Tuesday, October 02, 2007

A Project

One of my current research projects is thinking about the following question:

What changes to intellectual property laws would help entrepreneurs and entrepreneurship?

Note that the question is not what changes to the law would help inventors or innovators or artists or authors. The question is economic development. And the question is economic development at the local or regional level, rather than at the national level. Among other reasons, that's why I put the question out on this blog, rather than elsewhere.

Here are two possible examples:

One -- Change the law on non-competes. Pennsylvania law, like the law in most states, enforces non-competition agreements. "Non-competes" are contracts that companies sometimes require that their employees sign. The provision reads: If the employee leaves the company, he/she cannot go to work for a competitor for some specified period of time. PA law says that non-competes are valid so long as they are "reasonable" - they can last only for a reasonable length of time, and cover competitors in a reasonable geographic area, and limit their scope reasonably to employment in the employee's field.

What if non-competes were invalid? This is how California law works. Non-competes in CA are never enforced, except in connection with a sale of a business. Trade secret law is in full force, so an employee who goes to work for a competitor is free to do so but will be strung up by his/her thumbs if trade secrets go to the competitor as well. There is a fair amount of evidence that (i) California labor markets, particularly markets for technical and executive expertise, are more lively as a result, and (ii) the entrepreneurial marketplace is livelier to.

Two -- Change part of the law on mandatory assignment of inventions. Currently, the federal Bayh-Dole statute permits university researchers to patent inventions developed with federal research dollars. The theory is that the university community is more likely to be effective at commercializing the research than the federal government would be. In practice, virtually all research universities (like virtually all private industrial enterprises) require that faculty assign rights in their inventions to the university, in exchange for a share of any eventual royalties. Universities' tech transfer offices take over to try to commercialize the inventions. Yet universities are terrible at this. Just about all tech transfer offices are treated as profit centers by their universities (which is a bad idea), and just about all tech transfer offices just about break even annually. Entrepreneurially-minded faculty complain that tech transfer offices are obstacles to success, and private investors and entrepreneurs who would like to know more about university research find themselves stymied by tech transfer bureaucracies.

What if the law were changed to prohibit mandatory assignment of patentable inventions funded by federal research money? University faculty without the entrepreneurial spirit would still be free voluntarily to assign their inventions, but others would be free to keep their inventions and work directly with the private entrepreneurial community. Unlike the non-compete example, there is no evidence from some other jurisdiction that shows how a different system might work better. The current tech transfer system is clearly broken, however. There must be some way to do better.

These are thought experiments, not conclusions or even proposals. Thoughts?

7 comments:

Anonymous said...

http://www.entrepreneurship.fiu.edu/downloads/publications/papers/ResearchPapersinBusinessStudies.pdf

The paper essentially says that you need to get technological innovations to entrepreneurs, rather then public or non-profit agencies.

"Commercializing University Innovations: A Better Way", April 2007 also makes the point that TTOs are more gatekeepers than facilators.

Mike Madison said...

That's a useful link. Thanks.

The question that I'm thinking about is whether there are changes specifically to the legal system that are likely to make an entrepreneurial economy more productive.

Note that both of my hypotheticals are really employment law changes, not IP law changes!

Jefferson Provost said...

Bayh-Dole is a great example of how the road to hell is paved with good intentions.

Why not just repeal it, and put the government funded inventions back in the public domain. University professors generally don't need the promise of exclusive development and marketing rights as an incentive to invent things. Hundreds or even thousands of inventions in fields like computer science, are freely published every year, sans patents.

From an economic development standpoint, once a good idea exists, it's probably better policy to allow many companies to try to develop it than just one. The inventor who wishes to commmercialize his own ideas will still have the competitive advantage of his or her own expertise in the area.

Anyway, doesn't most of the economic activity around any particlar patented idea occur after it goes off-patent? Why not just skip that silly monopoly phase completely?

Jefferson Provost said...

Section 4.2.2 of the paper linked above suggests that giving university researchers more control over their inventions may not actually help much. I think it's a great characterization of the attitute of many on university science faculty toward entrepreneurship, and worth quoting at length:

The researchers quite clearly indicated that they are researchers not entrepreneurs, and that this choice appeared as much a lifestyle choice as that to become an entrepreneur. Many of the researchers expressed genuine frustration over the fact that somebody somewhere [government] seems to think they [the researchers] would have time to draft business plans
and go to cluster meetings. There simply is no time left over to even consider starting a firm. All time is consumed by teaching and research and if they had more time it would be used for research.

One former vice rector complained that representatives from regional development centres, science parks, and other government bodies meet with deans and rectors of universities. “But, that is not where the ideas are. They are with the researchers within the departments, but these guys never meet”.... It appeared as if professors and researchers deliberately stayed back from involvement in commercialization. The professors did not want to interact with the science park representatives because that would be time away from more important research. “We have tried to stay away as much as possible from their meetings, and we want it to be that way”. When asked what to do if one had a good idea that potentially could lead to a start-up company, a professor replied: “If you have a good idea, you go behind a corner and wait until it blows over, and then you go back to research”.

A strong conviction among researchers was that universities conduct research and do not start businesses. Those researchers, who want to start a firm, are welcome to do so, but then they are no longer part of the research community. The shared understanding among the researchers was that one cannot be a good researcher and a good entrepreneur. It is either or, but never both.

Mike Madison said...

Jeff,
I read that long quotation as consistent with my hypothetical proposal. Vest both ownership and control of commercializable research in individual faculty, rather than requiring disclosure to the university and assignment of any patentable material. If the faculty member wants to remain in classic research mode, having control means that's a viable option. For the faculty member who either wants to be an entrepreneur or wants to transfer ideas directly into an entrepreneurial community, having control opens a direct door to that option. Talk to VCs; they'd much rather talk directly to the researcher than talk to a tech transfer office.
Mike

Anonymous said...

For what it's worth, our own single-data-point experience at Vivisimo was that Carnegie Mellon did not hinder our desire to leave the university and start our own firm, way back in 2000. But I've heard plenty of horror stories from other entrepreneurs at diverse universities. Whatever the tech transfer policy is, it should be transparent, public, and fair, to avoid suspicion on the part of the inventor that their case will involve ad-hoc negotiation at which the inventor may not be experienced or skilled.

Raul Valdes-Perez

Anonymous said...

I don’t know much about legal issues, so perhaps I shouldn’t post, but I think one of the ideas here is the degree to which the legal recognition of an idea as property is a good one from an economic development standpoint. Jane Jacobs felt that upstate New York was a bad place from an entrepreneurial point of view, not because of the dearth of ideas (think Kodak, Xerox, IBM, Cornell, etc.) but because of the aggressiveness in how tightly property rights were held. From the research on Bayh-Dole, it sounds like granting universities a stake in the properties developed by their researchers, rather than stimulating the universities to act as entrepreneurs and aggressively seek out ways to exploit those ideas, instead resulted in the universities acting, well, like universities “managing” assets in a competent and predictable way.

West Virginia is looking at legislation that would exclude from taxes any business income derived from a patent or copyright created in West Virginia. The idea is that the property will be worth more in the state than outside of it, which may or may not influence the location decisions of some firms and entrepreneurs. But it seems unlikely, in and of itself, to have much of a spillover effect on the remainder of the economy.

I wonder how much of an effect the legal structure has on the commercializing innovation? The multi-state upper Appalachian region, stretching from Kentucky/West Virginia through upstate New York and non-coastal New England, seems to be struggling to compete on innovation. Is the legal framework in all of these states similar, and is their legal framework substantively different from others areas that are economically stronger?

The traditional view has always been that the lack of entrepreneurial activity in Western Pa. has been an issue of habit (there has not been an entrepreneurial tradition) or infrastructure (the institutions that support entrepreneurial behavior are simply not present). But I wonder if the blessing of past wealth has given the region a very conservative view of property assets, and that competence involves prudent management rather than innovative exploitation? If the asset is the exploitation, rather than the idea, it seems like that would promote more economic activity and spillover effects, given the right players.