How do university inventions get into practice?

For most readers of this paper, the striking increase in patenting and licensing activities at American research universities since the late 1970s is a familiar story. The number of patents issued to universities and colleges more than doubled between 1979 and 1984, did so again between 1984 and 1989, and nearly doubled once more during the 1990s. In 1980 approximately 20 universities had technology licensing and transfer offices. By 1990 that number was 200, and by 2000 nearly every major research university had one. University license revenues have increased from $220 million to $698 million from 1991 to 1997 alone (Association of University Technology Managers 1998). Most observers attribute these trends to the Bayh-Dole Act, which was passed in 1980. Several years ago a team of researchers, including many authors of this paper, together with collaborators at the University of California, began a study that explored in detail what was going on behind the figures in the three American research universities that then had the highest revenues from licensing: Columbia University, Stanford University, and the University of California. The study worked with the collection of ‘invention reports’ that all of these universities had been requiring their researchers to make out whenever their research came up with something that seemed of likely commercial value.