Learning and uncertainty in spatial resource management

Abstract

Natural resources such as fish, and wildlife have the ability to move across different areas within an ecosystem. Such movements are subject to random changes in environmental conditions (e.g., nutrients, temperature, oxygen). Although empirical evidence suggests that learning about such movements helps improve management, the related economic literature concentrates on scenarios in which the resource population lives in a closed area and cannot migrate. In this paper, we develop a spatial bioeconomic model to examine a renewable resource harvester’s responses to learning about fish movements. Our baseline is the scenario in which the harvester is fully informed about the distribution of fish movements. We find that introducing uncertainty and learning about fish movements critically affects extraction incentives. For instance, we show that uncertainty and learning may increase harvest in a patch and reduce harvest in another patch when the marginal harvesting cost function is constant. In the stock dependent marginal harvesting cost case, we delineate conditions under which uncertainty and learning increase harvest in all patches. We also show how harvest responses to learning change with the distribution of uncertainty.