Discussion Papers no. 133

An empirical study using microdata

R&D, spillovers and performance among heterogenous firms

Empirical research has established that there is a significant, positive relationship between productivity growth and R&D- expenditure at the firm level. Yet, while interesting, the conventional production function approach applied in such studies has some well known limitations.
This paper attempts to provide new insights into three main issues: (i) Heterogeneity in production relationships, in particular differences in innovative opportunities, across firms are emphasized throughout. (ii) There tends to be a correlation between investment decisions and the error term in production function regressions (even when specified in a growth rate form), that will bias the estimated parameters. The paper handles this problem more carefully than usual, by dealing explicitly with uncertainty and expectation- errors by means of instrumental variable estimation. (iii) R&D is an investment activity which involves significant adjustment costs.
This paper presents a new specifications of adjustment costs, where rapid changes in the R&D- program reduce the growth in knowledge capital. The results confirm the view that there are significant differences in innovative opportunities across firms (within narrowly defined industries). There is clear evidence that R&D- activity in a firm improves its performance. The results suggest that R&D- activities in competing firms will have a positive or negative effect on a firm's performance, depending on whether the firm is technologically advanced or not.

Om publikasjonen

Tittel

R&D, spillovers and performance among heterogenous firms. An empirical study using microdata

Ansvarlig

Tor Jakob Klette

Serie og -nummer

Discussion Papers no. 133

Utgiver

Statistics Norway, Research Department

Emne

Discussion Papers

Antall sider

40

Målform

Engelsk

Om Discussion Papers

Discussion papers comprise research papers intended for international journals and books. A preprint of a Discussion Paper may be longer and more elaborate than a standard journal article as it may include intermediate calculations, background material etc.

Kontakt