC investors’ demand for risk (or loss) sharing (e.g. Lerner 1994), our results may suggest that North American and UK VC firms perceive

cross-border deals to be risky. In contrast, it is domestic deals which are seen to be riskier by VC firms in the ROW.

Table 2 Descriptive statistics of deal characteristics

Variables North America Europe (Ex. UK) UK ROW

Dom Cross-border Dom Cross-border Dom Cross-border Dom Cross-border

Panel A

VC age (years)

Mean 4.64 13.65** 8.36 11.12* 10.56 13.05* 5.17 7.66*

Median 4.00 11.00** 7.50 10.00* 8.92 10.75* 4.54 5.00

Fund age (years)

Mean 2.35 3.94* 1.72 2.09* 1.88 2.89* 2.31 3.61*

Median 2.08 2.25 1.15 1.13 1.25 1.10 1.50 2.13

Investment size (USD million)

Mean 4.69 6.28** 4.39 2.16** 3.01 4.93* 5.15 7.13*

Median 1.15 2.15** 2.02 1.51** 1.86 2.00* 2.30 2.54

Investment duration (years)

Mean (years) 4.74 3.44** 3.88 3.20* 3.98 3.15* 3.90 3.10**

Median (years) 4.00 3.14* 3.32 3.11* 3.50 3.01* 3.42 3.08*

Liquidity

Mean 1.46 0.91*** 1.10 0.84** 1.29 1.01* 0.56 1.20**

Median 1.24 0.75*** 1.01 0.81* 1.15 0.86* 0.55 0.94**

Syndication (binary)

Yes 36.70 56.98** 48.56 49.43 32.97 72.97** 68.93 40.07**

No 63.30 43.09** 51.44 50.57 67.03 27.03** 31.07 59.94**

Financing stages (binary)

Early 72.26 67.23*** 59.47 67.84 53.06 66.06 67.56 61.52

Expansion 23.81 19.79*** 38.85 18.92*** 37.51 18.04** 19.61 21.73

Later 3.94 12.98*** 1.68 13.23*** 9.44 15.90* 12.83 16.75

Industry (binary)

Biotechnology 2.58 7.26** 6.89 12.95** 6.71 9.68** 7.191 2.91***

Consumer Goods and Services 16.77 11.33* 11.20 15.59 10.95 13.32 11.403 20.00***

Financials 2.38 2.26 2.22 4.54** 2.02 2.61 2.29 3.20

Industrials 16.42 19.53* 58.20 51.30 58.76 53.05 1.50 21.43

Information Technology 60.34 57.11 19.12 15.10 18.17 20.04 59.13 52.38*

Others 1.51 2.51* 2.37 0.52* 3.39 1.29* 0.48 0.09**

No of deals 3945 389 553 286 258 105 794 199

Panel B: logarithm transformations

VC age (years)

Mean 1.63 2.47 2.44 2.37 2.33 2.63 1.81 2.19

Median 1.55 2.32 2.38 2.27 2.26 2.47 1.66 1.69

Fund age (years)

Mean 0.94 1.49 0.75 0.81 0.82 1.09 0.87 1.39

Median 0.92 0.88 0.21 0.14 0.38 0.11 0.55 0.81

Investment size (USD million)

Mean 1.77 1.87 1.05 1.01 1.22 1.74 1.79 2.04

Median 0.38 0.84 1.01 0.74 0.78 0.88 0.88 1.02

This table provides descriptive statistics for domestic and cross-border deals by regions. VC age is the age (years in business) of a VC firm at the time of an initial investment in a portfolio company. Fund age is the age of a fund measured from the date of the fund’s initiation to the date of an investment. Investment size is the total amount invested by a VC firm in a portfolio company. Investment duration is the time between an initial VC investment and the VC exit date (in years). Liquidity is the stock market liquidity of a portfolio company’s country measured as the total value of shares traded on the stock exchange(s) divided by the country’s gross domestic product (GDP). Syndication is a binary variable equal to one for syndicated deals and zero otherwise. We show figures for each of three financing stages (early-, expansion-, and later-stage investments). Panel B shows natural logarithmic transformations of VC age, fund age, and investment size. We use a statistical test to assess whether cross-border characteristics differ significantly from domestic characteristics. The t-tests for the means are based on unequal samples and unequal variances. Differences in medians are tested using the Wilcoxon test.

***, **, and * indicate significance at the 1, 5, and 10% levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

585

Journal of International Business Studies

In terms of the breakdown of financing stages, we find a consistent pattern among all cross- border deals, with approximately two-thirds of investments at the early stage, one-fifth at the expansion stage, and the rest (approximately 13–16%) at the later stage. This pattern is similar to the breakdown across financing stages of domestic deals by VC firms in the ROW. This finding may suggest a degree of convergence in terms of a global investment pattern. The break- down in the three other regions is broadly similar, although there are some differences. Notably, North American VC firms have a greater tendency to invest at early stages domestically rather than abroad. This approach may reflect the aversion of such VC firms to the higher risk of cross-border early-stage investments compared with domestic early-stage deals. European and UK VC firms differ from the global pattern in terms of their domestic deals, with a greater preference for expansion- stage investments in domestic deals compared with other stages.

In terms of industry sector, there is evidence of investment clustering. North American and ROW VC firms are more likely to invest in the IT sector, while European and UK VC firms are more likely to invest in industrials. In contrast to these regional variations, there is little difference between domes- tic and cross-border deals in each region (except for the biotechnology industry).14

In conclusion, compared with domestic invest- ments, cross-border deals are conducted by older VC firms and later in the life of a VC fund. Cross- border deals are larger (based on investment size) and more likely to be syndicated at later financing stages and in the biotechnology sector. In terms of performance, we find that cross-border investments are exited more quickly but at the expense of returns and with lower IPO frequency. The lower IRR and the shorter holding periods may both be due to systematic differences between domestic and cross-border investments. These differences could be because VC firms target mature companies and later-stage financing when crossing borders. Hence the risk/return profile of these investments may differ from domestic VC investments. Using PME as a risk-adjusted measure of performance, we never- theless find that domestic investments outperform cross-border investments in almost all regions. In the next section, we examine the performance of domestic and cross-border investments in the con- text of a multivariate analysis.

Multivariate Analysis In order to conduct a formal test of Hypothesis 1 (which posits that cross-border and domestic VC investments generate the same returns), we esti- mate two-stage Heckman models. The first stage (reported in panel A of Table 3) estimates the probability of a given deal being a cross-border investment; the second stage regresses deal-level performance on the cross-border indicator and a range of control variables.15 Performance is mea- sured as IRR in panel B and PME in panel C. We control for risk by including indicators for the financing stage. Studies suggest financing stage as a suitable proxy for deal risk (Sahlman, 1990; Ruhnka & Young, 1991; Seppa & Laamanen, 2001; Cornelli, Kominek, & Ljungqvist, 2013). The results are shown in Table 3.16 Panel A of the

table shows the results for IRR.We find a statistically significant negative coefficient of the cross-border indicator in each of the four regressions (models I– IV). This finding is consistent with our univariate analysis reported in Table 1. Our multivariate anal- ysis confirms this cross-border effect in terms of lower IRR, even after controlling for risk and other potential determinants of performance. The magni- tude of this cross-border coefficient ranges from –0.27 for the UK to –0.156 (for the ROW). This result suggests that cross-border investments have signif- icantly lower IRR on average, statistically and eco- nomically, than comparable domestic investments for VC firms based in any of the four regions. The cross-border coefficient in North America (model 1) is –0.192, indicating that, all else being equal, cross- border investments by North American VC firms underperform equivalent domestic investments by 19% in terms of IRR.With regard toVCfirms inother regions, cross-border underperformance ranges from 16% for VC investors in the ROW to 27% for the UK. In panel B of Table 3, the dependent variable is

the risk-adjusted measure of performance (PME). Consistent with the IRR results in panel A and the univariate analysis (Table 1), we find that the PME for cross-border investments is lower than the PME for domestic investments. This result confirms that domestic investments outperform cross-border investments and that this performance difference persists after controlling for risk and other known performance determinants. Next, we examine whether this cross-border

effect is a long-term feature of the VC industry or whether it is concentrated in the early years of the internationalization of VC investments. The

Cross-border venture-capital investments Axel Buchner et al

586

Journal of International Business Studies

Table 3 Multivariate analysis (two-stage model)

Variables Model I: North America Model II: Europe (Ex. UK) Model III: UK Model IV: ROW

Coeff P value Coeff P value Coeff P value Coeff P value

Panel A Ln capital inflow 0.013*** (0.000) 0.027** (0.016) 0.012** (0.036) 0.003 (0.115) Ln VC age 0.002** (0.039) 0.001** (0.002) 0.011*** (0.000) 0.004 (0.234) Ln fund age 0.014* (0.091) 0.097* (0.074) 0.104** (0.017) 0.072* (0.092) Ln investment size 0.041** (0.041) 0.068*** (0.005) 0.020** (0.037) 0.020** (0.041) Liquidity 0.023*** (0.000) 0.002* (0.096) 0.027** (0.022) 0.012* (0.064) Syndication 0.016** (0.029) 0.003* (0.087) 0.009* (0.052) 0.006 (0.184) Industrya Present Present Present Present Financing stagea Present Present Present Present Investment yeara Present Present Present Present Pseudo R square 0.166 0.131 0.089 0.111 No. of obs 4334 839 363 993

Panel B Cross-border -0.192** (0.033) -0.232** (0.017) -0.270** (0.023) -0.156** (0.017) Post-1997 9 cross-border 0.047 (0.130) 0.173 (0.207) 0.036 (0.193) 0.250 (0.135) Post-1997 -0.339** (0.021) -0.379** (0.032) -0.117 (0.108) -0.252** (0.037) Ln VC age 0.011** (0.023) 0.015** (0.030) 0.018** (0.012) 0.050** (0.015) Ln fund age -0.141 (0.309) -0.126 (0.192) -0.233 (0.226) 0.264 (0.781) Ln investment size -0.238*** (0.000) -0.263*** (0.000) -0.320*** (0.000) -0.373*** (0.000) Investment duration 0.065** (0.030) 0.050** (0.045) 0.091** (0.030) 0.055* (0.088) Ln fund age top 25th 0.066 (0.398) 0.057 (0.223) -0.059 (0.762) -0.276 (0.652) Liquidity -0.188** (0.024) -0.273** (0.011) -0.178** (0.026) -0.422** (0.050) Syndication 0.014 (0.743) 0.012 (0.222) 0.104 (0.458) 0.353* (0.086) Constant 0.173*** (0.000) 0.037** (0.044) 0.152** (0.027) 0.154** (0.022) Lambda 0.605* (0.090) -0.477* (0.096) 0.201 (0.244) -0.117 (0.314) Industrya Present Present Present Present Financing stagea Present Present Present Present Investment yeara Present Present Present Present Adj R square 0.141 0.077 0.044 0.066 No. of obs 4334 839 363 993

Panel C Cross-border -0.018** (0.020) -0.020** (0.016) -0.023** (0.042) -0.016** (0.035) Post-1997 9 cross-border 0.002 (0.477) 0.004 (0.356) 0.007 (0.381) 0.041 (0.270) Post-1997 -0.019* (0.078) -0.024* (0.053) -0.021* (0.092) -0.061* (0.083) Ln VC age 0.001** (0.032) 0.007** (0.040) 0.001** (0.012) 0.002** (0.017) Ln fund Age -0.006 (0.306) -0.011 (0.144) -0.004 (0.766) 0.022 (0.254)

Ln investment size -0.015*** (0.000) -0.010*** (0.000) -0.009*** (0.000) -0.058*** (0.000) Investment duration 0.003** (0.021) 0.004** (0.024) 0.002** (0.029) 0.006* (0.056) Ln fund age top 25th 0.023 (0.391) 0.008 (0.499) 0.009 (0.756) -0.036* (0.065) Liquidity -0.034** (0.040) -0.032** (0.033) -0.017** (0.023) -0.029* (0.075) Syndication 0.003 (0.219) 0.004 (0.216) 0.005 (0.663) 0.036* (0.089) Constant 0.014*** (0.000) 0.018** (0.031) 0.016** (0.023) 0.014** (0.000) Lambda 0.021 (0.133) -0.034* (0.081) 0.010 (0.255) 0.014 (0.261) Industry Present Present Present Present Financing stage Present Present Present Present Investment year Present Present Present Present Adj R square 0.13 0.084 0.050 0.074 No. of obs 4334 839 363 993

This table presents the results of a two-stage Heckman model of VC performance. Stage 1 in panel A shows the probability of investing in cross-border deals using probit models. The dependent variable is equal to one if the investment is a cross-border deal and zero otherwise. At stage II, the dependent variable is IRR measured as annualized IRR in panel B. The dependent variable is PME in panel C. Capital inflow is the logarithm of the total amount of funds in the VC industry. Cross-border is one for cross-border investments and zero otherwise. We include a binary indicator, post-1997, which is equal to zero for investments up to 1997 and one thereafter. Cross-border x post-1997 is cross-border interacting with post-1997. Ln VC age is the natural logarithm of age (years in business) of a VC firm at the time of an initial investment in a portfolio company. Ln fund age is the natural logarithm of fund age measured from the date of a fund’s initiation to the date of an investment. Ln investment size is the natural logarithm of the total amount invested by a VC firm in a portfolio company. Investment duration is the number of years from an initial VC investment to VC exit. Ln fund age top 25th is the natural logarithm of fund age for funds above the 75th age percentile and zero otherwise. Liquidity is the stock market liquidity of a portfolio company’s country measured as the total value of shares traded on the stock exchange(s) divided by the country’s GDP. Syndication is a binary variable equal to one for syndicated deals and zero otherwise. a We include two binary financing-stage indicators for expansion- and later-stage investments (with early stage as the base). We also include dummies for a portfolio company’s industry and (investment) year.

***, **, and * indicate statistical significance at the 1, 5, and 10% levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

587

Journal of International Business Studies

literature finds that the period prior to the late 1990s is characterized by high returns on VC investments in North America, while the subse- quent period experienced significantly lower VC fund returns and large capital inflow resulting in the saturation of the VC industry (Harris et al., 2014). We examine whether this performance decline occurred equally among cross-border and domestic investments. A priori, we may expect that high capital inflow into a saturated industry leads to a decline in the performance of domestic and cross-border investments. However, some of this decline may be offset by positive learning effects derived from cross-border investments. Thus we may expect less of a performance decline among cross-border investments initiated in the late 1990s. Harris et al. (2014) show that the decline in VC performance started in the late 1990s. The litera- ture (e.g. Aizenman and Kendall 2012) also suggests that cross-border VC investing accelerated during the 1990s because of shortages in profitable domes- tic investment opportunities. We use 1997 as the cut-off point to divide our sample. Further, we include a binary indicator (post-1997) coded as zero for investments made up to 1997 and as one thereafter. We include post-1997 in our multivariate analyses both on its own (un-interacting) and interacting with the cross-border indicator.17

Table 3 reports a significant negative coefficient for the un-interacted post-1997 variable, suggesting that performance declined for all investments, namely for domestic and cross-border deals, after the late 1990s. The coefficients of the interaction terms cross-border and post-1997 are positive and in some cases as large (in absolute terms) as the cross- border coefficient. However, because these coeffi- cients do not differ from zero at conventional levels of statistical significance, it appears that the per- formance differential between cross-border and domestic investments remains unchanged post- 1997. This result leads us to conclude that the cross-border effect is a permanent feature.

We consider several further moderating factors.18

We report our results in Table 3 separately for four broad regions. This regional breakdown is based on the origin of the VC investors. We observe under- performance of cross-border investments in all four regions despite the differences between these regions in terms of the institutional and cultural environments in which VC firms operate. We further explore whether our results depend on the destination (as opposed to the origin) of cross- border deals. To this end, we separate the cross-

border indicator used in our multivariate analysis reported in Table 3 into three binary variables depending on deal destination. Focusing on the investments by North American VC firms, we find that cross-border deals in the ROW underperform domestic deals the most, by more than 10% (the coefficient of cross-border deals in the ROW is – 0.11). Investments in the UK underperform the least (2.5%), with the underperformance of deals in Europe at 6%. While the magnitude of underper- formance appears to vary by destination, we con- clude that cross-border investments consistently underperform domestic investments worldwide. Next, we examine whether our results differ for

subsets of VC firms, specifically whether more experienced VC investors show less cross-border underperformance. Measuring experience as VC Age, which represents the number of years a VC firm has been in business, we extend the models reported in Table 3 by including as an additional variable the cross-border indicator interacting with a binary variable coded one for the most experienced quartile of VC firms (and zero otherwise). We find the coefficient of this interaction term statistically not different from zero in all regions except ROW (where it is positive but statistically significant at only 10%). We conclude that there is no evidence of a differential effect of cross-border investing for top-tier VC investors. All of our multivariate analyses control for a

range of statistically significant determinants of performance. Specifically, the control variables include VC age, investment size, investment duration, fund age, liquidity, syndication, financing stage, year, and industry. Except for fund age and syndication, we find that all of these control variables have a statistically significant impact on VC firms’ perfor- mance. VC age increases IRR and PME in all regions, suggesting that more experienced VC backers are better at selecting, nurturing, and exiting invest- ments. Large investments are associated with lower IRR and PME in all regions. This suggests that VC providers require higher returns for smaller invest- ments to compensate them for higher (business) risk. We find that investment duration (i.e. the holding period, or time to exit, of VC backers) has a significantly positive impact on IRR and PME in all regions. It appears that VC backers require higher returns to compensate them for longer holding periods. Investments which are held for a longer period are likely to be early-stage invest- ments for which VC backers require higher average returns to compensate for (liquidity, etc.) risk.

Cross-border venture-capital investments Axel Buchner et al

588

Journal of International Business Studies

Another possible reason for this positive impact of Investment Duration on IRR could be the tendency of VC investors to hold promising (and ultimately profitable) investments longer in order to maximize the future gains from a successful exit.

Next, we examine the impact of market liquidity on IRR and PME. Cumming, Fleming, and Sch- wienbacher (2005) focus on variations in the liquidity of exit markets in terms of ‘liquidity risk’. They find that when the liquidity of exit markets is high, VC firms tend to invest more in later-stage rather than early-stage ventures. Cochrane (2005) documents that early-stage returns are higher than later-stage returns. Based on the foregoing studies, one may expect that the returns to VC firms are likely to be low when market liquidity is high. This reasoning may suggest that liquidity has a negative coefficient in our models. However, because we control for the financing stage, we should not expect to find an incremental negative impact of market liquidity on IRR and PME.19 Nevertheless, we find statistically significant negative coefficients of liquidity in all of the models. Our results suggest that over and above the impact of liquidity on VC investment decisions (specifically the choice of financing stage), more liquid markets motivate VC firms to ‘rush to exit’ at the expense of lower returns.

The coefficient of the inverse Mills ratio (lambda) is statistically significant at the 10% level in only some of the regions, specifically in the models for North America and Europe in panel A, and only in Europe in panel B. This finding suggests that adjusting for selection bias and endogeneity is only important in some of the regions, thereby confirm- ing that the cross-border effect which we document is not the result of differential deal selection by VC backers. Instead, our results show that cross-border deals underperform equivalent domestic deals. In ‘‘Robustness’’ section, we discuss alternative mod- elling approaches, including mixed-effects models and propensity score matching, and conclude that our finding of cross-border underperformance is robust.20

In conclusion, our results show that cross- border investments have significantly lower returns (in terms of IRR and PME) than domestic investments after controlling for risk, which has financing stage as a proxy, and a range of other potential determinants of performance. On the basis of our results, we do not find support for Hypothesis 1.

Alternative Measures of Performance We test differences between cross-border and domestic investments in terms of exit routes and time to exit. In the absence of detailed cash flow data, prior studies focus on exit success as a proxy of VC performance measurement. Our analysis examines the robustness of this approach in the context of examining cross-border investments. Some exit routes benefit VC investors more than

others. IPO exits and exits through M&A, such as trade sales or secondary buyouts, generate positive financial returns (IRR, as reported by Cumming, 2008), while write-offs (liquidations) typically result in the loss of the VC investments (Cumming, 2008; Dai et al., 2012). ‘Successful’ exits also enhance venture capitalists’ reputations and pro- vide opportunities to raise additional funds from limited partners (LPs). Our first measure of exit performance is the so-called ‘success ratio’, which reflects the frequency of ‘successful’ exits from portfolio companies. A narrow measure of the success ratio considers only IPO exits as successes and is defined as the ratio of IPO exits to all exits. A broader version of the success ratio which considers both IPO and M&A exits as successes is defined as the ratio of IPO and M&A exits to all exits. Prior studies have examined the success ratio as an indicator of VC performance (e.g. Hochberg et al., 2007; Chen et al., 2010; Cumming & Dai, 2010; Wang & Wang, 2012; Nahata et al., 2014; Dai & Nahata, 2016). Our second measure of performance is the length

of VC backers’ holding periods in terms of the time from the first VC investment in a given portfolio company to the exit of the VC backer. Prior evidence suggests that VC backers realize their investments, on average, after 7 years, with the time to exit ranging from 6.5 years to more than 8 years, depending on the exit route (Giot & Schwienbacher, 2007). Panel A of Table 4 shows that the M&A route is

the most common type of successful exit, followed by IPOs. Focusing on IPO exits, we find no signif- icant difference between cross-border and domestic investments in the proportion of IPO exits in all regions, except North America. With regard to North American VC firms, we find that IPO exits are less likely for cross-border investments, with only 6% of cross-border deals resulting in an IPO exit. On the basis of this success ratio, cross-border investments are less successful than domestic deals for North American VC firms. However, perhaps

Cross-border venture-capital investments Axel Buchner et al

589

Journal of International Business Studies

IPO frequency is not an appropriate indicator of success, given that an M&A exit can also be attractive in terms of returns. Defining the success ratio more broadly as the combined proportion of IPO and M&A exits relative to all exits, we find no significant difference between domestic and cross- border deals.

Panel B of Table 4 reports the investment-hold- ing period (the time to exit) by region and exit route. The results of the full sample show that irrespective of the location of the VC firms, cross- border investments are exited more quickly than domestic investments. This finding is consistent with the higher costs of screening, monitoring, and fostering cross-border investments, tipping the

balance towards quicker exits from cross-border than domestic investments. This result may also be driven by the propensity of VC firms to invest in cross-border ventures only later in the lives of their VC funds, leaving less time to realize investments before the funds are wound up (see Table 2). Examining the results separately by region and exit route, we find the same type of cross-border effect (a shorter cross-border holding period) for all of the investments by North American VC firms irrespec- tive of exit route. We also find the same effect for all IPO exits irrespective of region. However, outside North America, VC firms have longer holding periods in cross-border investments which are exited through M&A than for their corresponding

Table 4 Investment types and holding periods (time to exit)

North America Europe (Ex. UK) UK ROW

Dom Cross-border Dom Cross-border Dom Cross-border Dom Cross-border

Panel A

Full sample

Proportions 0.910 0.090*** 0.659 0.341** 0.711 0.289** 0.800 0.200***

No. of obs 3945 389 553 286 258 105 794 199

IPO

Proportions 0.218 0.064** 0.148 0.143 0.267 0.266 0.120 0.116

No. of obs 861 25 82 41 69 28 95 23

M&A

Proportions 0.399 0.501 0.618 0.486* 0.527 0.552 0.683 0.593

No. of obs 1575 195 342 139 136 58 542 118

Write-offs

Proportions 0.383 0.434 0.233 0.371 0.205 0.181 0.198 0.291

No. of obs 1509 169 129 106 53 19 157 58

Panel B

Full sample

Mean (years) 4.74 3.44** 3.88 3.20* 3.98 3.15* 3.90 3.10**

Median (years) 4.00 3.14* 3.32 3.11* 3.50 3.01* 3.42 3.08*

No. of obs 3945 389 553 286 258 105 794 199

IPO

Mean (years) 6.98 4.69** 6.08 4.68** 5.96 4.12** 4.73 3.12**

Median (years) 6.88 4.17** 4.50 4.02* 4.51 4.00* 4.25 3.01**

No. of obs 861 25 82 41 69 28 95 23

M&A

Mean (years) 4.49 3.10* 4.37 4.83* 3.39 4.03* 3.43 4.82**

Median (years) 3.97 3.09** 3.99 3.09* 3.01 3.83* 3.17 4.35**

No. of obs 1575 195 342 139 136 58 542 118

Write-offs

Mean (years) 2.13 1.90 2.02 2.02 2.85 2.45 1.83 2.11

Median (years) 1.87 1.71 1.62 1.01 2.09 1.87 1.75 1.57

No. of obs 1509 169 129 106 53 19 157 58

Panel A shows the exit numbers by exit type and region. Full-sample proportions are relative to total (domestic and cross-border) exits by region. Proportions for each of the exit routes (IPO, M&A, and write-offs) are relative to either domestic or cross-border exits by region. The Z-test for the proportions is based on an unequal sample and unequal variance. Panel B shows the means and medians of investment durations (in years) by the method of exit and the region of VC firms. The t-test for each mean is based on an unequal sample and unequal variance. The t-test for each median is based on the Wilcoxon test.

***, **, and * indicate significance at the 1, 5, and 10% levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

590

Journal of International Business Studies

domestic investments. Judging the performance of VC investments in terms of time to exit, it appears that cross-border investments are more successful than domestic investments, at least for North American VC firms.

Given the conflicting evidence on the exit success of cross-border investments by North Amer- ican VC firms in panels A and B, we suspect that exit behaviour is driven primarily by country- specific and other macro factors and does not reflect the selection and value-adding behaviour of North American VC backers.

In order to examine whether cross-border deals are exited more quickly even after controlling for macro factors and VC and deal characteristics, we estimate a multivariate Cox hazard model. Table 5 shows the results of this model with the holding period (time to exit) as the dependent variable and an indicator for cross-border deals, controlling for VC age, fund age, investment size, market liquidity (liquidity), syndication, financing stage, investment year, and industry of the portfolio company. We find that, all else being equal, VC firms exit cross-border investments significantly more quickly than

domestic investments. This effect is observed in all four regions and is strongest in the UK and weakest in North America. Next, we examine whether the exit behaviour of

VC firms and the cross-border effect on exits differ before and after the late 1990s. Speedier exits in the latter period may be due to a learning effect whereby VC firms become better over time at achieving quicker exits. Since VC cross-border investing is a more recent activity than domestic investments, we may expect this learning effect to impact cross-border deals in particular. To this end, we define a time indicator which takes the value of zero for all investments in the years up to 1997 and one thereafter. In the models shown in Table 5, we include the

post-1997 indicator and interact it with the cross- border indicator. The coefficient of post-1997 is positive and significant for Europe and the UK, showing increasing exit rates since the late 1990s in these regions possibly due to a learning effect. The coefficient of post-1997 interacting with cross-border is also positive and significant in all regions, suggesting that the increase in exit speed has been

Table 5 Cox proportional hazard model

Variables Model I: North America Model II: Europe (Ex. UK) Model III: UK Model IV: ROW

Coeff Hazard ratio Coeff Hazard ratio Coeff Hazard ratio Coeff Hazard ratio

Cross-border 1.375*** 3.955 1.689** 5.417 2.519*** 12.411 2.329*** 10.267

Cross-border 9 post-1997 2.324*** 10.221 0.248* 1.281 1.185** 3.269 2.522** 12.459

Post-1997 0.276 1.318 0.915** 2.496 0.773* 2.167 0.377 1.458

Ln VC age -0.018** 0.982 0.007* 1.007 0.002 1.002 0.002 1.002

Ln fund age 0.153** 1.165 0.116* 1.122 0.380** 1.462 0.285* 1.330

Ln investment size 0.221** 1.247 0.132** 1.141 0.127** 1.135 0.294** 1.341

Ln fund age top 25th 0.030 1.030 0.167** 1.182 0.077 1.080 0.040 1.041

Liquidity 0.258*** 1.295 0.853*** 2.346 0.806** 2.238 0.205** 1.228

Syndication 0.482*** 1.619 0.103* 1.108 0.145* 1.156 0.138 1.148

Theta 0.799** 0.208** 0.163** 0.998***

Industrya Present Present Present Present

Financing stagea Present Present Present Present

Investment yeara Present Present Present Present

Pseudo R-square 0.166 0.141 0.090 0.111

No. of obs 4334 839 363 993

Censored obs 512 316 105 255

This table shows a Cox proportional (frailty) model of the VC exit rates estimated using maximum likelihood estimation. Cross-border equals one for cross-border investments and zero otherwise. We include an interaction term of cross-border with post-1997, another interaction term of cross-border with Ln VC age, and a third with Ln investment size. Ln VC age is the natural logarithm of age (years in business) of a VC firm at the time of an initial investment in a portfolio company. Ln fund age is the natural logarithm of fund age measured from the date of a fund’s initiation to the date of an investment. Ln investment size is the natural logarithm of the total amount invested by a VC firm in a portfolio company. Ln fund age top 25th is the natural logarithm of the age of funds for funds above the 75th age percentile and zero otherwise. Liquidity is the stock market liquidity of a portfolio company’s country measured as the total value of shares traded on the stock exchange(s) divided by the country’s GDP. Syndication is a binary variable equal to one for syndicated deals and zero otherwise. Theta is an indicator of heterogeneity (frailty). aWe include two binary financing-stage indicators for expansion- and later-stage investments (with early stage as the base). We also include dummies for a portfolio company’s industry and (investment) year for the years up to 1997.

***, **, and * indicate statistical significance at the 1, 5, and 10% levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

591

Journal of International Business Studies

particularly pronounced for cross-border investments.

Consistent with prior studies (Giot & Schwien- bacher, 2007; Espenlaub et al., 2015), we find that larger deals are exited more quickly. Investment size may increase the VC backers’ marginal costs of continuing with an investment (in terms of mon- itoring and advisory costs) relative to marginal benefits, thereby tipping the balance in favour of an earlier exit.

Fund age at the time of VC investments is also a significant determinant of exit rates. Investments from funds which are more mature at the time of the investments have higher exit rates than invest- ments made from younger funds. In some regions, the effect of fund age is particularly strong for the oldest quartile of funds (as shown by Fund age x top 25th).

Finally, we examine the impact of the liquidity in the stock market of the investment region (i.e. the region of the portfolio company’s country of origin). Based on the findings of Black and Gilson (1998), Cumming et al. (2006), and Wang and Wang (2012), we expect more liquid capital mar- kets to facilitate speedier exits and higher exit rates. Consistent with this conjecture, we find that Liquidity has a strongly significant positive impact on exit rates in all four regions.

As a final control variable, we include in Table 5 an indicator of syndicated investments. Consistent with Giot and Schwienbacher (2007), we find that syndicated investments have higher exit rates than other investments. This suggests that by combining expertise and networks, syndicates are able to facilitate speedier exits.

After controlling for a number of factors, our results show that cross-border investments are exited more quickly than domestic investments. However, if we were to conclude from this that cross-border investments are more successful than domestic investments, this would be at odds with our analysis of returns performance given in Table 3. Our results suggest that VC investors may choose to invest abroad to benefit from shorter holding periods, but they do so at the expense of returns. Our results highlight that alternative mea- sures of performance may result in fundamentally different assessments of success.

The Effect of Distance So far our analysis examines the difference between domestic and cross-border investing using a simple binary indicator. Our results suggest that

foreignness (as captured by our cross-border indica- tor) is a liability. In this section, we try to unpack the aspects of foreignness which drive our results. In our earlier analysis, we treat investments by US VC firms in Canada the same as a US VC firm investing in China. Clearly, the latter process is significantly costlier to an investing US VC firm than the former because of the greater geographical distance involved as well as cultural and regulatory differences, and differences in political risk. In panel B of Table 6, our analysis breaks down

cross-border investing into ventures in neighbour- ing countries (which share a land border) and non- neighbouring countries.21 In addition to cross-bor- der (which continues to be coded as one if a VC firm invests in a foreign country and zero otherwise), we now also include the binary variable cross-border (not sharing) to capture the effect of investments in non-neighbouring countries. Cross-border (not shar- ing) is coded as one if the deal is located in a non- neighbouring foreign country and zero otherwise. Panel B of Table 6 reports the results of a mixed-

effects model of IRR for North America and Europe.22 In both regions, we find the coefficient of cross-border (not sharing) to be significant and negative, with a coefficient which is almost three times as large in absolute terms than that of the cross-border indicator. The latter coefficient is slightly smaller than in our earlier analysis reported in Table 3 (models I and II of panel A) but is still statistically significant and negative. The difference between the cross-border and cross-border (not shar- ing) coefficients is statistically significant, suggest- ing that cross-border investments involving countries which do not share a land border under- perform compared with domestic investments sig- nificantly more than cross-border investments in neighbouring countries. We conclude that there is a significant negative ‘cross–cross-border effect’.23

Next, we examine whether the cross-border effect is driven by geographic distances, cultural dispari- ties, or institutional differences between VC inves- tors and portfolio companies. Studies report mixed evidence on the impact of geographic distances, cultural disparities, and institutional differences between VC firms and portfolio companies on the likelihood of a successful exit, typically an IPO exit (Chen et al., 2010; Cumming & Dai, 2010; Nahata et al., 2014; Chemmanur et al., 2016; Dai & Nahata, 2016). Consistent with past evidence (Li et al., 2014; Nahata et al., 2014; Dai & Nahata, 2016), we predict that geographic distances have no signifi- cant effect after controlling for other distance

Cross-border venture-capital investments Axel Buchner et al

592

Journal of International Business Studies

Table 6 Multivariate analysis of VC performance

Panel A: descriptive statistics Cross-border (all regions)

Mean Median STD

Cross-border (not sharing) 0.793 1.00 0.231 Geographic distance (miles) 4651 4255 2610 Cultural distance (#) 9.87 10.91 5.44 Regulatory quality (diff) 2.00 4.00 37.00 Political Stability (diff) 5.00 1.00 41.00 Legal system (diff) 0.351 0.000 0.403 Logarithmic values

Geographic distance 8.755 8.355 1.184 Cultural distance 2.105 2.187 0.645

Panel B: Mixed-effects models of IRR in (non-)

neighbouring countries

North America Europe

Coeff P value Coeff P value

Cross-border -0.151** (0.024) -0.205** (0.033) Cross-border not sharing border -0.417*** (0.001) -0.588*** (0.000) Ln VC age 0.020** (0.011) 0.024** (0.021) Ln fund age -0.156 (0.254) -0.121 (0.301) Ln investment size -0.305*** (0.000) -0.286*** (0.000) Investment duration 0.082*** (0.000) 0.074*** (0.000) Ln fund age top 75th 0.085 (0.251) 0.099 (0.351) Liquidity -0.267*** (0.000) -0.302*** (0.000) Syndication 0.025 (0.254) 0.021 (0.251) Constant 0.221*** (0.000) 0.042*** (0.000) Pseudo R-square 0.131 0.110 No. of Obs 4334 839 Industry, stage, investment yeara Y Y

Panel C: Mixed-effects models of

IRR with distance measures

Model I: North America Model II: Europe (Ex. UK) Model III: UK Model IV: ROW

Coeff P value Coeff P value Coeff P value Coeff P value

Cross-border dummy -0.178* (0.073) -0.105* (0.092) -0.132* (0.061) -0.101 (0.118) Ln (geographic distance) -0.181* (0.080) -0.120* (0.090) -0.113* (0.058) -0.173* (0.055) Ln (cultural distance) -0.294* (0.084) -0.257* (0.094) -0.276* (0.092) -0.368* (0.082) Regulatory quality (diff) -0.004*** (0.004) -0.005** (0.046) -0.002** (0.021) -0.002* (0.091) Political stability (diff) -0.002* (0.075) -0.003* (0.059) -0.002* (0.072) -0.003* (0.052) Legal system (diff) -0.068 (0.478) -0.103 (0.329) -0.050 (0.601) -0.011 (0.293) Post-1997 -0.518*** (0.000) -0.250** (0.024) -0.483*** (0.000) -0.362*** (0.008) Ln VC age 0.018*** (0.001) 0.030*** (0.000) 0.028*** (0.000) 0.055*** (0.000) Ln fund Age -0.155 (0.254) -0.215 (0.132) -0.196 (0.161) -0.072 (0.688) Ln investment size -0.293*** (0.000) -0.246**** (0.000) -0.332*** (0.000) -0.500*** (0.000) Investment duration 0.082*** (0.000) 0.082*** (0.000) 0.080*** (0.000) 0.034 (0.168) Ln fund age top 25th 0.095 (0.394) 0.075 (0.544) 0.039 (0.727) 0.037 (0.364) Liquidity -0.269*** (0.000) -0.360 (0.308) -0.267*** (0.000) -0.263*** (0.005) Syndication 0.014 (0.291) 0.015 (0.187) 0.011 (0.390) 0.364** (0.025) Constant 0.205*** (0.000) 0.166** (0.041) 0.174*** (0.000) 0.144*** (0.000) Industry Present Present Present Present Financing stage Present Present Present Present Investment year Present Present Present Present Pseudo R square 0.137 0.120 0.109 0.117 No. of obs 4334 839 363 993

This table presents the results of mixed-effects models of VC performance. Panel A shows the descriptive statistics for the additional variables used in panels B and C which measure the absence of a shared land border (cross-border not sharing), geographic distance, cultural disparity, and regulatory differences between a VC firm’s and portfolio company’s home countries. Panel B shows the impact of cross-border investing on performance in countries with and without shared land borders. The dependent variable is IRR and is measured as annualized IRR. Cross-border is one for cross-border investments and zero otherwise. Cross-border (not sharing) is one for cross-border investments in countries without a land border with the VC firm’s home country and zero otherwise. All other variables are defined as in Table 3. Panel C shows the impact of measures of distance and regulatory and institutional differences on performance (IRR). Geographic distance is measured as the distance between the capitals of the respective countries of VC firms and portfolio companies. Cultural distance is determined as the differences in the Hofstede measures of culture (from Geert Hofstede’s website) between the home countries of VC firms and their portfolio companies. Regulatory quality (diff) and political stability (diff) are the differences in the average scores of regulatory quality and political stability (as in Kaufmann, Kraay, & Mastruzzi, 2007, based on data from the World Bank). Legal system (diff) is a binary variable which takes a value of one if the countries of a VC firm and portfolio company have different legal systems and zero otherwise (based on data from Rafael La Porta’s website). Other control variables are as defined in Table 3.

***, **, and * indicate 1, 5, and 10% conventional levels, respectively. a The mixed-effects analyses in panels B and C model investment year and portfolio company industry as fixed effects and financing stage as random.

Cross-border venture-capital investments Axel Buchner et al

593

Journal of International Business Studies

measures (see Hypothesis 2a). However, we expect that cultural disparities and institutional differ- ences reduce exit performance, as predicted by Hypotheses 2b and 2c (see ‘‘Literature, Conceptual Framework, and Hypotheses’’ section).

Unlike all prior studies of cross-border perfor- mance, our analysis measures performance based on deal-level returns (IRR) rather than a proxy of performance based on exit success (Hochberg et al., 2007). We use five different measures of distance, disparity, and difference: the logarithm of geo- graphic distances (in miles); the cultural disparities between the countries of VC firms and portfolio companies, calculated as differences between coun- tries in terms of Hofstede et al.’s (2010) four cultural dimensions (power distance, individual- ism, masculinity, and uncertainty avoidance); and three measures of institutional differences between the home countries of VC firms and their portfolio companies (differences in regulatory quality, polit- ical stability, and the legal systems). See ‘‘Literature, Conceptual Framework, and Hypotheses’’ section and Appendix Table 9 for details of variables and data sources.

Consistent with Hypotheses 2b and 2c (see ‘‘Literature, Conceptual Framework, and Hypothe- ses’’ section), but contrary to the null Hypothesis 2a, our results in panel C of Table 6 show that all five measures which capture geographic distances, cultural disparities, and differences in institutional environments have significant negative impacts on the returns performance of VC deals. The cross- border indicator also remains significant at the 10% level (except in model IV for the ROW). This suggests that a residual negative impact of foreign- ness remains over and above the negative impacts of geographic distances, cultural disparities, and

institutional differences. In the next section, we examine further possible explanations for this cross-border effect.

Extensions In order for our conclusion that cross-border investments underperform domestic investments to be valid, we need to demonstrate that the differences in returns are not merely due to differ- ences in risk. This part of our analysis examines the relative risk of cross-border and domestic deals. Note that there is no generally accepted measure of risk in the context of PE investments. This lack of a common measure is because one cannot observe a time series of market valuations for non-traded assets (such as PE stakes); thus, standard methods which are used to assess risk for standard traded assets are infeasible. As a result, in PE research, there are no standard approaches to correct for risk. Following Degeorge, Martin, and Phalippou (2016), we approximate investment risk in several ways. First, we examine the VC providers’ risk of wealth loss based on the investment multiples of domestic and cross-border investments in each of the four regions. Column (1) of Table 7 shows the propor- tions of domestic and cross-border VC investments with investment multiples of zero; namely, deals which result in bankruptcies (write-offs) and thus a complete loss of wealth for VC providers. The second column (entitled ‘Capital loss’) shows the proportions of deals with investment multiples of less than one which result in (at least) a partial loss of wealth for VC providers. Using tests of differ- ences in proportions, we conclude that there are no statistically (or economically) significant differ- ences between domestic and cross-border invest- ments for both these measures of risk. Next, we

Table 7 Alternative measures of risk (domestic vs. cross-border deals)

Regions Bankruptcy Capital loss Systematic risk

Domestic

(%)

Cross-

border

(%)

Domestic

(%)

Cross-

border

(%)

Domestic deals

(local index of

VC) (1)

Cross-border deals

(local index of

portfolio) (2)

Cross-border

deals (local

index of VC) (3)

Diff

(1) - (2)

Diff

(2) - (3)

North

America

17 20 29 31 2.519 2.402 1.343 1.112 2.113**

Europe 15 18 25 32 2.749 2.661 1.394 1.021 3.022***

UK 18 19 22 24 2.692 2.511 1.241 1.512 2.590**

ROW 23 24 29 33 2.681 2.784 2.567 -1.041 1.481

This table shows a comparison of domestic and cross-border deals using alternative measures of risk. Bankruptcy is calculated as the fraction of investments where the investment multiple is equal to zero. Capital loss is the fraction of investments where the multiple is less than one. Systematic risk is measured as the sensitivity of IRR to the local market index of VC firms with regard to domestic deals (1), the local market index of portfolio companies with regard to cross-border deals (2), and the local market index of VC firms with regard to cross-border deals.

***, **, * indicate 1, 5, and 10% conventional levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

594

Journal of International Business Studies

measure the systematic risk of VC investments in terms of their betas; namely, in terms of the sensitivity of the IRR of VC investments relative to returns on the stock market indices for the home countries of the portfolio companies. Following Axelson, Strömberg, and Sorensen (2013) and Degeorge et al. (2016), we estimate beta as the slope in a regression of IRR on the corresponding stock market return. We find no statistically signif- icant differences between the betas of domestic and cross-border investments in any of the four regions.

The final part of the analysis reported in Table 7 addresses the question of whether cross-border investing helps VC firms to diversify fund risk. Column (3) of Table 7 shows the betas of cross- border investments relative to the stock markets of VC providers’ home countries. We compare the resulting ‘home betas’ in column (3) to the ‘host betas’, shown in column (2), which are calculated relative to the stock markets in the portfolio companies’ countries. If VC providers are able to diversify their portfolios by investing abroad, we expect the IRR of cross-border deals to be less sensitive to the VC firms’ domestic stock indices than to the stock indices of the portfolio compa- nies’ countries. In other words, we expect to find the home betas to be lower than the host betas. Our results confirm that the home betas are statistically significantly lower than the host betas. While the host betas range between 2.5 and 2.8, the home betas range between 1.2 and 1.4 for all regions except in the ROW. Our results suggest that VC firms from North America, Europe, and the UK significantly reduce their exposure to (home) mar- ket risk by investing abroad. Consequently, they should expect lower returns from cross-border than from domestic investments.

Our final analysis examines another possible motive for VC firms to invest in cross-border deals (besides diversification). Past evidence suggests that higher capital inflow into domestic VC industries has led to more competition among VC firms for promising deals. Gompers and Lerner (2000) refer to this as ‘money chasing deals’. In a saturated domestic VC market characterized by an excess of funds and a shortage of investment opportunities, VC firms may seek investment opportunities abroad (Schertler & Tykvová, 2011). We estimate a probit model to investigate the likelihood of a VC firm investing in cross-border deals. This investiga- tion is the first stage of the Heckman model reported in panel A of Table 3. Our explanatory variable of interest measures aggregate capital

inflow into the domestic VC market (ln capital inflow). If cross-border investments are driven by limited domestic investment opportunities and saturation of the domestic VC market, we expect a positive coefficient for ln capital inflow in the probit model. The results of the probit model show a significant positive coefficient of ln capital inflow, which is consistent with the saturation argument. Overall, the results of our extended analysis show

that VC firms benefit from cross-border investing by achieving portfolio diversification and overcom- ing shortages of domestic investment opportunities in saturated markets. These benefits may explain the attraction of cross-border investing to VC investors, despite the poor returns performance of cross-border deals relative to domestic investments.

Robustness We explore whether our models are subject to bias due to the non-randomness of VC exit decisions by using an approach similar to that of Cumming et al. (2009). In the sample used for our foregoing baseline analyses, we include only fully exited investments. As a robustness check, we start with a sample including all (fully exited, partially exited, and unexited) investments. We estimate a Heck- man model for IRR similar to that reported in Table 3. In addition to the probit model of a VC firm’s decision to invest cross-border (as in Table 3), the extended Heckman model includes a second probit model of a VC firm’s decision to exit its investment fully.24 We find no evidence of signif- icant bias; thus, our results are qualitatively unchanged in all regions. The coefficients for our variable of interest, the cross-border indicator, with regard to the four regions remain broadly in line with those reported in our baseline analysis in Table 3 (see Appendix Table 11). Our baseline analysis controls for deal syndica-

tion but not for possible endogeneity of syndica- tion choice. In order to check the robustness of such possible endogeneity, we estimate a Heckman model for IRR with a probit stage which models the syndication decision, following Tian (2012).25 The results of this Heckman model remain qualitatively unchanged with the coefficient of interest, for the cross-border indicator, broadly the same as in our baseline analysis in Table 3 (see Appendix Table 11). Our baseline analysis includes standard control

variables for VC firms and funds; however, we examine whether omitting certain other VC con- trol variables biases our results. Specifically, we

Cross-border venture-capital investments Axel Buchner et al

595

Journal of International Business Studies

control for the ‘busyness’ of VC fund managers, following the literature on the trade-off between VC fund size and investment monitoring quality (Kanniainen & Keuschnigg, 2003; Cumming, 2006; Cumming & Walz, 2010). Busyness is defined as the portfolio (fund) size per fund manager and is intended to measure the impact of lower amounts of managerial time and resources available for any one investment in larger funds. Adding busyness to the second stage of the Heckman model of IRR reported in Table 3, we find that busyness has a statistically significant (at 5%) negative coefficient which ranges between –0.07 in ROW and –0.12 in North America. However, comparing the cross- border coefficients of our baseline model with those of the extended model, we find that our key result, the cross-border effect, remains virtually unchanged in all regions (see Appendix Table 11).

Our baseline model in Table 3 includes year dum- mies for the investment years and controls formarket conditions (in terms of market liquidity) at the times ofexit.Next,weexaminewhether further controls for changes inmarket conditionsduring the investment- holding periods (i.e. from investments to exits) affect our results. To this end, we measure the average returns of the Morgan Stanley Capital International (MSCI) index and the average risk-free returns between the dates of investment and the dates of exit. Including theseadditional controlvariablesdoes not qualitatively affect our main results of interest; namely, our estimates of the coefficient of the cross- border indicator (see Appendix Table 11).

Finally, we examine the robustness of our results using propensity score matching. We match each cross-border deal with an equivalent domestic deal based on the propensity score estimated using VC age, investment size, fund age, investment duration, liquidity, syndication, and financing stage. Using caliper radius matching, we classify a domestic investment as a match for a cross-border invest- ment when the propensity scores for both invest- ments vary by no more than 1% (following e.g. Dehejia & Wahba, 2002). We employ t-tests and Wilcoxon tests to confirm that the cross-border and matched domestic investments are not signifi- cantly different in terms of their mean and median characteristics. Using the sample of matched obser- vations, we estimate a mixed-effects model (similar to the models reported in Table 3). Consistent with our prior results in Table 3, we find that the cross- border coefficient is lower than in our earlier analyses; however, it is still both statistically and economically significant (see Appendix Table 11).26

Overall, our results on the underperformance of cross-border investments remain robust. With regard to North America, we find cross-border coef- ficients ranging from -0.119 to -0.196, from which we conclude that, all else being equal, the cross-border investments of North American VC firms underperform their domestic investments by between 12 and 20% in terms of IRR. Globally, our results are similarly robust, indicating cross-border investment underperformance between 22 and 25% for European VC investors, 25 and 28% for UK VC investors, and comparatively lower under- performance between 15 and 17% in the ROW.

CONCLUSION Against the background of increasing cross-border financial flow, we examine the returns performance of cross-border VC investments. Because of data limitations, the literature analyses cross-border performance in terms of the likelihood of successful VC exit and reports conflicting results. The main contribution of our study is that we are able to estimate actual returns using detailed cash flow data which were unavailable to prior studies. We calculate IRR and PME for 6529 deals conducted by VC firms in North America, the UK, Continental Europe, and the ROW during 1971–2008. We show that cross-border deals underperform domestic deals in terms of IRR by 12–28% depending on the region and research design. Similarly, we find significant cross-border underperformance in PME. Comparing deals before and after the late 1990s, performance is lower for all investments in the latter period; however, the performance differential between domestic and cross-border investments remains unchanged. Geographic distance, cultural disparity, and insti-

tutional differences between the home countries of a VC investor and a portfolio company negatively affect cross-border returns but do not fully explain cross-border underperformance. Moreover, our analysis confirms that differences in risk are unlikely to explain the variation in returns between cross-border and domestic investments. Further analysis shows that VC firms benefit from cross- border investing by achieving portfolio diversifica- tion and overcoming shortages of domestic invest- ment opportunities in saturated markets. These benefits may explain the attraction of cross-border investing to VC firms despite the poor returns performance of cross-border deals relative to domestic investments. While VC firms may choose

Cross-border venture-capital investments Axel Buchner et al

596

Journal of International Business Studies

to invest abroad for many reasons, our results suggest that they should not expect to achieve higher deal-level returns when they do so.

Our results are of clear relevance to VC fund managers (general partners) in guiding their port- folio decisions, including their considerations of relocating portfolio companies (Cumming et al., 2009). The results should also prove useful to investors (limited partners) in VC funds who need to understand the relative performance of cross- border and domestic investments in order to make sound fund-selection decisions and influence investment patterns. The negative impact of insti- tutional differences on VC returns which we observe will be of interest to policymakers and regulators.

ACKNOWLEDGEMENTS We thank Douglas Cumming, Ludovic Phalippou, Armin Schwienbacher, and participants at the T2S 2013 Conference (Bergamo) as well as seminar participants at Technical University of Munich for helpful comments. We thank the seminar participants at Liverpool University School of Management, Manchester Business School, and University of Passau for their helpful comments.

NOTES

1With respect to their analysis of exit as a measure of VC success, Bengtsson and Hsu (2015) point out that an ‘important caveat for this part of our analysis is that we equate the investment outcome with the com- pany’s exit mode due to data limitations. This outcome variable is a coarse measure of investment perfor- mance, though it is commonly used in the entrepreneurship literature’ (p. 340; italics added).

2While previous studies have used CEPRES (e.g. Franzoni et al., 2012; Krohmer et al., 2009; Cumming et al., 2010; Cumming & Walz, 2010), none of these previous studies compare the returns of domestic and cross-border investments.

3CEPRES is a private data provider established in 2001 which offers information on VC deals worldwide.

4We focus on fully exited (realized) deals to avoid issues related to the accuracy of the estimated net asset values (NAVs) of unrealized deals or timing issues about when the NAVs are reported. We examine the sensitivity of our results to the inclusion of partial exits and non-exits in the robustness section

5Appendix Table 8 provides a breakdown of the distribution of VC deals by region during 1971–2009.

6We observe the countries of origin of VC firms and portfolio companies at the time of the investments. A limitation of our data is that we do not observe relocations to other countries by either the VC firms or the portfolio companies after the initial investments.

7Note that the IRR estimated in our analysis are gross returns as opposed to returns net of fees and the costs (transaction, search, and monitoring) incurred by VC firms when undertaking and managing investments.

8In our analyses, we winsorize IRR at 1%. 9We expect that aggregate capital inflow into the

VC industry of a given country makes it more likely that VC firms invest abroad as competition among VC firms for domestic investments becomes more intense. This causes VC firms to search out investment oppor- tunities abroad.

10The CEPRES database we use only shows whether an investment is syndicated or not. Unfortunately, we cannot distinguish between domestic and foreign syndication.

11For detailed discussions of Hofstede measure see Beugelsdijk, Kostova, and Roth (2017).

12Median IRR for the ROW also show that domestic investments outperform cross-border investments. The average and median returns on domestic investments by ROW VC firms are comparable to those reported for Asia–Pacific VC firms during 1989–2001 in Cumming et al. (2009).

13Based on median age, the results are broadly the same with the exception of the ROW, where the difference in medians is only half a year.

14Panel B of Table 2 shows the natural logarithms of VC age, fund age, and investment size which are used in our multivariate analysis.

15We focus here on the second stage of the two- stage models. The first stage involves a probit model with cross-border as the dependent variable and aggregate capital inflow as an instrument, as outlined in ‘‘Data and Methodology’’ section. The results of this first stage are discussed in greater detail in the extensions of our baseline analysis in ‘‘Extensions’’ section.

16There is no evidence of multicollinearity among the variables used in our study. Appendix Table 10 provides a correlation matrix for the variables used in the analysis.

17We also use 1998 and 1999 as cut-off points. Our results remain unchanged.

18The results of the additional analyses, including deal destination and VC experience as moderating

Cross-border venture-capital investments Axel Buchner et al

597

Journal of International Business Studies

factors, are not tabulated here but are available from the authors on request. We are grateful to an anony- mous referee for suggesting this analysis of moderat- ing factors.

19We thank an anonymous referee for helping to clarify our interpretation.

20The results of propensity score matching are discussed in ‘‘Robustness’’ section. The unreported mixed-effects models are presented in Table 7 but without the measures of distance. The results of the unreported analyses are available from the authors on request.

21Panel A of Table 6 reports descriptive statistics for the additional variables used in the analyses presented in panels B and C of the same table.

22The mixed-effects analyses in panels B and C of Table 6 model investment year and industry of the portfolio company as fixed effects and financing stage as random effects. Our choice between fixed and random effects is based on likelihood ratio tests. Following common practice in the relevant literature (e.g. Chemmanur et al., 2016; Degeorge et al., 2016), industry fixed effects are included to control for unobserved time-invariant industry characteristics. Our results may be biased if the unobserved industry

characteristics are time variant. We thank an anony- mous referee for noting this issue.

23We thank an anonymous referee for suggesting this term and the associated analysis.

24Full exits are coded one, while partial exits and no exits are coded zero. Explanatory variables are invest- ment duration and economic conditions at exit, as in Cumming et al. (2009). The estimated coefficient of the lambda of this full-exit probit model is statistically insignificant.

25Syndication is modelled as a function of the logarithms of capital inflow, VC age, fund age, and investment size. Unfortunately, the database we use does not allow us to differentiate between domestic and foreign syndication.

26Because propensity score matching is solely based on observable characteristics, it only addresses con- cerns of selection bias and endogeneity due to observable differences between treatment and control groups (e.g. Dehejia & Wahba, 2002). In contrast, the two-stage Heckman models given in Table 3 allow for endogeneity due to unobservable differences and omitted variables.

REFERENCES Admati, A., & Pfleiderer, P. 1994. Robust financial contracting and the role of venture capitalists. Journal of Finance, 49: 371–402.

Aizenman, J., & Kendall, J. 2012. The internationalization of venture capital. Journal of Economic Studies, 39: 488–511.

Amit, R., Brander, J., & Zott, C. 1998. Why do venture capital firms exist? Theory and Canadian evidence. Journal of Business Venturing, 13: 441–466.

Axelson, U., Strömberg, P., & Sorensen, M. 2013. The alpha and beta of buyout deals: A jump CAPM for long term illiquid investments. London School of Economics and Political Science, Working Paper. London School of Economics.

Bengtsson,O.,&Hsu,D. 2015. Ethnicmatching in theUS venture capital market. Journal of Business Venturing, 30: 338–354.

Bertoni, F., & Groh, A. 2014. Cross-border investments and venture capital exits in Europe. Corporate Governance An International Review, 22: 84–99.

Beugelsdijk, S., Kostova, T., & Roth, K. 2017. An overview of Hofstede-inspired country-level culture research in interna- tional business since 2006. Journal of International Business Studies, 48: 30–47.

Black, S., & Gilson, J. 1998. VC and the structure of capital markets: Banks versus stock markets. Journal of Financial Economics, 47: 243–277.

Cao, J., Cumming, D., Qian, M., & Wang, X. 2015. Cross border LBOs. Journal of Banking & Finance, 50: 69–80.

Chemmanur, T., Hull, T., & Krishnan, K. 2016. Do local and international venture capitalists play well together? A study of international venture capital investments. Journal of Business Venturing, 31: 573–594.

Chen, H., Gompers, P., Kovner, A., & Lerner, J. 2010. Buy local? The geography of successful and unsuccessful venture capital expansion. Journal of Urban Economics, 67: 90–110.

Cochrane, J. 2005. The risk and return of venture capital. Journal of Financial Economics, 75: 3–52.

Cornelli, F., Kominek, Z., & Ljungqvist, A. 2013. Monitoring managers: Does it matter? Journal of Finance, 68: 431–481.

Coval, J., & Moskowitz, T. 1999. Home bias at home: Local equity preference in domestic portfolio. Journal of Finance, 54: 2045–2073.

Cumming, D. 2006. The determinants of venture capital portfolio size: Empirical evidence. Journal of Business, 79: 1083–1126.

Cumming, D. 2008. Contracts and exits in venture capital finance. Review of Financial Studies, 21: 1947–1982.

Cumming, D., & Dai, N. 2010. Local bias in venture capital investments. Journal of Empirical Finance, 17: 362–380.

Cumming, D., Fleming, G., & Schwienbacher, A. 2005. Liquid- ity risk and venture capital finance. Financial Management, 34: 77–105.

Cumming, D., Fleming, G., & Schwienbacher, A. 2006. Legality and venture capital exits. Journal of Corporate Finance, 12: 214–245.

Cumming, D., Fleming, G., & Schwienbacher, A. 2009. Corpo- rate relocation in venture capital finance. Entrepreneurship Theory and Practice, 33: 1121–1155.

Cumming, D., & Johan, S. 2007. Regulatory harmonization and the development of private equity market. Journal of Banking & Finance, 31: 3218–3250.

Cumming, D., Knill, A., & Syvrud, K. 2016. Do international investors enhance private firm value? Evidence from venture capital. Journal of International Business Studies, 47: 347–373.

Cumming, D., Schmidt, D., & Walz, U. 2010. Legality and VC governance around the world. Journal of Business Venturing, 25: 54–72.

Cross-border venture-capital investments Axel Buchner et al

598

Journal of International Business Studies

Cumming, D., & Walz, U. 2010. Private equity returns and disclosure around the world. Journal of International Business Studies, 41: 727–754.

Dai, N., Jo, H., & Kassicieh, S. 2012. Cross-border venture capital investments in Asia: selection and performance. Journal of Business Venturing, 27: 666–684.

Dai, N., & Nahata, R. 2016. Cultural differences and cross- border venture capital syndication. Journal of International Business Studies, 47: 140–169.

De Prijcker, S., Manigart, S., Wright, M., & De Maeseneire, W. 2012. The influence of experiential, inherited and external knowledge on internationalization of venture capital firms. Journal of International Business Review, 21: 929–940.

Degeorge, F., Martin, J., & Phalippou, L. . 2016. On secondary buyouts. Journal of Financial Economics, 120: 124–145.

Dehejia, R., & Wahba, S. 2002. Propensity score matching methods for nonexperimental causal studies. The Review of Economics and Statistics, 84: 151–161.

Devigne, D., Manigart, S., & Wright, M. 2016. Escalation of commitment in venture capital decision making: Differentiat- ing between domestic and international investors. Journal of Business Venturing, 31: 253–271.

Devigne, D., Vanacker, T., Manigart, S., & Paeleman, I. 2013. The role of domestic and cross-border venture capital investors in the growth of portfolio companies. Small Business Economics, 40: 553–573.

Espenlaub, S., Khurshed, A., & Mohamed, A. 2015. Venture capital exits in domestic and cross-border investments. Journal of Banking & Finance, 53: 215–232.

Franzoni, F., Nowak, E., & Phalippou, L. 2012. Private equity performance and liquidity risk. Journal of Finance, 67: 2341–2373.

Giot, P., & Schwienbacher, A. 2007. IPOs, trade sales and liquidation: Modelling VC exits using exits using survival analysis. Journal of Banking & Finance, 31: 697–702.

Gompers, P. 1995. Optimal investment, monitoring, and the staging of venture capital. Journal of Finance, 50: 1461–1489.

Gompers, P., & Lerner, J. 2000. Money chasing deals? The impact of fund inflows on private equity valuations. Journal of Financial Economics, 55: 281–325.

Gompers, P., & Lerner, J. 2004. The venture capital cycle. Cambridge and London: MIT Press.

Gorman, M., & Sahlman, W. 1989. What do venture capitalists do? Journal of Business Venturing, 4: 231–248.

Guler, I., & Guillén, M. 2010. Knowledge, institutions and organizational growth: The internationalization of US venture capital firms. Journal of International Business Studies, 41: 185–205.

Hain, D., Johan, S., & Wang, D. 2015. Determinants of cross- border venture capital investments in emerging and devel- oped economies: The effects of relational and institutional trust. Journal of Business Ethics, 138: 743–764.

Harris, R., Jenkinson, T., & Kaplan, S. 2014. Private equity perfor- mance: What do we know? Journal of Finance, 69: 1851–1882.

Hesketh, S., Skrondal, A., & Pickles, A. 2005. Maximum likelihood estimation of limited and discrete dependent variable models with nested random effects. Journal of Econometrics, 128: 301–323.

Hochberg, Y., Ljungqvist, A., & Lu, Y. 2007. Whom you know matters: Venture capital networks and investment perfor- mance. Journal of Finance, 62: 251–301.

Hofstede, G. 1980. Culture’s consequences: International differ- ences in work-related values. Beverly Hills, CA: Sage.

Hofstede, G., Hofstede, G. J., & Minkov, M. 2010. Cultures and organizations: Software of the mind, 3rd edn. New York: McGraw-Hill.

Humphery-Jenner, M., & Suchard, Jo. 2013. Foreign venture capitalists and the internationalization of entrepreneurial companies: Evidence from China. Journal of International Business Studies, 44: 607–621.

Hursti, J., & Maula, M. 2007. Acquiring financial resources from foreign equity capital markets: An examination of factors

influencing foreign initial public offerings. Journal of Business Venturing, 22: 833–851.

Jääskeläinen, M., & Maula, M. 2014. Do networks of financial intermediaries help reduce local bias? Evidence from cross- border venture capital exits. Journal of Business Venturing, 29: 704–721.

Kanniainen, V., & Keuschnigg, C. 2003. The optimal portfolio of start-up firms in venture capital finance. Journal of Corporate Finance, 9: 521–534.

Kaplan, S., & Schoar, A. 2005. Private equity performance: Returns, persistence and capital flows. Journal of Finance, 60: 1791–1823.

Kaufmann, D., Kraay, A., & Mastruzzi, M. 2007. Governance matters VI: Governance indicator for 1996–2006. World Bank Policy Research.

Kirkman, L., Lowe, B., & Gibson, B. 2006. A quarter century of culture’s consequences: A review of empirical research incor- porating Hofstede‘s cultural values framework. Journal of International Business Studies, 37: 285–320.

Kogut, B., & Singh, H. 1988. The effect of national culture on the choice of entry mode. Journal of International Business Studies, 19: 411–432.

Krohmer, P., Lauterbach, R., & Calanog, V. 2009. The bright and dark side of staging: Investment performance and the varying motivations of private equity firms. Journal of Banking & Finance, 33: 1597–1609.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. 1998. Law and finance. Journal of Political Economy, 106(6): 1113–1155.

Lerner, J. 1994. Venture capitalists and the decision to go public. Journal of Financial Economics, 35: 293–316.

Li, Y., Vertinsky, I., & Li, J. 2014. National distances, interna- tional experience, and venture capital investment perfor- mance. Journal of Business Venturing, 29: 471–489.

Li, Y., & Zahra, A. 2012. Formal institutions, culture, and venture capital activity: A cross-country analysis. Journal of Business Venturing, 27: 95–111.

Mäkelä, M., & Maula, M. 2008. Attracting cross-border venture capital. The role of a local investor. Entrepreneurship and Regional Development, 20: 237–257.

Manigart, S., & Wright, M. 2013. Reassessing the relationships between private equity investors and their portfolio compa- nies. Small Business Economic, 40: 449–479.

Meuleman, M., Jääskeläinen, M., Maula, M., & Wright, M. 2017. Venturing into the unknown with strangers: Substitutes of relational embeddedness in cross-border partner selection in venture capital syndicate. Journal of Business Venturing, 32: 131–144.

Meuleman, M., & Wright, M. 2011. Cross-border private equity syndication: Institutional context and learning. Journal of Business Venturing, 26: 35–48.

Nahata, R., Hazarika, S., & Tandon, K. 2014. Success in global venture capital investing: Do institutional and cultural differ- ences matter? Journal of Financial and Quantitative Analysis, 49: 1039–1070.

Portes, R., & Rey, H. 2005. The determinants of cross-border equity flows. Journal of International Economics, 65: 269–296.

Poterba, J., & French, K. 1991. Investor diversification and international equity markets. American Economic Review, 81: 222–226.

Reuer, J., & Ragozzino, R. 2014. Signals and international alliance formation: The roles of affiliations and international activities. Journal of International Business Studies, 45: 321–337.

Rubinstein, M. 1976. The valuation of uncertain income streams and the pricing of options. Bell Journal of Economics and Management Science, 7: 407–425.

Ruhnka, J., & Young, J. 1991. Some hypothesis about risk in venture capital investing. Journal of Business Venturing, 6: 115–133.

Cross-border venture-capital investments Axel Buchner et al

599

Journal of International Business Studies

Sahlman, W. 1990. The structure and governance of venture capital organizations. Journal of Financial Economics, 27: 473–521.

Sapienza, H. 1992. When do venture capitalists add value? Journal of Business Venturing, 7: 9–27.

Sapienza, H., Manigart, S., & Vermeir, W. 1996. Venture capitalist governance and value added in four countries. Journal of Business Venturing, 6: 439–469.

Schertler, A., & Tykvová, T. 2011. Cross-border venture capital flows and local ties: Evidence from developed countries. The Quarterly Review of Economics and Finance, 51(1): 36–48.

Schertler, A., & Tykvová, T. 2012. What lures venture capital inflows? Journal of International Money and Finance, 31: 1777–1799.

Seppa, T., & Laamanen, T. 2001. Valuation of venture capital investments empirical evidence. R&D Management, 31: 215–230.

Sivakumar, K., & Nakata, C. 2001. The stampede toward Hofstede’s framework: Avoiding the sample design pit in cross-cultural research. Journal of International Business Studies, 32: 555–574.

Sojli, E., & Tham, W. 2017. Foreign political connections. Journal of International Business Studies, 48: 244–266.

Sorensen, M., & Jagannathan, R. 2015. The public market equivalent and private equity performance. Financial Analysts Journal, 71: 43–50.

Sorensen, O., & Stuart, T. 2001. Syndication networks and the spatial distribution of venture capital investments. The Amer- ican Journal of Sociology, 106: 1546–1588.

Taussig, M. 2017. Foreignness as both a global asset and a local liability: How host country idiosyncrasies and business

activities matter. Journal of International Business Studies, 48: 498–522.

Tian, X. 2012. The role of venture capital syndication in value creation for entrepreneurial firms. Review of Finance, 16: 245–283.

Tykvová, T., & Schertler, A. 2011. Cross-border venture capital flows and local ties: Evidence from developed countries. The Quarterly Review of Economics and Finance, 51: 36–48.

Wang, L., & Wang, S. 2012. Economic freedom and cross- border venture capital performance. Journal of Empirical Finance, 19: 26–50.

Wright, M., Lockett, A., & Pruthi, S. 2005a. Internationalization of Western venture capitalists into emerging markets: Risk assessment and information in India. Small Business Economics, 19: 13–29.

Wright, M., Pruthi, S., & Lockett, A. 2005b. International venture capital research: From cross-country comparisons to crossing borders. International Journal of Management Review, 7: 135–165.

Wright, M., & Robbie, K. 1998. Venture capital and private equity: A review and synthesis. Journal of Business Finance and Accounting, 25: 521–570.

Wu, Z., & Salomon, R. 2017. Deconstructing the liability of foreignness: Regulatory enforcement actions against foreign banks. Journal of International Business Studies 48: 837–861.

Wuebker, R., Kraeussl, R., & Schulze, W. 2016. Is venture capital really a local business? Evidence from US cross-border invest- ment (unpublished working paper).

Zaheer, S. 1995. Overcoming the liability of foreignness. Academy of Management Journal, 38: 341–363.

APPENDIX See Tables 8, 9, 10, and 11.

Cross-border venture-capital investments Axel Buchner et al

600

Journal of International Business Studies

T a b le

8 D is tr ib u ti o n o f ve n tu re

ca p it a l d e a ls b y re g io n fr o m

1 9 7 1 to

2 0 0 9

In ve st m e n t

y e a rs

A ll d e a ls

(n u m b e r)

N o rt h A m e ri ca

(N A )

E U

(E x . U K )

U K

R O W

D o m e st ic

d e a ls

(n u m b e r)

C ro ss -

b o rd e r (C

B )

(n u m b e r)

C B to

to ta l

N A (%

)

D o m e st ic

d e a ls

(n u m b e r)

C ro ss -

b o rd e r (C

B )

(n u m b e r)

C B to

to ta l

E U

(% )

D o m e st ic

d e a ls

(n u m b e r)

C ro ss -

b o rd e r

(C B )

(n u m b e r)

C B to

to ta l

U K (%

)

D o m e st ic

d e a ls

(n u m b e r)

C ro ss -

b o rd e r

(C B )

(n u m b e r)

C B to

to ta l

R O W

(% )

1 9 7 1 – 1 9 8 0

3 5

0 0

– 0

0 –

0 0

–

1 9 8 1 – 1 9 9 0

8 5 0

6 9 1

5 0

6 .7 5

3 9

1 6

2 9 .0 9

1 1

1 2

5 2 .1 7

2 9

2 6 .4 5

1 9 9 1 – 2 0 0 0

4 4 3 5

2 5 9 5

1 9 4

6 .9 5

4 2 4

2 1 6

3 3 .7 5

1 7 8

7 7

3 0 .2 0

6 2 1

1 3 0

1 7 .3 1

2 0 0 1 – 2 0 0 9

1 2 0 9

6 2 4

1 4 5

1 8 .8 6

9 1

5 4

3 7 .2 4

6 7

1 7

2 0 .2 4

1 4 5

6 6

3 1 .2 8

T o ta l

6 5 2 9

3 9 4 5

3 8 9

8 .9 7

5 5 4

2 8 6

3 4 .0 5

2 5 6

1 0 6

2 9 .2 8

7 9 5

1 9 8

1 9 .9 4

Cross-border venture-capital investments Axel Buchner et al

601

Journal of International Business Studies

Table 9 Definitions of variables

Variable Definition of variable

Cross-border A dummy variable taking a value of one if a portfolio company and the VC firm are located in different

countries and zero otherwise

Cross-border not

sharing

A dummy variable taking a value of one if a portfolio company is located in a country which does not share a

border with the home country of the VC firm, and zero otherwise

Geographic distance Measured as the distance in miles between the capitals of the countries of a VC firm and portfolio company.

We obtain the data from the CEPII website (www.cepii.fr/anglaisgraph/bdd/distances.htm)

Cultural distance As in Dai and Nahata (2016), we follow the approach of Kogut and Singh (1988) and use the Hofstede

measures of culture (i.e. power distance, individualism, masculinity, and uncertainty avoidance) to compute

the cultural distance between a VC firm and a portfolio company. The data are obtained from Geert Hofstede’s

website (www.geerthofstede.nl). The following Cartesian distance measure is used to calculate cultural dis-

parity: Hofstede cultural difference = P4

i¼4 CLocal;i�Ccross�border;ið Þ2

� �1 2

4 ;where Clocal, i is the local VC firm’s culture based

on measure i and Ccross-border, i is the cultural measure of a portfolio company based on measure i

Regulatory distance This is the difference in the regulatory quality score between the home countries of a VC firm and portfolio

company. The score value for each country ranges between 0 and 100. The data are obtained from the

Worldwide Governance Indicator (WGI) of the World Bank website (http://data.worldbank.org/data-catalog/

worldwide-governance-indicators)

Political distance This is the difference in the political stability score between the home countries of a VC firm and portfolio

company. The score value for each country ranges between 0 and 100. The data are obtained from the WGI of

the World Bank website (http://data.worldbank.org/data-catalog/worldwide-governance-indicators)

Legal system

difference

As in Chemmanur et al. (2016), this is a dummy variable coded 1 if the (common or civil law) origins of the

legal systems of the home countries of a VC firm and portfolio company differ, and zero otherwise. The data

are obtained from Rafael La Porta’s website (http://faculty.tuck.dartmouth.edu/rafael-laporta/research-

publications)

Capital inflow The aggregate amount of capital inflow into the VC industry in the home country of a VC firm in the year of

investment. The data are obtained from the Preqin database

Ln VC age Natural logarithm of the age (years in business) of a VC firm at the time of an initial investment in a portfolio

company

Ln investment size Natural logarithm of the total amount invested by a VC firm in a given portfolio company (in USD million)

Ln fund age Natural logarithm of a fund’s age measured in years from the date of the fund’s initiation to the date of a VC

investment in a given portfolio company

Ln fund age top

25th

Natural logarithm of a fund’s age for older funds, specifically funds above the 75th age percentile, and zero

otherwise

Investment duration The time between an initial VC investment and the VC exit date (in years)

Liquidity Stock market liquidity of a portfolio company’s country of origin measured as the ratio of the value of shares

traded on the country’s stock exchange(s) divided by the country’s GDP

Syndication A binary variable equal to one if the investment is syndicated and zero otherwise.

IRR The internal rate of return (IRR) based on detailed information of investment cash flow, as described in ‘‘Data

and Methodology’’ section. IRR is winsorized at 1% in all analyses

PME Public market equivalent, as described in ‘‘Data and Methodology’’ section

Post-1997 Binary indicator equal to zero for investments up to 1997 and one thereafter

Year A set of binary indicator variables for the year of a VC investment

Financing stage Binary variables for a financing stage. We include an indicator for investments in expansion stages (zero

otherwise) and another for investments in later stages (zero otherwise). Early-stage investments are used as the

base

Industry A set of binary industry variables for a portfolio company’s primary industry focus. We include indicators for

biotechnology, consumer goods and services, financials, industrials, and information technology, treating

‘others’ as the base

Cross-border venture-capital investments Axel Buchner et al

602

Journal of International Business Studies

http://www.cepii.fr/anglaisgraph/bdd/distances.htm

http://www.geerthofstede.nl

http://data.worldbank.org/data-catalog/worldwide-governance-indicators

http://faculty.tuck.dartmouth.edu/rafael-laporta/research-publications

T a b le

1 0

C o rr e la ti o n m a tr ix

V a ri a b le s

(1 )

(2 )

(3 )

(4 )

(5 )

(6 )

(7 )

(8 )

(9 )

(1 0 )

(1 1 )

(1 2 )

(1 3 )

Fu n d a g e

(1 )

V C

a g e

(2 )

0 .1 6 9

In ve st m e n t si ze

(3 )

- 0 .1 1 5

0 .2 8 1

IR R

(4 )

0 .0 2 5

- 0 .0 0 3

- 0 .0 3 7

P M E

(5 )

- 0 .0 0 6

- 0 .0 1 2

- 0 .0 0 5

0 .0 0 3

H o ld in g p e ri o d

(6 )

0 .0 1 9

0 .2 0 8

0 .5 1 5

- 0 .0 3 2

- 0 .0 1 2

Li q u id it y

(7 )

0 .0 1 1

- 0 .0 5 8

0 .0 3 2

- 0 .0 2 0

- 0 .0 0 2

0 .1 3 3

S y n d ic a ti o n

(8 )

- 0 .0 9 2

0 .0 1 8

0 .0 3 4

0 .0 2 0

- 0 .0 1 9

0 .1 1 9

0 .0 1 9

IP O

(9 )

0 .0 2 0

- 0 .0 0 2

- 0 .1 4 9

0 .1 6 3

- 0 .0 0 3

- 0 .2 5 0

- 0 .0 0 6

0 .0 2 2

M & A

(1 0 )

0 .0 1 3

- 0 .0 5 7

- 0 .0 3 3

0 .1 4 5

- 0 .0 0 8

- 0 .0 4 6

- 0 .0 3 5

- 0 .0 3 9

- 0 .2 3 1

E a rl y

(1 1 )

- 0 .0 3 3

0 .1 1 8

- 0 .0 5 0

- 0 .0 3 9

0 .0 1 8

0 .0 7 2

- 0 .0 3 9

0 .1 4 1

- 0 .0 8 8

- 0 .0 2 0

E x p a n si o n

(1 2 )

0 .0 3 4

- 0 .1 4 7

0 .0 3 3

- 0 .0 0 3

- 0 .0 0 4

- 0 .1 3 5

0 .0 5 9

- 0 .1 4 4

0 .0 5 3

- 0 .0 2 2

- 0 .7 3 4

La te r

(1 3 )

0 .0 0 5

0 .0 1 4

0 .0 3 0

0 .0 5 9

- 0 .0 0 4

0 .0 6 5

- 0 .0 1 7

- 0 .0 2 3

0 .0 6 1

0 .0 5 7

- 0 .5 2 6

- 0 .1 9 2

Cross-border venture-capital investments Axel Buchner et al

603

Journal of International Business Studies

ABOUT THE AUTHORS Axel Buchner is an Associate Professor of Finance at the University of Passau, Germany. His research focuses on private equity, venture capital, corporate finance, asset pricing, and derivatives valuation. His work has been published in leading academic journals. He holds a PhD in Finance from the Technical University of Munich and a Master’s in Business Administration from the University of Munich.

Susanne Espenlaub is a Professor of Finance at Alliance Manchester Business School, University of Manchester (UK). She holds a PhD and MPhil from Oxford University. Her research focuses on initial public offerings, event-study methodology, corpo- rate contracting, corporate governance, venture capital, and share repurchase.

Arif Khurshed is a Professor of Finance at Alliance Manchester Business School, University of Manch- ester (UK). He graduated from the University of Delhi and received his PhD from the University of Reading. His research interests include initial public offerings, venture capital, trade credit, corporate debt, and share repurchase programmes.

Abdulkadir Mohamed is an Associate Professor of Finance at the Cranfield School of Management, UK. His research focuses on private equity, venture capital, corporate finance, and survival analysis. His work has been published in leading academic journals. He holds a PhD and MSc in Finance from Manchester Business School.

Accepted by Becky Reuber, Area Editor, and Douglas Cumming, Guest Editor, 9 September 2017. This article has been with the authors for three revisions.

Table 11 Robustness

Coefficient of cross-border indicator Model I: North

America

Model II: Europe

(Ex. UK)

Model III: UK Model IV: ROW

Coeff P value Coeff P value Coeff P value Coeff P value

(1) Baseline Heckman model (from

Table 3)

-0.192** (0.033) -0.232** (0.017) -0.270** (0.023) -0.156** (0.017)

(2) Heckman model with two probit

stages (for exit decision and for cross-

border investing)

-0.196** (0.041) -0.248** (0.028) -0.278** (0.033) -0.166** (0.025)

(3) Heckman model with instrumental

variable for syndication

-0.1843** (0.045) -0.2204** (0.019) -0.2592** (0.026) -0.1498** (0.019)

(4) Heckman model with additional stage

II variables (market conditions, busyness)

-0.179** (0.044) -0.229** (0.020) -0.254** (0.025) -0.145** (0.018)

(5) Mixed-effects model with propensity

score matching

-0.119** (0.032) – – – – – –

This table shows the coefficients of cross-border indicators for different models by regions. The results are reported by the baseline Heckman model (1), the three-stage Heckman model (2), the instrumental variable for syndication (3), market conditions and busyness (4), and propensity score matching (5). The control variables are as reported in the prior tables.

***, **, and * indicate significance at the 1, 5, and 10% levels, respectively.

Cross-border venture-capital investments Axel Buchner et al

604

Journal of International Business Studies

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.

Cross-border venture capital investments: The impact of foreignness on returns

Abstract
Introduction
Literature, Conceptual Framework, and Hypotheses
Data and Methodology

Data Sources and Sample
Methodology

Results

Univariate Analysis
Multivariate Analysis
Alternative Measures of Performance
The Effect of Distance
Extensions
Robustness

Conclusion
Acknowledgements
References
Bib1