1) As Jack commented above, university inventions are often solutions in search of a problem (and usually only partial solutions at that). Identifying promising technologies seems unlikely to be improved by third party tech transfer agents.

2) Crossing the chasm to commercial success is generally a matter of execution, not invention. The talent and capital necessary to take a new technology to market seem to be far more limiting than the efficiency of technology transfer deals.

3) Nobody really wants to front the costs for patent protection and pre-market commercialization activities. Why would third party agents be any different or make this any better?

4) Industry, investors, and frustrated researchers often claim that universities want too much, too soon, with no risk. But where are the “incentives” for third party transfer agents going to come from? Aren’t they just another transaction cost (since the university’s percentage is supposedly unchanged)? It is almost impossible to correctly allocate risk, investment, and return when speculating on the combination of technology, IP, execution, and market factors over extended development cycles. I suspect that much of the frustration over these transactions on all sides results from this fundamental problem, not “monopolization” by TTOs. What we need are better data, deal models, and experience not another person in need of a return.

The big question is whether Bayh-Dole allocated too much value to the role of the university. It’s possible, but it is also possible that “free agent” advocates are overlooking the importance of the university system as a whole. It isn’t clear to me that separating the tech transfer role from the responsibility to the wider mission and values of the academic institution will ultimately strengthen the system. Researchers would end up serving two masters, some becoming industry researchers who happen to have a university address. Might this just encourage a few elite researchers and institutions to gravitate toward the commercialization model while others simply revert to ivory towers or deallocate space and staffing for research altogether?

Perhaps this is a step forward to an open innovation economy, with technology brokers serving to efficiently mate problems, solutions, teams, and capital. But agents and brokers already exist. There isn’t anything in Bayh-Dole that prevents them. If they were the silver bullet to better commercialization results, universities would already be motivated to allow competing mercenary agents to mine their research. Does anyone know of an actual proof of concept that free agent tech transfer improves transfer efficiency or is this pure speculation?

I *can* see instances where this plan could assist faculty entrepreneurs–public universities sometimes cap out the number of patents they can fund in a given year, and the rest of the gang are on their own. But the process at strong-TTO institutions is more complex, and the article doesn’t address specifics. What incentive exists for universities to give up the power to control licensing carve-outs? I’m missing something here.

Bayh-Dole (or more correctly, the way research institutions have chosen to implement Bayh-Dole) may create hurdles for swift licensing, but is this truly because of a disclosure bottleneck, or because US universities drive harder IP bargains than foreign competitors?

One good thing–some of my friends who are IP attorneys have been concerned about the future prospects of their field given the USPTO morass and the potential elimination of software patents. If free agency becomes real, their billable hours will likely skyrocket as the conflict sandbox is sorted out.