The determinants of research and development and intellectual property usage among Australian companies, 1989 to 2002
Melbourne Institute Working Paper No. 27/04
Date: September 2004
This paper traces the innovation pathways of new creations from R&D activity through to intellectual property (IP) applications using enterprise panel data from 1989 to 2002. Our estimation method explicitly addresses the selection issues associated with missing R&D data which is a common problem among this type of data set. We find that R&D activity is a highly path dependent process that relies heavily on firm specific effects. These firm specific effects were subsequently found to be correlated with managerial style – more aggressive and intuitive managers have higher R&D ceteris paribus – and extensive use of incentive schemes for employees within the firm. In addition, we find that R&D is higher when the previous year’s enterprise debt ratio is lower, the speed of technological change is faster, the firm’s ability to absorb knowledge spillovers is greater and the product market is less contestable. Furthermore, these firms appear to be using the various methods of appropriation, IP and non-IP, as complementary packages to capture the quasi-rents from previous R&D expenditure rather than as substitutes.