³ÉÈË´óƬ

A pervasive assumption in the economics of technological change is that sponsors of university research seek fundamental output. Indeed, Academia rewards fundamental results through reputation and salary. However, the majority of US academic research sponsors have utilitarian goals which conflict with the goal of producing fundamental results. This paper evaluates the impact of this goal conflict on the effectiveness of knowledge procurement and its impact on academic output. A simple model characterizes attrition of relationships between a utilitarian sponsor and academically motivated researchers. I test implications of the model with a new dataset that tracks the academic productivity of 221 university researchers funded by NASA's aerospace engineering program in 1981; and then check for attrition from the program in 1988. A selection model is estimated to characterize the 79% attrition rate while a productivity analysis tests if directed grants suppress academically-valued output, as measured by publications and citations. For the researcher, results suggest that maintaining a relationship with NASA's program comes at the expense of academic reputation and that this motivates a search for less restrictive funding. For its part, NASA's program identifies researchers who will produce relevant knowledge with difficulty. The results highlight potential inefficiencies in researcher-sponsor relationships and more generally in contracting for services in the absence of monitoring or an ability to specify output. These empirical implications should be important. considerations for students of the economics of contracts, technological change and research policy.