DOE Human Genome Program Contractor-Grantee Workshop VIII
February 27-March 2, 2000  Santa Fe, NM

139. An Economic Analysis of Intellectual Property Rights Issues Concerning the Human Genome Program

David J. Bjornstad¹ and Steven Steward²

¹Oak Ridge National Laboratory, Oak Ridge, TN and ²University of Tennessee

dub@ornl.gov

Information produced by the human genome program and by private investors is given asset status through the patent process. It is then traded among agents who assemble it along with other factors of production to produce products for final consumption. This project seeks to examine the efficiency of markets in effecting this assemblage and the role of patent policy and other public policies in changing this efficiency. It does this by constructing economic models and studying them using the methods of experimental economics.

Efforts during the first year have sought to develop a description of this overall process and to create a logic for studying its components. We view value being added to the stock of intellectual information through a series of phases. The first phase is the development of the base genome itself. The second phase is the identification of the functionality of the genome. The third phase is the identification of the significance of individual variability from the base genome. The fourth phase is the development of mitigating responses to variability. The final step is recovery of value by sales for final consumption. Patent policy establishes property rights across this continuum. Thus, the debate over the patentability of an EST may be framed according to the manner in which information on the base genome must be combined with information on functionality to define patent utility. Recovery of value through final consumption requires acquiring property rights across the four phases. Hence, the demand for final consumption is passed backward through the phases as a derived demand. In other words, the information about the base genome acquires value in combination with function, variability and mitigating response, as well as investment alternatives. Agents bid for patent rights not for their own sake but because of the combinatorial value.

We model this activity through a market in which agents make choices between investing in R&D to acquire property rights or by purchasing established property rights. We view agents as having resource endowments that may be "money," property rights, or some combination thereof. They interact over a series of time periods during which they can choose to invest in R&D or to buy or sell property rights. They interact through a market environment that has a number of fixed properties and several variable properties. Variable properties are used to implement alternative government policies. Government develops patent policies to guide the establishment of intellectual assets and R&D policies to guide its contribution to the stock of knowledge. For example, government may set the standard for the number of pieces of information necessary to acquire a patent, with different requirements leading to different behavior by agents. Government may affect the potential value of R&D investments by agents through its own R&D investment strategies. Other policy parameters include the degree to which information, like R&D success, is made public, by the "technologies" through which R&D investment may take place, and by the role of uncertainty, for example, as an attribute of information quality. Uncertainty can enter the market in a number of additional ways as well, each of which complicates the analysis.

A key aspect of the market is the requirement that information from the different stages be assembled in specific combinations. This is similar to an airplane trip requiring rights to takeoff and land at specific times at specific airports or the assemblage of an urban site requiring the lease or purchase of a specific configuration of properties. For example, if ESTs are patented, several EST may to be required to form a gene. This, in turn, must be combined with information from other stages. For any given family of final consumption products, total value to be distributed among the agents is limited by the demand for the final consumption products. These distributions will be used to define market efficiency and departures from it. We are interested in how market forces, technical attributes and government policies lead to different distributions of rent.

We are now completing our study of the individual parts of this system and are preparing parameters for portions of the market that can be studied using experimental economics.