Recharging the SEMATECH concept?
Posted by Evan Herrnstadt on December 18, 2008
Furthermore, the comparison to Sematech isn’t terribly exciting. A super-quick Google Scholar search reveals things like “Sematech induced members to cut their overall R&D spending on the order of $300 million per year” (Irwin & Klenow, 1996) and others. Lets hope that isn’t the result the battery people are seeking (getting someone else to pay for R&D they would otherwise do).
I agree to a large extent. For those who are unfamiliar with SEMATECH, the story begins in the heady year of 1986. I was 2 years old, the Bangles were on the verge of releasing “Walk Like an Egyptian”, and Duchess Sarah Ferguson married a Prince which paved the way for future stardom as a Weight Watchers spokeswoman. SEMATECH was an effort by U.S. semiconductor manufacturers to regain their competitiveness in the face of growing Asian dominance. It was initially funded in equal parts by the constituent firms and the Defence Advanced Research Projects Agency (DARPA). The basic idea behind a consortium is that since some discoveries will spread across the industry anyway, a joint investment can spread the costs across firms while still obtaining the final result. In addition, collaboration can reduce redundant innovation; in principle, innovation increases, costs fall, or both.
However, there are always pitfalls in such a collective action case. How do you determine a firm’s contribution to the consortium, in both funds and manpower? Is there to be a centralized facility, or will personnel exchange among labs? What prevents freeriding? In the end, SEMATECH lost its DARPA funding and restructured, but not before its focus had significantly shifted. Because so much of the most crucial R&D was proving intensely proprietary, the primary function of SEMATECH evolved toward encouraging R&D by semiconductor manufacturing equipment firms. Thus, the proprietary R&D could remain in-house, while the consortium members benefited roughly equally from the advances they funded. Of course, the phrase “in principle” comes to mind again. Because SEMATECH’s facilities were centralized, it was crucial for firms to send engineers and technicians in to aid technology transfer. Thus, large, vertically integrated firms benefited disproportionately while smaller companies were stretched thin trying to send even a few experts out to pick up the innovations.
So how does this lesson apply to batteries? From an ungated Reuters story (check it out if you couldn’t get the WSJ one):
The alliance, which includes battery industry giants such as 3M Co and Johnson Controls-Saft, intends to secure $1 billion to $2 billion in U.S. government funding over the next five years to build a manufacturing facility with an “open foundry” for the participants to pursue the goal of perfecting lithium-ion batteries for cars.
To me, this sounds like a consortium-run sandbox for trying out new techniques for lithium-ion battery production. Will the firms be able to effectively separate improving the efficiency of battery manufacturing from proprietary technological advances? I’m not sure, since it seems to me that some of the most profitable innovations will be process-based until a totally new battery paradigm arises. So I’m pretty skeptical myself, though I don’t really know enough about battery technology to make a good assertion. Anti-climactic, I know, but I hope you all enjoyed my story about the good old days.
Interestingly, USCAR is an existing auto industry umbrella organization of consortia that used to work on batteries but shifted a lot of money and energy to hydrogen. You might be surprised about the lack of wisdom in that decision until you look at the membership: Chyrsler, GM, Ford. Maybe we should also have a consortium to research fuel-efficient private jets.