Venture capital and high-tech start-ups in Europe: a systematic review of the empirical evidence

ABSTRACT

Empirical literature on venture capital (VC) and VC-related policies posit that VC can play a major role in helping high-tech start-ups grow and innovate. Motivated by the VC policies recently adopted in Europe and by the heterogeneity of the VC effects across countries and regions, we assess this hypothesis in Europe. Using systematic literature review methods, we summarize the results of 34 firm-level studies published between 2000 and 2023 that estimate value-added effects of VC on funded firms in 17 European countries. The results show a clear preponderance of positive effects on firm expansion and investment. The evidence on the effects on innovation and productivity is more limited and mixed. Regarding the channels of these effects, there is considerable evidence that VC reduces credit constraints of funded firms and limited, but supportive, evidence that VC reduces other barriers to growth. The type of VC emerges as a key moderating factor. Private, independent VC tends to have the largest effects. Government VC tends to have lower, but positive, effects and it often complements private VC. Based on the results of the review and on the current landscape of VC in Europe, we provide recommendations for policy and future research.

KEYWORDS:

• Venture capital
• start-ups
• firm growth
• public policy
• Europe
• systematic literature review

1. Introduction

Increasing the supply of venture capital (VC) has become a major focus of policies that aim to support high-tech start-ups in Europe (European Commission 2016, 2021). These policies, which include InvestEU, European Scale-up Action for Risk Capital (ESCALAR) and European Innovation Council, involve large public resources.¹ Currently, even more ambitious policies are discussed.² These policies are motivated by the VC investment gap in Europe (Durufle et al. 2017; European Investment Bank 2020; Quas et al. 2022) and by the contribution of VC to the development of the high-tech sector the United States (US) (Akcigit et al. 2022; Schnitzer and Watzinger 2022). However, recent reviews (Haslanger, Lehmann, and Seitz 2023; Lohwasser 2020) show that the effects of VC vary significantly across countries and regions and that they tend to be higher in North America than in Europe. This makes the extrapolation of the VC effects in the US to the European context problematic. Moreover, the empirical literature on the effects of VC is highly heterogeneous in terms of the effects estimated, international contexts studied and methods used. This makes it difficult to assess the preponderance of evidence for different effects and their relevancy for the European context.

Motivated by the VC-related policies recently adopted in Europe and the heterogeneity of the VC effects, this study assesses the preponderance of evidence on VC effects on different aspects of firm performance and for different types of VC in Europe. It aims to shed light on the following questions. Which firm level outcomes are more likely to be achieved by increasing VC in Europe? What are the effects of different types of VC? Which policy relevant aspects have not received sufficient attention in the academic literature?

European VC market has several distinctive characteristics compared to the US market. First, it is smaller, both in absolute and relative terms³, as shown in Figure 1.

Figure 1. VC as a percentage of GDP.

Structural and Demographic Business statistics, OECD (2020).

Second, it is more fragmented along national borders due to differences in legislation and the “home bias” of VC investors (Botsari, Crisanti, and Lang 2019; European Commission 2021). Third, it is geographically concentrated, with 10 EU countries accounting for 90% of all VC (Nepelski, Piroli, and De Prato 2016). Finally, government VC (GVC) and bank VC (BVC) play more important roles in Europe than in the US and their effects on firm performance tend to be lower than the effects of private VC (PVC) and independent VC (IVC) (Andrieu 2013; Quas et al. 2022). Some of these characteristics are linked to a slower development of VC market and will become less prominent over time. However, other characteristics, like, the fragmentation of VC markets and the role of GVC, are structural and are likely to persist. As start-ups tend to raise capital locally, these characteristics result in thin VC markets in many EU countries, fewer scale-ups,⁴ lower performance, with negative consequences for the size and competitiveness of high-tech sector (Aernoudt 2017; Duruflé, Hellmann, and Wilson 2017).

To inform the European Union (EU) policies discussed above and similar national and regional interventions, this review summarizes the most recent and rigorous estimates of VC effects in Europe. It reviews the results of 34 published articles covering 17 European countries. The focus is on articles that estimate value-added effects of VC, which are the most relevant for policy. The results are summarized using systematic literature review instead of meta-analysis due to the heterogeneity in the methods used and outcomes studied, which reflect different aspects of firm performance.

This review makes several contributions to the literature. First, it has a geographical focus on Europe, as it aims to inform VC-related policies in Europe. Other reviews (Haslanger, Lehmann, and Seitz 2023; Lohwasser 2020; Rosenbusch, Brinckmann, and Müller 2013; Tykvová 2018b) take a worldwide perspective and rely, largely, on estimates of VC effects in North America and Asia and on estimates obtained using cross-country samples where firms in these regions account for large shares. While they provide important insights, there is evidence that VC effects differ systematically across regions and countries (Grilli, Latifi, and Mrkajic 2019; Haslanger, Lehmann, and Seitz 2023; Lohwasser 2020; Tykvová 2018a). This renders problematic generalizing the results of these studies to the European context. We add to the literature on heterogeneity of the effects of VC across countries and regions by evaluating the preponderance of evidence on these effects in Europe, a relatively under-researched context.

Second, this review summarizes the effects of GVC,⁵ which plays a major role in Europe (Andrieu 2013; Quas et al. 2022), but received limited attention in earlier reviews. These reviews and meta-analyses either examine the effects of VC, without distinguishing between different subtypes (Lohwasser 2020), or focus on the effects of PVC (Haslanger, Lehmann, and Seitz 2023). By summarizing the effects of GVC in Europe, we contribute to the literature on the effects of different types of VC (Chemmanur, Loutskina, and Xuan 2014; Guerini and Quas 2016; Masucci et al. 2021; Quas et al. 2022) and to the debate on the use of GVC as a policy instrument.

Third, this is the first systematic literature review on the effects of VC that focuses on the value-added effects of VC on firm performance, and, thus on studies that account for selection into VC. Previous reviews (Lohwasser 2020; Manigart and Wright 2013; Peneder 2012; Rosenbusch, Brinckmann, and Müller 2013) cover studies using a variety of methods. However, Rosenbusch, Brinckmann, and Müller (2013) and Lohwasser (2020) argue that the methodological heterogeneity affects the comparability of the estimates, leading to inconclusive results. Our methodological focus narrows the scope of the review, but addresses this problem. This focus is in line with the increased importance given to selection bias in applied economics (Angrist and Pischke 2010), recent trends in literature reviews (Dvouletý, Srhoj, and Pantea 2022; Kersten et al. 2017) and in the empirical literature on VC effects (Tykvová 2018b). This study adds to these strands of literature and to policy debate on VC, as value-added effects may justify policy interventions.

Finally, this study provides a nuanced and comprehensive picture of the effects of VC in Europe, by examining the effects on different aspects of firm performance that can be affected differently by VC. Several meta-analyses on VC effects on firm performance (Haslanger, Lehmann, and Seitz 2023; Lohwasser 2020; Rosenbusch, Brinckmann, and Müller 2013) highlight this aspect as a potential problem for summarizing the effects on firm performance. We also examine the channels and the moderating factors of these effects. Thus, this study contributes to the literature that documents the heterogeneity of the VC effects across different firm level outcomes (Bronzini, Caramellino, and Magri 2020; Colombo and Murtinu 2017; Haslanger, Lehmann, and Seitz 2023; Lohwasser 2020; Rosenbusch, Brinckmann, and Müller 2013) and to policy debate, by providing information on which outcomes are more likely to be achieved by increasing VC.

The review reaches several substantive conclusions. First, there is a clear preponderance of the positive effects of VC on firm expansion and investment, but evidence on other outcomes, like innovation, productivity, and positive exits, is less conclusive. Second, the type of VC emerges as a key moderating factor. Private, independent VC tends to have the largest effects. GVC tends to have positive, but lower effects. However, there is evidence of complementarity between the two types of VC. Third, there is considerable evidence that financial resources provided by VC reduce financial constraints and limited, but supportive evidence that non-financial resources provided by VC reduce other barriers to growth.

These findings lend support to policies that aim to increase supply of VC to help high-tech start-ups expand. The evidence is less conclusive for the use of these policies to support innovation or productivity growth. The results for different types of VC underscore the importance of taking into account the type of VC investor and complementarities between different types of VC. For the academic literature, the review highlights literature gaps related to the effects of VC on firms in the first years of their lifecycle and in small European countries, the international dimension of VC and the evaluation of specific VC-related policies. It also suggests a need for further research to better understand how and in what settings VC increases innovation and productivity.

The study is organized as follows. Section 2 describes the effects of VC on firm performance and the challenges related to their estimation. Section 3 reports the methodology used to summarize the empirical literature. Section 4 presents the results of the review. Section 5 discusses the conclusions of the review, policy implications and avenues for future research.

2. Conceptual framework

2.1. VC and firm performance

The literature proposes three explanations for the positive relationship between VC and performance of funded firms: (i) selection of the most promising firms by VC investors (ii) provision of financial resources and (iii) provision of non-financial resources.

VC investors consider the pre-investment screening of the start-ups a key part of the investment process and allocate important resources to this activity (Gompers et al. 2020). They rely on purposefully developed screening procedures and on accumulated experience in evaluating high-tech start-ups (Gompers et al. 2020; Masucci et al. 2021; Quas, Martí, and Reverte 2021). Selection is often based on characteristics of founders and management (Bertoni, Colombo, and Grilli 2011, 2013; Colombo and Grilli 2010), innovation capabilities (Akcigit et al. 2022; Bertoni, Colombo, and Grilli 2013; Bock, Huber, and Jarchow 2018; Caselli, Gatti, and Perrini 2009; Engel and Keilbach 2007; Peneder 2010) and past performance (Akcigit et al. 2022). However, different types of VC investors give different weight to different aspects. GVC investors often emphasize social goals, like, job creation or supporting start-ups in lagging regions (Croce, Martí, and Reverte 2019). IVC investors focus on aspects related to the success of the Initial Public Offering (IPO) and Corporate VC (CVC) tend to emphasize strategic and technological fit (Chemmanur, Loutskina, and Xuan 2014; Masucci et al. 2021).

Provision of financial resources is one of the main reasons high-tech start-ups seek VC. They are often credit constrained due to their limited credit record and collateral and to the high risk of the projects they typically pursue. These projects are difficult to assess for banks and other traditional sources of external finance (Bertoni, Colombo, and Croce 2010). VC investors can overcome these challenges due to their purposely developed pre-investment screening procedures and accumulated experience with evaluating such firms and projects (Andrieu 2013; Quas, Martí, and Reverte 2021). VC investors may differ in the capital supplied and the horizon of their investment. IVC have strict contractual limits on both, while CVC can be more flexible (Chemmanur, Loutskina, and Xuan 2014). VC may also facilitate access other sources of external finance. Selection for VC investment may “certify” the start-ups’ potential for investors with limited experience in evaluating them (Bertoni, Croce, and Guerini 2015; Guerini and Quas 2016; Manigart and Wright 2013; Tykvová 2018b).

VC investors often provide non-financial resources, like business know-how and management expertise. Start-ups often lack this type of expertise, but is essential for a successful scale-up as expansion often involves an increase in the complexity of business processes, which may require changes in internal procedures. VC investors provide this expertise by participating directly in the management of the firm, by providing advice and by connecting the firm with potential clients, suppliers, and other investors (Davila, Foster, and Gupta 2003; Duruflé, Hellmann, and Wilson 2017). VC investors differ the most in the non-financial resources provided. GVC provide less business know-how due to more limited business experience (Bertoni and Tykvová 2015). IVC tend to provide mainly business know-how, monitoring and active participation in the firm management and CVC tend to provide expertise and resources linked to the parent corporation (Chemmanur, Loutskina, and Xuan 2014).

2.2. Methodological challenges

For policy, the relevant effects are those that occur post-investment due to the provision of financial and non-financial resources, as they may justify interventions to support VC. They are called “value-added” effects, as they reduce constraints on firm growth and would not occur in the absence of VC. In contrast, the “selection” effect is due to the selection of the most promising firms into VC and could occur even in the absence of VC (Bertoni, Colombo, and Grilli 2011; Croce, Martí, and Murtinu 2013; Tykvová 2018b).

The distinction between selection and value-added effects is important for the estimation of VC effects for two reasons. First, the selection effect is, likely, sizable in the case of VC, as VC investors are highly effective in pre-investment screening and selecting the best performing firms (Gompers et al. 2020). Second, the determinants of selection into VC, like, the human capital of founders and management (Bertoni, Colombo, and Grilli 2011, 2013; Colombo and Grilli 2010), innovation capabilities (Akcigit et al. 2022; Bertoni, Colombo, and Grilli 2013) and past performance (Akcigit et al. 2022) are also key determinants of firm performance (Akcigit et al. 2022; Coad and Hölzl 2012). Thus, not accounting for the selection into VC, may bias the estimated effects of VC. This reduces the comparability between estimates that account for selection and those that do not (Lohwasser 2020; Rosenbusch, Brinckmann, and Müller 2013; Tykvová 2018b).

Disentangling these two effects empirically is challenging. It requires finding credible counterfactuals for the firms that received VC that indicate what would have happened to them in the absence of the VC investment. Few studies find such counterfactuals, like firms rejected in late stages of VC selection process or firms benefiting from similar types of finance (Bronzini, Caramellino, and Magri 2020; Quas, Martí, and Reverte 2021). In the absence of such counterfactuals, most studies establish counterfactuals by matching VC-backed firms to non-funded ones based on their observable characteristics. These methods control for observable characteristics of the firms, but unobserved characteristics can still bias the estimates. To address this possible source of bias, recent studies combine matching with methods that account for firm unobserved characteristics, like, difference-in-differences.

To ensure comparability and policy relevancy of the results, we focus on the studies that explicitly control for selection into VC, at least based on observed characteristics.

3. Methodology

The results are summarized using systematic literature review, which is preferred to meta-analysis due to the heterogeneity in the methods and in the measures of firm performance used in the literature. These measures may reflect different aspects of firm performance and are not directly comparable.

The search protocol reflects the focus on the value-added effects of VC, on firm performance and on Europe. It follows the methodology used in similar reviews (Dvouletý, Srhoj, and Pantea 2022; Kersten et al. 2017). It has five parts and each part contains keywords related to: (1) VC finance/investment, (2) Europe, the European Union, and countries in Europe, (3) firm performance measures, (4) the characteristics of the funded firms and (5) econometric methods that account for selection into VC. The exact search code is reported in Appendix. The search was restricted to studies published between 2000 and 2023 in peer-reviewed journals. The time restriction is motivated by the focus on recent studies, whose results are informative for current policies. We include only published studies because they underwent a peer-review process and their results can be considered final.

Since the focus is on published studies, the code was applied to the two most used academic databases in Social Sciences: Web of Science (Clarivate Analytics 2023) and Scopus (Elsevier 2023). The initial search took place between 28 February 2020 and 6 March 2020. After merging the two searches and removing duplicates, we obtained a total of unique 115 articles. The compliance of the articles with the objectives of the review regarding estimated effects, the geographical scope and the methods used was assessed independently by the authors using a scale from 1 (least relevant) to 5 (most relevant) based on the abstracts of the studies. Those below the mean value (2.3) were considered of low relevance and excluded, similar to Dvouletý, Srhoj, and Pantea (2022). For the articles with a score above the mean, we also reviewed the references and the journals in which they were published to find additional relevant studies.⁶ Three additional studies were suggested by anonymous referees. Articles in this sample were examined in detail, based on the full text of the articles. The main reasons for exclusion at this stage were: (i) the econometric methods used did not account for selection into VC, (ii) the data used was not at firm level and (iii) the data included firms in countries outside Europe. In June 2023, this search was repeated to include newer studies.

The final sample contains 34 studies, which is comparable to similar reviews (Dvouletý, Srhoj, and Pantea 2022; Srhoj, Vitezić, and Wagner 2023). For these studies, we extracted information on: authors, year of publication, VC indicators used, the country, period and the characteristics of the sample analyzed, outcome variables, econometric methods used and the main findings. The main results are classified into positive, negative and insignificant, which allows us to compare results of different studies and evaluate the preponderance of evidence. This information is reported in Table 1.

Table 1. Review of empirical studies on the effect of VC on performance of funded firms.

Study	Country/region	Period	Treatment	Sample	Outcome variables	Empirical Methods	Main results
Bottazzi and Da Rin (2002)	Europe	1997–2000	VC	511 firms listed on Euro.nm: 292 VC-backed and 219 non-VC-backed	Sales and employment growth	Matching and DiD	Positive effects on sales and employment
Engel and Keilbach (2007)	DE	1995–1998	VC	142 VC-backed and 21,375 non-VC-backed young firms	Patent applications, probability of applying for a patent, employment growth.	Matching (PSM)	Insignificant effects on patent applications and probability of applying for a patent, positive effects on employment growth
Bottazzi, Da Rin and Hellmann (2008)	AT, BE, DK, FI, FR, DE, GR, IE, IT, LU, NL, NO, PT, ES, SE, CH, UK	1998–2005	VC, IVC and CVC	1652 firms, members of EVCA	Positive exit (IPO or acquisitions)	IV	Positive effects, contingent on VC involvement in management, which is more prevalent for IVC.
Caselli et al. (2009)	IT	1995–2004	VC	37 VC-backed and 37 non-VC-backed firms	Patents, sales growth	Matching (PSM)	Insignificant effects on patents, positive effects on sales growth
Colombo and Grilli (2010)	IT	2000–2004	VC	439 new technology-based firms: 46 VC-backed and 393 non-VC-backed.	Employment and sales growth	Heckman procedure, switching regression, IV	Positive effects on employment and sales growth
Bertoni et al. (2010)	IT	1994–2003	IVC and CVC	New technology-based firms: 52 VC-backed (19 IVC, 33 CVC) and 327 non-VC backed.	Investment and sensitivity of investment to cashflow	GMM	Positive effects for CVC and IVC on investment. Only IVC reduces the sensitivity of investment to cashflow.
Peneder (2010)	AT	2002–2005	VC	209 young firms: VC-backed and non-VC-backed	Employment and sales growth, share of sales from innovation in total sales	Matching (PSM)	Positive effects on employment and sales growth, insignificant effects on share of sales from innovation in total sales
Bertoni et al. (2011)	IT	1994–2003	VC	538 new technology-based firms (58 VC-backed).	Employment and sales growth	GMM, Heckman procedure	Positive effects on employment and sales growth.
Arqué-Castells (2012)	ES	1999–2008	VC	Firms focused on innovation: 233 VC-backed and 91,381 non-VC-backed.	Probability of patenting, number of patents	DiD, probit, Poisson	Positive effects on probability of patenting and on number of patents.
Bertoni, Colombo and Grilli (2013)	IT	1994–2003	IVC and CVC	531 new technology-based firms: 23 IVC-backed, 24 CVC-backed and 484 non-VC-backed.	Employment and sales growth	GMM	Positive effects for IVC and CVC on employment and sales growth.
Bertoni et al. (2013)	ES	1995–2004	VC and PE	324 firms in low/medium tech manufacturing and services: 246 VC-backed and 78-PE backed firms.	Investment, sensitivity to internal cashflow	FE, GMM	Negative effects of VC on the sensitivity of investment to internal cashflow. PE has the opposite effect.
Croce et al. (2013)	BE, FI, FR, IT, ES, UK	1984–2010	VC	696 young firms in hightech sectors: 267 VC-backed firms and 429 non-VC-backed	Productivity	Matching (PSM), FE, GMM	Positive effects on TFP and capital productivity, insignificant effects on labor productivity.
Grilli and Murtinu (2014)	BE, FI, FR, DE, IT, ES, UK	1993–2010	IVC, GVC and syndicates of the two types of VC.	538 IVC-backed, 239- GVC-backed, 126 backed by syndicates of both types of VC, and the same numbers of non-VC backed firms.	Employment and sales growth	Matching (PSM), FE, GMM	Positive effects for IVC and IVC-led syndicates on sales growth. Insignificant effects for GVC on sales growth, insignificant effects on employment for IVC and GVC.
Engel and Stiebale (2014)	FR, UK	2000–2007	PE	18085 firms in France (of which 2797 PE-backed) in France and 7543 firms in the UK (of which 2239 PE-backed).	Investment, sensitivity to internal cashflow	GMM	Positive effects on investment and negative effect on the sensitivity to internal cashflow.
Colombo, Piva and Rossi-Lamastra (2014)	BE, FI, FR, DE, IT, ES, UK	1994–2004	VC, IVC, CVC and GVC	New technology-based firms: 3172 non-VC, 258 VC-backed (138 IVC and 78 CVC/GVC)	Sensitivity of employment to changes in sales	Matching (PSM)	IVC increases the sensitivity of employment to changes in sales, while CVC/GVC have the opposite effects.
Alperovych et al. (2015)	BE	1998–2007	VC, GVC and PVC	Young firms: 515 VC-backed (272 PVC-backed and 243 GVC- backed) and 515 non-VC-backed.	Productivity	Matching (PSM)	Negative effects for VC, driven by the negative effects for GVC.
Croce et al. (2015)	BE, FI, FR, IT, ES, UK	1994–2004	BVC and IVC	High tech firms: 51 BVC-backed matched to 332 non-VC-backed firms and 77 IVC-backed firms matched to 402 non-VC-backed firms.	Default risk, debt/sales	Matching (PSM)	Insignificant effects of BVC and IVC on default probability and positive effects of BVC on debt exposure.
Bertoni and Tykvová (2015)	BE, FI, FR, DE, IT, ES, UK	1984–2010	GVC, IVC and syndicates of the two types of VC.	Young, biotech start-ups: 665 (125 VC-backed and 540 non-VC-backed).	Patents, citation-weighted patents	FE, GMM	Insignificant effects for GVC, positive effects for IVC and syndicates of GVC and IVC.
Bertoni et al. (2015)	BE, FI, FR, IT, ES, UK	1995–2004	VC	Young high-tech firms: 128 VC-backed and 233 non-VC-backed.	Investment and sensitivity to internal cashflow	Matching (PSM), GMM.	Negative effects on sensitivity of investment to internal cashflow, especially for follow-on investment rounds.
Guerini and Quas (2016)	BE, FI, FR, DE, IT, ES, UK	1993–2010	GVC	High-tech entrepreneurial firms: 183 GVC-backed and 8094 non-GVC backed.	Attracting PVC and positive exit (IPO or acquisition)	Matching (PSM), Heckman procedure, switching regression	Positive effect on receiving PVC and on positive exit, contingent on the firm also receiving PVC
Colombo et al. (2016)	BE, FI, FR, DE, IT, ES, UK	1995–2010	VC, IVC, CVC, BVC, and GVC	8346 new technology-based firms in high tech sectors (of which 758 VC-backed)	EU-funded R&D partnerships	Matching (PSM), RE, Heckman procedure.	Positive effects for VC, BVC, GVC and IVC and insignificant effects for CVC.
Croce and Martí (2016)	ES	1995–2006	PE	615 family firms (257 PE-backed and 358 non-PE backed) and 1315 non-family firms (728 of PE-backed and 587 non-PE-backed)	Productivity growth	Matching (PSM), RE and GMM	Positive effects for family firms.
Cumming et al. (2017)	BE, FI, FR, DE, IT, ES, UK	1991–2010	IVC, GVC and syndicates of IVC and GVC	8370 young, high-tech firms (420 IVC-backed, 127-GVC backed and 62 backed by syndicates of GVC and IVC investors)	Positive exit (IPO or trade sale)	Multinominal logit model, matching (PSM), IV	Positive effects for IVC and syndicates of IVC and GVC, insignificant effects for GVC.
Colombo and Murtinu (2017)	BE, FI, FR, IT, ES, UK	1992–2010	IVC, CVC and syndicates of the two types of VC	High-tech entrepreneurial firms: 759-VC backed and 7611 non-VC backed	Productivity, sales, fixed assets, payroll and employment	FE, IV, GMM	Positive effects on sales and productivity for both types of VC, but higher effect for IVC in the short run. Positive effects for IVC on employment and payroll and insignificant effects of CVC on these outcomes. Insignificant effects of both types of VC on assets.
Bock et al. (2018)	DE	1997–2012	VC	31 VC-backed firms and 67 non-VC-backed firms	Employment and sales growth	IV, Heckman procedure.	Insignificant effects on employment and sales growth
Croce et al. (2019)	ES	2005–2013	GVC and PVC	Young firms: 173 GVC-backed, 211 PVC-backed and 888 non-VC-backed.	Employment growth	Matching (PSM), IV, GMM	Positive effects for PVC and GVC, the magnitudes of the two effects depend on the phase of the business cycle.
Corsi and Prencipe (2019)	BE, DK, FI, FR, DE, IT, LU, NO, PT, ES, SE, UK	2009–2013	VC/PE	Young SMEs in medium-high and high-tech manufacturing: 223 firms non-VC/PE-backed and 103 VC/PE- backed.	Patents	Poisson quasi likelihood, IV	Positive effects of VC/PE.
Antoni, Maug, and Obernberger (2019)	DE	2002–2008	PE	2420 PE-backed and 24,147 non-PE-backed firms	Employment	Matching, DiD	Negative effects, linked to organization streamlining.
Bronzini, Caramellino, and Magri (2020)	IT	2004–2014	VC, IVC and CVC	Start-ups who sought VC: 101 VC-backed and 258 non-VC-backed, rejected in late stages of VC screening.	Employment, labor costs, assets, sales, wage, probability of patent application, number of patent applications, equity and debt, survival	DiD	Positive effects on employment, labor costs, assets, innovation, equity and debt, insignificant effects on sales, survival and wages. The positive effects on size and innovation are driven by IVC.
Quas et al. (2021)	ES	2005–2013	VC	Young SMEs: 915 VC-backed, 1551 non-VC-backed, who received participative loans from public sources	Employment, sales, assets growth	Matching (PSM, CEM), DiD, GMM	Positive effects on sales, assets and employment growth. Larger effects for more experienced VC investors.
Sardo et al. (2020)	Western Europe	2006–2015	VC	1161 small firms in knowledge intensive services	Debt/assets	GMM	VC reduces dependence on debt and lowers adjustment costs.
Murtinu (2021)	BE, FI, FR, IT, ES, UK	1994–2004	GVC	166 GVC-backed firms and 1122 non-GVC- backed firms	Productivity	Matching (PSM), DiD and IV	Positive effect, working through increased responsiveness to tax policy changes.
Huergo and López (2022)	ES	2001–2018	GVC	Newly created firms: 129 GVC backed and 143 non-GVC- backed	Employment, sales and labor productivity	Matching (PSM)	Positive impact of GVC on sales and labor productivity growth. Inconclusive results for employment growth.
Balz, Brinkmann and Kanbach (2023)	Europe (VICO)	2005–2018	CVC and IVC, different types of CVCs	86 CVC-backed and 297 IVC-backed firms	Productivity	switching regression	Larger positive effects for IVC. Large heterogeneity of effects among different types of CVCs.

Studies are reported in chronological order. The abbreviations in the table stand for: VC – venture capital, BVC -bank venture capital, CVC – corporate or captive venture capital, GVC – government venture capital, IVC – independent venture capital and PVC- private venture capital, PSM – propensity score matching, GMM – generalized method of moments, FE – fixed effects regressions, TFP – total factor productivity, IV – instrumental variables, DiD – difference-in-differences, CEM – coarsened exact matching.

4. Results of the systematic literature review

4.1. The coverage of the review

Figure 2 shows the number of articles by the year of publication. Most articles were published after 2010. Likely, this reflects improved data availability on firm performance and VC and increased use of methods that control for selection into VC in recent years (Tykvová 2018b).

Figure 2. Sample decription: number of articles by year of publication.

Note: Own calculations

Given the interdisciplinary scope of the topic, the studies are published in entrepreneurship, innovation, finance, and management journals. Most articles were published in: Small Business Economics (six), Journal of Business Venturing (five), Research Policy (five), Journal of Financial Economics (two), Journal of Corporate Finance (two), European Financial Management (two). One article was published in each of the following journals: Economic Policy, Economics of Innovation and New Technology, Entrepreneurship Theory and Practice, Industry and Innovation, Journal of Business Venturing Insights, Journal of Economics & Management Strategy, Journal of Empirical Finance, Journal of Technology Transfer, Managerial and Decision Economics, Strategic Entrepreneurship Journal, Structural Change and Economic Dynamics and Venture Capital. Figure 3 shows this distribution.

Figure 3. Sample description: number of articles by journal of publication.

Own calculations.

The geographical coverage of the selected articles is unbalanced. The countries covered are: Italy and Spain (nineteen studies), Belgium, France and United Kingdom (fourteen studies), Finland (thirteen studies), Germany (eleven studies), Austria, Denmark, Luxembourg, Norway, Portugal and Sweden (two studies) and Greece, Ireland, Netherlands and Switzerland (one study). In addition, three studies mention only that they cover European or Western European countries, without specifying the countries covered (Balz, Brinkmann, and Kanbach 2023; Botazzi and Da Rin 2002; Sardo, Serrasqueiro, and Félix 2020). This unbalanced coverage, likely, reflects the concentration of VC activity in Europe (Nepelski, Piroli, and De Prato 2016) and data availability.⁷ It shows a lack of the studies on smaller EU countries and on Central and Eastern European (CEE) countries. These countries tend to have thinner VC markets, which may lead to different VC effects (Bertoni, D’Adda, and Grilli 2016). Moreover, Figure 1 shows that VC investment has increased rapidly in some of these countries, like, Estonia and Ireland, and documenting their experience could be relevant for VC-related policies in other small European countries.

Most of the studies use samples which are exclusively or predominantly composed of firms in high-tech/knowledge intensive sectors. Only one study (Bertoni, Ferrer, and Martí 2013) focuses on low-tech sectors. Even those that cover all sectors focus on innovative firms, defined as “research-based firms” or “new technology-based firms”. This coverage is consistent with the sectoral distribution of VC in the EU (Nepelski, Piroli, and De Prato 2016).

Most studies indicate a focus on young firms, but the definition of “young” varies between newly created (Huergo and López 2022) to less than 25 years old (Corsi and Prencipe 2019). Only two studies examine very young firms: Huergo and López (2022) examine newly created firms and Croce, Martí, and Reverte (2019) examine firms that are younger than 5 years. Thus, few studies cover firms in the first years of their lifecycle. Likely, this is due to poor data coverage of these firms in the commercial databases and to the econometric methods used, which require data on firm performance for several years before receiving VC. This is a notable gap, as the effects of VC may differ by age (Rosenbusch, Brinckmann, and Müller 2013) and very young firms account for a third of VC-backed firms in the EU (Nepelski, Piroli, and De Prato 2016).

Overall, the coverage of the literature mirrors the concentration of VC activity in Europe, at sectoral and geographical level, but not the firm demographics distribution. Therefore, the findings of the review should be considered representative, mainly, for high-tech start-ups in large EU economies, after the early stages of their lifecycle.

4.2. Firm performance measures, VC treatment and econometric methods

The most studied measure of firm performance is expansion, measured as employment (fifteen studies), sales/revenue growth (twelve studies) and assets growth (three studies). Other extensively examined outcomes are: innovation (eight studies), productivity (seven studies), investment and sensitivity to internal cashflow (five studies), access to other sources of external finance (six studies) and exits (five studies).⁸ Table 1 shows a trend among the most recent articles to examine the effects on several aspects of firm performance.

The most used measure of VC is a dummy variable that indicates whether a firm received VC (eighteen studies). Among these, twelve report only estimates for this indicator, while six also report the effects of subtypes of VC. Three studies (Antoni, Maug, and Obernberger 2019; Croce and Martí 2016; Engel and Stiebale 2014) use similar indicators for Private Equity (PE)⁹, one study (Corsi and Prencipe 2019) uses an indicator for receiving either VC or PE and another study (Bertoni, Ferrer, and Martí 2013) examines the differences between VC and PE. Ten studies examine GVC, either exclusively (Guerini and Quas 2016; Huergo and López 2022; Murtinu 2021), or in comparison with PVC (Alperovych, Hübner, and Lobet 2015; Bertoni and Tykvová 2015; Colombo, D’Adda, and Pirelli 2016; Colombo, Piva, and Rossi-Lamastra 2014; Croce, Martí, and Reverte 2019; Grilli and Murtinu 2014). Eight studies distinguish between IVC and CVC (Balz, Brinkmann, and Kanbach 2023; Bertoni, Colombo, and Grilli 2013; Bottazzi, da Rin, and Hellmann 2008; Bronzini, Caramellino, and Magri 2020; Colombo, D’Adda, and Pirelli 2016, Colombo and Murtinu 2017; Colombo, Piva, and Rossi-Lamastra 2014; Croce, D’Adda, and Ughetto 2015). Two studies focus BVC (Colombo, D’Adda, and Pirelli 2016; Croce, D’Adda, and Ughetto 2015).

Given the focus on value-added effects, all studies use methods that control for selection into VC. The most used econometric methods are as follows: matching (nineteen studies), generalized methods of moments (thirteen studies), instrumental variables (eight studies), difference-in-differences (six studies), fixed-effects regression (five studies) and Heckman procedure (five studies). Earlier studies tend to use methods that account only for selection based on observed characteristics, like matching, while more recent studies combine these methods with methods that account for unobserved firm characteristics, like, difference-in-differences and generalized methods of moments.

4.3. Review of the results of the effects of VC

4.3.1. Firm expansion

The most documented effect of VC is on firm expansion. Employment/payroll growth is the most used measure of firm expansion. While employment is not the main objective for VC investors, its widespread use is due to its reliability as an indicator of firm growth (Coad and Hölzl 2012) and its policy importance (Croce, Martí, and Reverte 2019). Fifteen studies use this measure (Antoni, Maug, and Obernberger 2019; Bertoni, Colombo, and Grilli 2011; Bertoni, Colombo, and Grilli 2013; Bock, Huber, and Jarchow 2018; Bottazzi and Da Rin 2002; Bronzini, Caramellino, and Magri 2020; Colombo and Grilli 2010; Colombo and Murtinu 2017; Colombo, Piva, and Rossi-Lamastra 2014; Croce, Martí, and Reverte 2019; Engel and Keilbach 2007; Grilli and Murtinu 2014; Huergo and López 2022; Peneder 2010; Quas, Martí, and Reverte 2021). Nine studies find positive effects of VC on employment/payroll, three studies find insignificant effects (Bock, Huber, and Jarchow 2018; Grilli and Murtinu 2014; Huergo and López 2022), one study (Antoni, Maug, and Obernberger 2019) finds negative effects, linked to organization streamlining and one (Colombo and Murtinu 2017) finds that the effects depend on the type of VC. In a similar vein, Colombo, Piva, and Rossi-Lamastra (2014) find that IVC increases sensitivity of employment to changes in sales, while GVC and CVC decrease it, likely reflecting the importance of efficiency and non-financial objectives for different VC investors.

The second most used measure of firm expansion is sales/revenue growth. Eight studies (Bertoni, Colombo, and Grilli 2011, 2013; Bottazzi and Da Rin 2002; Caselli, Gatti, and Perrini 2009; Colombo and Murtinu 2017; Grilli and Murtinu 2014; Huergo and López 2022; Peneder 2010; Quas, Martí, and Reverte 2021) find positive effects of VC on sales/revenue growth, three (Bock, Huber, and Jarchow 2018; Bronzini, Caramellino, and Magri 2020; Colombo and Grilli 2010) find insignificant effects and one finds positive effects for IVC, but insignificant effects for GVC. A less analyzed aspect is assets growth. Two studies find positive effects of VC on this measure (Bronzini, Caramellino, and Magri 2020; Quas, Martí, and Reverte 2021) and one finds insignificant effects (Colombo and Murtinu 2017). The effects on assets growth tend to be higher for IVC compared to CVC (Bronzini, Caramellino, and Magri 2020).

The results show a clear preponderance of positive effects of VC on firm expansion, across different measures of expansion and different types of VC, likely, reflecting the importance of firm expansion for both VC investors and funded firms (Duruflé, Hellmann, and Wilson 2017).

4.3.2. Investment, financial constraints and access to other source of external finance

The VC and PE effects on financial constraints are also well documented. Four studies (Bertoni, Colombo, and Croce 2010; Bertoni, Croce, and Guerini 2015; Bertoni, Ferrer, and Martí 2013; Engel and Stiebale 2014) find that they reduce sensitivity of investment to internal cashflow. The evidence is stronger for VC compared to PE (Bertoni, Ferrer, and Martí 2013) and for IVC compared to CVC (Bertoni, Colombo, and Croce 2010).

There is also evidence on “certification” effect of receiving VC. Several studies show that VC enables firms to access external finance from banks and increases their debt exposure (Bronzini, Caramellino, and Magri 2020; Croce, D’Adda, and Ughetto 2015), although Sardo, Serrasqueiro, and Félix (2020) find the opposite effect. There is also evidence of improving access to finance from public authorities (Colombo, D’Adda, and Pirelli 2016) and equity investors (Bottazzi, da Rin, and Hellmann 2008; Bronzini, Caramellino, and Magri 2020; Cumming, Grilli, and Murtinu 2017).

Taken together, these results show that VC reduces firms’ financial constraints both directly and indirectly by facilitating access to other sources of external finance.

4.3.3. Innovation

The VC effects on innovation is also extensively analyzed. Various measures of innovation are used: application for patents or the number of patents (Arqué-Castells 2012; Bertoni and Tykvová 2015; Bronzini, Caramellino, and Magri 2020; Caselli, Gatti, and Perrini 2009; Corsi and Prencipe 2019; Engel and Keilbach 2007), the share of innovation sales in total sales (Peneder 2010) and participation in R&D projects (Colombo, D’Adda, and Pirelli 2016). The results show mixed effects. For variables related to patents, two studies (Arqué-Castells 2012; Corsi and Prencipe 2019) find positive effects, two studies (Caselli, Gatti, and Perrini 2009; Engel and Keilbach 2007) find insignificant effects and two studies (Bertoni and Tykvová 2015; Bronzini, Caramellino, and Magri 2020) find that positive effects for IVC, but insignificant effects for GVC and CVC, respectively. The study examining the effects on sales from innovation finds insignificant effects (Peneder 2010). However, there is evidence that VC helps funded firms participate in EU-funded R&D partnerships (Colombo, D’Adda, and Pirelli 2016) and descriptive evidence that VC is linked to engaging in-house R&D (Da Rin and Penas 2017).

This review finds some evidence of positive effects of VC on innovation, but given the heterogeneity of outcomes studied and the mixed results, it is difficult to draw general conclusions. Several studies argue that the main way that the main way VC investors affect innovation is by selecting innovative firms and helping them commercialize their innovations and, thus, increasing the returns to innovation (Duruflé, Hellmann, and Wilson 2017; Engel and Keilbach 2007; Peneder 2010).

4.3.4. Productivity

Seven studies examine the VC effects on different measures of firm productivity. Four studies (Colombo and Murtinu 2017; Croce and Martí 2016; Croce, Martí, and Murtinu 2013; Huergo and López 2022; Murtinu 2021) find positive effects of VC or PE on funded firms’ productivity, while three studies (Alperovych, Hübner, and Lobet 2015; Balz, Brinkmann, and Kanbach 2023; Colombo and Murtinu 2017) find that VC effects on productivity depend on VC investor type and that the effectsy tend to be higher for PVC compared to GVC and for IVC compared to CVC. Overall, the evidence on the effects of VC on productivity is mixed and the results suggest that the effects may differ between different types of VC.

4.3.5. Successful exit

Five studies examine the effects of VC on different types of exits. The most relevant exits for VC investors are IPOs and trade sales. Three studies examine these exists. Bottazzi, Da Rin, and Hellmann (2008) find that VC has a positive effect on successful IPO, and that the effect is mediated by the active involvement of the VC investors in the management of the firm. Cumming, Colombo, and Murtinu (2017) find that IVC and syndicates of IVC and GVC have positive effects on IPO or trade sales, but that GVC alone has insignificant effects. Guerini and Quas (2016) find that GVC has a positive effect on the likelihood of receiving PVC and on the likelihood of a successful IPO or acquisition, if the firm has also received PVC. Several studies examine other exits. Croce, D’Adda, and Ughetto (2015) find that BVC and IVC have insignificant effects on the probability of default, but that BVC tend to select firms with lower default risk pre-investment. Bronzini, Caramellino, and Magri (2020) find an insignificant effect of VC on survival. Taken together, the evidence on exists is limited and heterogenous in terms of outcomes studied, making it difficult to draw conclusions on the preponderance of one type of effects.

4.4. Channels

Most studies estimate the overall effect of the VC without delving into the channels though which VC affects firm performance. Only few studies provide evidence on specific channels and they examine them separately, which prevents conclusions on their relative importance.

There is substantial evidence that financial resources provided by VC investors reduce the financial constraints of funded firms either directly (Bertoni, Colombo, and Croce 2010; Bertoni, Croce, and Guerini 2015; Bertoni, Ferrer, and Martí 2013; Croce, D’Adda, and Ughetto 2015; Engel and Stiebale 2014) or indirectly, by enabling access to external finance from banks (Bertoni, Croce, and Guerini 2015; Bronzini, Caramellino, and Magri 2020; Croce, D’Adda, and Ughetto 2015), public authorities (Colombo, D’Adda, and Pirelli 2016) or equity investors (Bottazzi, da Rin, and Hellmann 2008; Cumming; Colombo and Murtinu 2017).

Three studies examine the effects of non-financial resources provided by VC investors. Bottazzi, Da Rin, and Hellmann (2008) show that involvement of VC investors in management, recruiting and monitoring has a positive effect on the IPO success. Bronzini, Caramellino, and Magri (2020) and Quas, Martí, and Reverte (2021) show that VC-backed firms grow faster than firms receiving comparable finance from other sources, and attribute the difference to non-financial resources provided by VC investors. Additionally, Quas, Martí, and Reverte (2021) show that more experienced VC investors, who have more business expertise, tend to have larger effects on firm expansion. Overall, the evidence on these channels is limited but supportive of the hypothesis that VC investors provide important non-financial resources.

4.5. Effects of different types of VC investors

The differences between VC investors in their objectives and resources provided may influence the effects they have on funded firms. In this section, we summarize the results for specific types of VC investors.

Given the role played by the GVC in Europe, a key distinction is between GVC and PVC. GVC investors tend to focus on social goals, like, job creation or supporting entrepreneurship in lagging regions (Bertoni, Colombo, and Croce 2010; Croce, Martí, and Reverte 2019). Since they have less business experience, they provide less business know-how (Bertoni and Tykvová 2015). However, they may provide benefits related to closeness to policy makers (Murtinu 2021). Studies that focus exclusively on GVC find positive effects on sales (Huergo and López 2022) and productivity (Huergo and López 2022; Murtinu 2021). Murtinu (2021) also suggest a specific mechanism for this effect: faster reaction to tax policy changes. Several studies investigate how GVC and PVC compare and interact. They show that GVC tends to have lower effects on sales (Grilli and Murtinu 2014), innovation (Bertoni and Tykvová 2015), productivity (Alperovych, Hübner, and Lobet 2015) and successful exits (Cumming, Grilli, and Murtinu 2017). For employment, Croce, Martí, and Reverte (2019) find that the relative magnitude of the effects depends on the phase of the business cycle and Colombo, Piva, and Rossi-Lamastra (2014) find that GVC reduces sensitivity of employment to decreases in sales. Finally, there is evidence on complementarity between GVC and PVC: GVC has positive effects on attracting PVC (Guerini and Quas 2016) and syndicates of GVC and PVC tend to have the largest effects (Bertoni and Tykvová 2015; Cumming, Grilli, and Murtinu 2017).

Another important comparison is between IVC and CVC. IVC investors have strict time horizons and are motivated by financial returns from IPOs. They are more focused on firm growth and provide non-financial resources related to active participation in management and monitoring (Chemmanur, Loutskina, and Xuan 2014). CVC have longer time horizons and provide non-financial resources related to the expertise of the parent firm (Chemmanur, Loutskina, and Xuan 2014; Masucci et al. 2021). In line with these insights, this review finds evidence that IVC investors are more involved in the active management of the firm (Bottazzi, da Rin, and Hellmann 2008) and that they have larger effects on expansion (Bertoni, Colombo, and Grilli 2013; Bronzini, Caramellino, and Magri 2020; Colombo and Murtinu 2016) and successful IPOs (Bottazzi, da Rin, and Hellmann 2008). However, there is also evidence of a greater focus of IVC on efficiency, with possible negative effects on employment (Colombo, Piva, and Rossi-Lamastra 2014). Consistent with tighter time horizons, there is evidence that IVC tends to have higher effects in the short term, than CVC (Bertoni, Colombo, and Grilli 2013; Colombo and Murtinu 2016). As most studies estimate short-term effects, the larger effects of IVC, may reflect their shorter time horizons.

5. Discussion and conclusions

Economic literature and policy initiatives posit that VC can play a key role in helping innovative start-ups grow and innovate, by providing finance and business know-how. We assess this hypothesis in Europe by reviewing 34 recent firm-level studies that estimate value-added effects of VC on funded firm in Europe.

The review finds a large preponderance of positive effects of VC on firm expansion and on reducing firms’ credit constraints. The evidence on other outcomes, like innovation, productivity, and positive exits, is more limited and mixed. Likely, these findings reflect the focus of both VC investors and start-ups on expansion (Duruflé, Hellmann, and Wilson 2017) and the idiosyncratic nature of innovation and productivity growth. The results also indicate that PVC, especially IVC, tends to have the largest positive effects. GVC tends to have positive, but lower effects than PVC, and it often complements PVC.

Policy-wise these results have several implications. The large preponderance of positive effects of VC on firm expansion and investment lends support to policies that use VC to support scaling-up of high-tech start-ups. Scaling-up is a major focus of the recently adopted VC-related policies in the EU. Perhaps the most direct way to increase supply of VC is to provide GVC. However, the effects of the GVC are still debated in the literature (Bottazzi and Da Rin 2002; Quas et al. 2022) with earlier evidence suggesting limited or even negative effects (Leleux and Surlemont 2003) and more recent evidence suggesting positive effects and complementarities between PVC and GVC (Bai et al. 2021; Guerini and Quas 2016). The findings of this review support the use of GVC as a policy instrument, especially when the objective is scaling-up and leveraging other VC investments. However, the results for the PVC suggest a need for policy to go beyond the provision of GVC and to create a favorable environment for private VC. For these purposes, prior literature emphasizes the importance of further development of the IPO markets and reducing fragmentation of VC markets (Grilli, Latifi, and Mrkajic 2019; Tykvová 2018a).

Our review highlights several gaps in the empirical literature on the VC effects in Europe. The first gap relates to the geographical and firm demography coverage. More evidence is needed on the effects of VC on firms in smaller European countries and in the first years of lifecycle. The second gap relates to the effects of VC on innovation and productivity, on which evidence is limited and mixed. Since innovation and productivity growth are the ultimate drivers of economic growth, it is essential to gain a better understanding of the effects of VC on these outcomes. Related to this, more evidence is needed on the channels through which VC investments affect start-ups and their relative importance. The third gap relates to international dimension of VC, which represents a major trend in the VC activity and VC-related policies (Devigne et al. 2018), but is not analyzed by any study in the review. Finally, more evidence is needed on specific VC-related policies. Policy debates often focus on advantages and disadvantages of specific policy interventions, such as different way of supplying GVC (Alperovych, Quas, and Standaert 2018) or tax treatment of specific investments, while the academic research has concentrated, by and large, on measuring the average the effects of receiving VC. To bridge this gap, more research is needed on the effectiveness, advantages, and disadvantages of specific VC-related policies.

Acknowledgments

The authors thank Laure de Batz, Mircea Epure and Ayaz Zeynalov, two anonymous referees, and the editor for helpful comments and suggestions and Natalija Neskoroďana, Marek Pokorný and Zuzana Vrbková for excellent research assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work on this study was supported by the Internal Grant Agency of the Faculty of International Relations, Prague University of Economics and Business under the grant [F2/42/2020].

Notes

1. European Innovation Council has a budget of €10 billion, InvestEU has a budget of €26 billion and aims to mobilize €372 billion in additional investment.

2. EU Unicorns Group proposed the creation of a €100 billion EU sovereign tech fund to provide expansion investments for innovative start-ups in the EU (Quas et al. 2022).

3. It accounts for only 15% of the global VC, while the US accounts for 70% (Nepelski, Piroli, and De Prato 2016). At national level, Figure 1 shows that VC investment accounts for less 0.15% of the GDP in all EU countries compared to 0.63% in the US.

4. They may also lead to promising start-ups seeking finance in the US. A recent high profile case includes the Romanian founded unicorn, UiPath, (https://www.ft.com/content/05fb8ebd-26b1-48ca-be29-1ede7381da3b)

5. We focus on stand-alone GVC. Governments may also provide funds also as limited partners and this option is widely used in Europe (Alperovych, Quas, and Standaert 2018), but due to the limited number of studies estimating its effects, we cannot cover it in this review. We thank an anonymous referee for this observation.

6. The search includes also studies that were published only online at the time of the search.

7. Most studies are based on VICO dataset, which covers only seven EU member states (Belgium, Finland, France, Germany, Italy, Spain and United Kingdom). Additionally, countries with high quality national level datasets, like, Italy and Spain are more often studied.

8. We do not review measures of firm performance that are covered by only one study.

9. Studies found by the search which examined the effects of PE were considered relevant and were included in the review, in line with Bottazzi and Da Rin (2002) and Tykvová (2018a).

Appendix

Search code

1) Treatment (abstract/title/keywords):

“Venture capital” OR “VC investment*” OR “VC investor” OR “VC fund*” OR “VC financ*” OR “VC*backed” OR “seed funding” OR “initial VC” OR “funding round*” OR IVC OR GVC “government* venture capital” OR IVC OR “independent venture capital” OR CVC OR

“corporate venture capital” OR “syndicated investments”

2) EU MEMBER STATES (28 countries)

AND (anywhere)

Austria* OR Belgium* OR Bulgaria* OR Croatia* OR Cyprus* OR Czech* OR Czech Republic* OR Denmark* OR Estonia* OR Finland* OR France* OR Germany* OR Greece* OR Hungary* OR Italy* OR Ireland* Latvia* OR Lithuania* OR Luxembourg* OR Malta* OR Netherlands* OR Holland* OR Poland* OR Portugal* OR Romania* OR Slovakia* OR Slovenia* OR Spain* OR Sweden* OR United Kingdom* OR “European Union” OR “EU” OR “Europe” OR “European”

3) FIRM CHARACTERISTICS AND (anywhere)

“start up” OR “start-up” OR “startup” OR “young firm” OR “new firm” OR “innovative firm” OR “high tech” OR “high-tech” OR “IT” OR “ICT” OR “tech” OR “new technology-based firm” OR “NTBF” OR “knowledge intensive” OR “innovation”

4) OUTCOME VARIABLES:

AND (anywhere)

“firm growth” OR “firm performance” OR “economic performance” OR revenue* OR turnover OR sales OR “value added” OR “value-added” OR employment OR employee* OR productivity OR “scale-up” OR “scaleup” OR “scaling up” OR assets

5) METHODS

AND (anywhere)

“counterfactual evaluation” OR “treatment effect” OR “causal effect” OR “propensity score”

OR “matching” OR “regression discontinuity” OR “dif-in-dif” OR “difference-in-differences”

OR “difference in differences” OR “instrumental variable*” OR “identification strategy” OR

“GMM” OR “Generalized Method of Moments”