5 Conclusion.

"Entrepreneurship" is frequently advocated in the public debate as the solution to all (or at least most of) our trouble. However, often we do not know precisely what is meant with it, nor is it clear, what the different empirical measures of entrepreneurship are capturing. In this paper, we have tried to compare the theoretical ideas with what people are actually measuring.

First, we found that the different theoretical definitions of entrepreneurship reflect a diverse set of ideas about the role of entrepreneurship in the economy, involving aspects such as innovation, uncertainty-bearing, opportunity-seeking, and management. While the different perceptions of entrepreneurship may all be important for the economy--we should expect them to have rather different consequences. Thus, while both the uncertainty-bearer and the innovator are needed for the functioning of the economy, they perform different roles. This also implies that the appropriate set of policy instruments depend on which aspect of entrepreneurship that should be promoted.

Second, these different aspects are to varying degrees reflected in the various empirical measures. While some of these are good indicators of what we have termed "Knightian entrepreneurship," others may better capture (some) aspects of Schumpeterian or Kirznerian entrepreneurship. In general, stock measures seem better suited at measuring Knight's ideas, while flow measures to a large extent capture those of Schumpeter and Kirzner.

This distinction needs only be of academic interest if all empirical measures show the same picture. Thus, if the measures reflecting Knightian entrepreneurship are highly correlated with Schumpeterian indicators, measuring one or the other would not matter much. And it could be conjectured that policies promoting one would also promote the other.

However, we found considerable differences when comparing the levels of entrepreneurship across countries using the different measures. Thus, the different empirical measures not only reflect different theoretical aspects, they also provide very different pictures of the entrepreneurial activity level across countries. Hence, we cannot in general identify countries which are less entrepreneurial. Instead, some countries are entrepreneurial in a Knightian sense while less so in a Schumpeterian sense.

Many policies aimed at increasing entrepreneurial activities, rely on measures of entrepreneurial activity to document weak spots--or less entrepreneurial areas. Given that the empirical measures reflect rather different aspects of entrepreneurship--policy-makers should always be aware of the appropriateness of the empirical measure used as a measure of their perception of entrepreneurship in the given situation. The growth prospects and ability to create new jobs from Schumpeterian entrepreneurship is most likely to differ from that of Knightian entrepreneurship.

Furthermore, the substantial differences across measures highlight the problems involved in constructing composite indicators based on two or more of the specialized measures. Such composite indicators are likely to be very sensitive to the weights attached to the included indicators.

More research is definitely needed in this area to provide both better measures of the different aspects of entrepreneurship and to further clarify the relationships between the different measures and our understanding of entrepreneurship.

Jens Iversen (1), Rasmus Jorgensen (2) and Nikolaj Malchow-Moller (3)

(1) Centre for Economic and Business Research (CEBR) and University of Aarhus, Denmark, jiv@cebr.dk

(2) Centre for Economic and Business Research (CEBR) and University of Copenhagen, Denmark

(3) Centre for Economic and Business Research (CEBR) and University of Southern Denmark, Denmark