How does entrepreneurial activity affect the supply of informal investors?

Abstract

This paper examines the prevalence and the determinants of informal entrepreneurial investment activity (including investors in firms of family and friends, and business angels), using a dataset of more than 175,000 individuals – including some 4000 informal investors – in 28 highly developed countries over the period 2002–04. We distinguish between micro-level and macro-level determinants. The results uncover a positive virtuous circle where the demand for informal investment tends to generate its own supply as a result of micro and macro factors. Our results also suggest that higher levels of entrepreneurial activity at the country level increase the probability that venture capital and informal investment work in tandem with one another as complements rather than substitutes. Overall, we find that entrepreneurial activity whether ongoing or having resulted in exit appears to boost the supply of informal investors. This effect applies to both friends and family (F&F) and business angel investor types.

Keywords:

• informal investment
• venture capital
• entrepreneurship
• business angels

Acknowledgements

We are very grateful for the input of the editors and two anonymous referees who helped to improve the paper. Thanks are also owed to Siri Terjesen who provided helpful comments on an earlier draft. The paper has been written in the framework of the research programme SCALES, carried out by EIM and financed by the Dutch Ministry of Economic Affairs.

Notes

1. We use the term ‘school of thought’ to denote relevant theories/perspectives (often from diverse fields of economics and management) which have common arguments and implications relevant for an understanding of the determinants of informal investment activity.

2. We only label those individuals who respond to the (second) question of the relationship to the investee, as informal investor. Hence, respondents indicating to be an informal investor based on the first question (‘have you provided funds’) but who did not respond to the second question are labelled ‘no informal investor’. Vice versa, those who answered no on the first question, but did indicate a type of relationship with an investee, are counted as informal investor. This way we correct the responses of individuals who misunderstood the first question.

3. We realize that some business angels will invest in firms of friends and family as well.

4. More specifically, data for the variable ‘venture capital investment’ are taken from datasets provided in the ‘GEM Members Area’. For the year 2004 we have taken the variable labelled ‘Classic VC invested domestically $US (1000)’ from the financing dataset of 2004. The same variable is used for 2002, but this is taken from the National Venture Capital Data reported in the year 2003 and labelled briefly as ‘Domestic $US (1000): total classic’. We made an estimation to obtain data for venture capital investment in the year 2003. This estimation is based on classic VC investment growth rates from 2003 to 2004, published in the GEM 2005 executive report (Minniti, Bygrave, and Autio 2006, 49).

5. In order to obtain a measure for our first education category (low), we summated the GEM education variables ‘none’ and ‘some secondary education’. The GEM variable ‘secondary education’ is used for our second education category (middle). Our third education category (high) is obtained by adding the GEM variables ‘post-secondary education’ and ‘graduate experience’.

6. In particular, we use a variable labelled ‘interest rate, banks prime lending, per cent per annum, period average’.

7. Because we have a discrete left-hand side variable, we cannot include too many discrete variables on the right-hand side. In order to limit the number of discrete variables on the right-hand side, we use continent of country dummies instead of regular country dummies (note that all micro variables on the right-hand side are also of a discrete nature). The four macro-level variables in our model, together with the continent of country dummies, should capture the bulk of the macro-level variation in the data though.

8. Specifically, the lower developed countries include the countries categorized by the World Bank as ‘low-income economies’, ‘lower-middle-income economies’, or ‘upper-middle-income economies’, while the higher developed countries correspond to ‘high-income economies’.

9. This class is an aggregate of the four categories ‘close family member’, ‘some other relative, kin or blood relation’, ‘work colleague’ and ‘friend or neighbour’, as identified in the GEM survey.

10. For this estimation we also control for sector of investment. Note that this is not possible in the normal business angel prevalence regressions, because individuals who are no business angel – by definition – have no sector to invest in.

11. Kutsuna and Harada (2004) point out that informal investment in countries with bank-based financial systems are different from countries with market-based financial systems. Our analysis does not distinguish between these types of financial regimes. A fruitful line of future research might be to ascertain whether the effects found in this paper are uniform or different across different financial market regimes.