Interview with Suzanne Scotchmer

Jacques Cremer of the Toulouse Network for Information Technology (TNIT) recently interviewed Suzanne Scotchmer. They discussed her research on intellectual property and interest in innovation. Scotchmer is Professor of Economics, Professor of Law and Professor of Public Policy at the University of California, Berkeley. Her main academic interest at the moment is the economics, policy and law of innovation, including intellectual property.

TNIT: How did you begin working on intellectual property and R&D? What do you find particularly interesting/challenging in this field?

Scotchmer: I was appointed to the Berkeley faculty in 1986, in the midst of the biotechnology evolution. Disputes over intellectual property rights were in the newspapers daily, mostly involving the breadth of rights, patentable subject matter, and the problems that erupt when science proceeds in its natural way, with each discovery building on others.

There was a chasm between the way economists had conceptualized R&D and how it plays out in the lives of scientists and engineers: economists had mostly considered innovations in isolation. Thinking of innovation as a cumulative process led me to realize the contradiction between rapid technological change, which often occurs through rapid turnover in the market, and rewarding innovators, which requires enough longevity to recover costs. It was a puzzle to figure out how intellectual property rights can be structured to resolve this contradiction. I was hooked!

TNIT: You began life as a theorist. Is there any connection between your life as a theorist and your current interest in innovation?

Scotchmer: My work on R&D has deepened my interest in economic theory, and vice versa. Economic theory has two large branches. General equilibrium theory studies how interrelated markets clear in anonymous interactions. Game theory studies strategic behavior where interactions are not anonymous, and where agents’ strategies affect each other directly rather than through prices. Game theory and general equilibrium theory have not been unified in any grand unified theory. Nowhere is this more apparent than in the R&D context, which, in my view, has not been adequately integrated into either branch of economic theory.

Innovation does not fit into traditional models of general equilibrium because those models do not accommodate the introduction of new products or of cost-reducing innovations. Endogenous growth theory has made some strides, but only with a sterile stylization of the incentive problem that, in my view, is neither descriptive nor prescriptive.

Game theory, especially mechanism design, is the more natural tool for constructing theories of R&D incentives. However, mechanism design must begin from a well posed problem. In posing the problem, it seems to me that the most vexing and interesting aspect of the R&D problem is lost. For instance, to design an optimal patent system we should try to model the objectives of the legislators and all the constraints that they face when trying to influence the innovation process. An impossible task.

TNIT: Can you be more precise about the main difficulties in conducting this analysis. What are the most interesting questions?

Scotchmer: Innovation has two parts. Before there can be an investment in R&D, there must be an idea of what to invest in. The important first step, which economists have not much contemplated, is to understand where investment opportunities themselves come from. Once the investment problem is articulated, then the problem becomes “well posed”, and traditional tools of mechanism design can be applied to find the best incentive mechanism.

Traditional questions that mechanism design is well suited to are, for example: Which firm can offer the lowest cost? Which of several firms can offer the highest quality innovation? How do we know whether the innovation is worth its R&D cost?

How should the efforts of several simultaneous contributors be coordinated? Should rewards be given as prizes? Patents? Contract payments? What should rewards be contingent on?

For me, the most interesting question is “where do investment opportunities come from?” rather than “how do we give incentives to invest once the opportunity is realized?” Think for a moment about Mark Zuckerberg and Facebook. Would it have been possible to design an incentive mechanism to encourage the spark of the idea of Facebook? The essential step was the act of imagination that led to it, not the investment in server farms that reified it. Economists have no theory of imagination, and hence no theory for how to stimulate imagination.

TNIT: Many studies of R&D conclude that investments in R&D have a greater rate of return than investments in ordinary capital, such as plants and equipment. Why would that be so? Shouldn’t firms therefore invest more in R&D, and less in physical capital?

Scotchmer: My own interpretation is not that firms are irrational or short sighted, but rather that they don’t know what to invest in. All economics is about scarcity. In R&D, it is not the resources that are scarce as much as the imagination that is scarce. The high rates of return are not returns to resource investment, but rather to the scarcity of investment opportunities. It is imagination that is scarce.

Neither branch of economic theory has addressed the problem of stimulating imagination. By starting with a well posed problem, game theory suppresses the act of imagination, and general equilibrium theory either does not accommodate the introduction of new knowledge or takes the investment opportunities as given.

TNIT: You recently worked on open-source collaboration. How do you think this will change the way we write papers?

Scotchmer: Academic life at its best has always been “open source.” We are free to use the work of others with appropriate attribution, and we must return the privilege. Perhaps that is why academics find open source so congenial.

TNIT: Some of your co-authors are law scholars. How would you convince a young economist about the benefits of collaborating with lawyers?

Scotchmer: The most precious asset for a young scholar is a good problem to work on. Lawyers and judges, as opposed to academics, deal with hard problems not of their choosing. This forces them to work on important, new problems. The Microsoft and Google antitrust disputes are good examples. These firms fell into conflict with the competition authorities largely because neither the relevant market theories nor the right competition policies had been worked out. Their disputes continue to be a fertile ground for new thinking in economics.

TNIT: And now some quick opposites. Touch Type or Secretary?

Scotchmer: I am waiting for a direct digital line from my brain to yours.

TNIT: Facebook, or address book?

Scotchmer: I lose one and get lost in the other.

TNIT: JSTOR or paper copies in library?

Scotchmer: Where is the library?

TNIT: Coffee or mineral water?

Scotchmer: Do they put caffeine in mineral water?

TNIT: Twitter or not?

Scotchmer: Do I really have that much to say? (Answer from the editors of the Newsletter: yes, you do!)

TNIT: And finally, any plans for a new book?

Scotchmer: Yes, of course, don’t we all?