Venture capital, innovation activities, and economic growth: are feedback effects at work?

Abstract

This study makes an attempt to assess the relationship between venture capital, innovation activities and per capita economic growth in the European Economic Area (EEA) countries between 1989 and 2014. We use three indicators of venture capital and eight activities of innovation to examine this assessment. The study principally highlights whether Granger causality runs between these variables both ways, one way, or not at all. Using a panel vector error correction model (VECM), we find that both venture capital and innovation activities contribute to long-run per capita economic growth. These results provide imperative policy implications for these EEA countries.

Keywords:

• Venture capital
• innovation activities
• per capita economic growth
• Granger causality
• VECM
• EEA countries

JEL Classifications:

• O43
• O16
• E44
• E31

Acknowledgements

A part of this paper was presented at the Australia–Middle East Conference on Business and Social Sciences, Dubai, 17–18 April 2016, and adjudged the Best Paper. We thank the conference organizers for this appreciation. The authors also thank the editor and reviewers of this journal for their insightful comments, which have improved the quality of this paper.

Additionally, one of the authors (Rana P. Maradana) acknowledges financial support from UGC through UGC (NET).

Notes

1. Between venture capital and economic growth (see, inter alia, Carvell, Kim, Ma, & Ukhov, 2013; Füss & Schweizer, 2012; Pradhan et al., 2016b; Zhang, Zhnag, Wang, & Huang, 2013), between venture capital and innovation (see, inter alia, Hirukawa & Ueda, 2011; Lahr & Mina, 2016; Obrimah, 2015), and between innovation and economic growth (see, inter alia, Cameron, 1998; Fan, 2011; Galindo & Méndez, 2014; Pradhan et al., 2016).

2. The EEA unites the European Union Member States and the three EEA EFTA (European Free Trade Area) States, namely, Iceland, Liechtenstein and Norway. The total listed countries in this group number 31 and they are Austria, Belgium, Bulgaria, Croatia, Republic of Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Liechtenstein, Lithuania, Luxembourg, Malta, the Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, and the United Kingdom.

3. This sample is chosen depending upon the data availability for our study objectives.

4. These are PATR, PATN, RDEX, RRDA, HTEX and STJA.

5. The procedural details of having CIIN are discussed by Kaneva and Untura (forthcoming), Mirshojaeian Hosseini and Kaneko (2012), Menyah, Nazlioglu, and Wolde-Rufael (2014) and Pradhan, Arvin, Norman, and Hall (2014). Appendix B describes how we use an approach called principal component analysis to arrive at CIIN.

6. Variation is with respect to time series (1989–2014) only, while country inclusion (19) remains the same in each specification.

7. It is to observe whether a time series variable is non-stationary and possesses a unit root (see, inter alia, Bierens, 2001; Bhargava, 1986).

8. It is to observe the relationship between non-stationary time series variables. If two or more time series variables all have a unit root, i.e. I (1), but a linear combination of them is stationary, I (0), then the series is said to be cointegrated (see, inter alia, Engle & Granger, 1987).

9. FMOLS is a non-parametric approach, taking into account the feasible correlation between the error term and the first differences of the regressor as well as the presence of a constant term to deal with corrections for serial correlation (see, inter alia, Maeso-Fernandez, Osbat, & Schnatz, 2006; Pedroni, 2001).

10. The reported unit root test results are with respect to the BR test at the first difference level. The other test results have not been included due to space constraints.

11. The reported cointegration test results are with respect to only the KR cointegration test results are with respect to only the KR cointegration test because of space restrictions.

12. DOLS is a parametric approach, which adjusts the errors by augmenting the static regression with leads, lags and contemporaneous values of the regressor in first difference (see, inter alia, Kao and Chiang, 2000).