Foreign Investors & Domestic Company Growth: Evidence from US Venture Capital Investments in Sweden

There is a long-standing debate about the advantages and disadvantages of foreign investors. Advocates emphasize the benefits of foreign capital, expertise and networks, whereas critics worry about hollowing out domestic economic activities. This debate also pertains to the financing of start-ups by venture capitalists (VCs). The US is the epicentre of the venture capital (VC) industry, and in recent decades US VCs have taken an increased interested in global investment opportunities (Aizenman and Kendall, 2012). Policy makers in many countries therefore face the question of whether or not to encourage foreign VC investments, especially from the US (Bradley et al., 2011).

Of primary concern to most policy makers is the question of employment creation, which is typically the main reason of looking at VC in the first place (Davis et al., 2014; Samila and Sorenson, 2011). In addition, there is also an interest in business activity, typically measured through sales growth. Finally, there is a concern about exit, especially whether acquirers are domestic or foreign. All these policy issues ultimately come down to some factual questions: whether foreign VCs have a different impact than domestic VCs? In this paper we set out to provide empirical evidence about the differential effects of foreign VC investments. We specifically evaluate the main criticism about foreign VCs, that they care less about the domestic growth of the companies they invest in.

To focus our research question, we consider the investments of US VCs in Sweden. We focus on US investors, because the US has the largest and most mature VC market and is the most powerful source of foreign VC investments. We have several reasons for looking at Sweden. First, it is one of the most developed VC markets, always ranked among the top 10 countries in terms of VC to GDP ratio (OECD, 2017). We are therefore dealing with an institutional environment that is mature and has credible domestic VCs as an alternative to US VCs. Second, Sweden has high-quality data, including detailed data on employment and sales of all private companies. We are therefore able to measure the effect of US VCs with great precision. Lerner and Tåg (2013) provide a detailed description of institutional details of the Swedish VC market.

It should be mentioned upfront that our analysis only looks at the activities of Swedish start- up companies in Sweden. Our data sources cannot measure their activities outside of Sweden, so we cannot observe employment creation in the US (or elsewhere outside Sweden). We may well be underestimating the total growth effect of US VCs on Swedish companies. However, our interest here is specifically on the domestic growth of Swedish start-ups, which is the central concern for policy makers, and which our data covers accurately. Note also that our data contains one indirect measure of moving activity abroad, namely whether Swedish companies establish foreign subsidiaries or not.

All employment and sales data come from the Swedish Companies Registrations Office. The VC data on investments and exits come from Thompson One. Our sample covers the period 1998 to 2012. The main regression models consider the effect of US VC investments on company growth (such as employment and sales) over the subsequent five years. The analysis naturally controls for other factors, such as company industry, location, and stage, as well as calendar time and economic cycles. We also consider the effect of US VC investments on subsequent fundraising and exit.

Endogeneity is a central concern, as the investments of US VCs should not be treated as exogenous. Our analysis combines two well-established instrumental variable approaches. First, we consider supply shocks to the VC industry in the US. From the perspective of Sweden these are exogenous, as investment opportunities in Sweden are unlikely to drive US VC fundraising. Moreover, we control for US and Swedish GDP growth, so as to account for relevant macro-economic shocks. This approach is similar to Nanda and Rhodes-Kropf (2013). Second, we consider local market conditions, using the instrumental variable methods of Ackerberg and Botticini (2002). This approach leverages VC supply shocks to local markets, defined by industries and local geographies. Our instruments are highly significant in the first- stage regression that predict the presence of US VCs and pass all the standard specification tests.

Our main results are as follows. We find a strong positive employment effect for companies backed by US VCs, both in the simple OLS and in the instrumented 2SLS regressions. The effects are economically large and increase over time. We obtain a similar set of results for company sales. There is evidence for an increase in profitability and the presence of foreign subsidiaries in the long-run. We also find receiving funding from a US VC increases the likelihood of raising additional rounds of financing. However, we do not find any statistically significant effect of US VC on the probability of exit. Most surprising, we do not find that having US VCs increases the likelihood of US exits, such as getting acquired by a US company or listing on a US stock exchange. We perform a variety of robustness checks and find that our core results are very stable.

References

Ackerberg, D. A. & Botticini, M. (2002) 'Endogenous Matching and the Empirical Determinants of Contract Form'. Journal of Political Economy, 110(3), pp. 564–591.
Aizenman, J. & Kendall, J. (2012) 'The internationalization of venture capital'. Journal of Economic Studies; Glasgow, 39(5), pp. 488–511.
Bradley, D. et al. (2011) ‘Currying Favor with Top Venture Capital Firms: The Role of IPO Underpricing and All-Star Coverage’. Unpublished Working Paper.
Davis, S. J. et al. (2014) 'Private Equity, Jobs, and Productivity'. The American Economic Review, 104(12), pp. 3956–3990.
Lerner, J. and Tåg, J. 2013. ‘Institutions and Venture Capital’. Industrial and Corporate Change, 22:153–182.
Nanda, R. & Rhodes-Kropf, M. (2013) 'Investment cycles and startup innovation'. Journal of Financial Economics, 110(2), pp. 403–418.
OECD (2017) Entrepreneurship at a Glance 2017. See Figure 7.1. ‘Venture capital investments as a percentage of GDP’.
Samila, S. & Sorenson, O. (2011) 'Venture Capital, Entrepreneurship, and Economic Growth'. The Review of Economics and Statistics, 93(1), pp. 338–349.