PhD Graduate
Year Admitted
2018
Year Graduated
2026
Faculty Advisor(s)
Chris Costello (Social Science Group)
Committee
Andrew Plantinga, Steve Gaines
Dissertation Title & Abstract
The Design and Implementation of Index Insurance in Fisheries
Environmental shocks, such as marine heatwaves and harmful algal blooms, greatly damage fishing communities. Fishers have limited access to financial tools to ameliorate income loss due to diminished fishery productivity. Index insurance is a promising new tool to help protect fishers during environmental downturns. However, designing effective index insurance contracts for fisheries presents distinct and unresolved challenges. Unique fishery characteristics such as vulnerability to overexploitation, complex environmental to harvest relationships, and interaction with management have stymied deployment. This dissertation addresses each of the limitations in three chapters to enable effective and sustainable use of index insurance in fisheries.
Chapter one models fisher behavior changes induced through insurance moral hazard. My results show standard fishery production models in unconstrained settings lead insurance to incentivize increased harvest. Insurance will not be viable in the long run if it leads to overexploited stocks. Insurance may reduce harvest if fishers possess more adaptive risk mitigation strategies than standard model assumptions. Using estimates from Norwegian fisheries, index insurance may reduce harvest by 2\% or increase harvest by as much as 20\% depending on the type of insurance contract, risk characteristics of harvest, and fisher risk aversion.
Chapter two compares multivariate contracts built through machine learning algorithms with univariate contracts built through standard linear regression, using the California Market Squid fishery as a case study. Only one of twelve single-index contracts was viable in a private market, compared to three of four machine learning contracts. Community welfare would have improved had contracts built through machine learning been available over the past decade.
Chapter three examines how fisher decisions on insurance coverage influence management decisions on quota allocations. If management sufficiently mitigates risk exposure, fishers may not need insurance. Yet management decisions can harm fishing communities in the short run while protecting the long run — a gap index insurance can fill. Using stochastic dynamic optimization, I find that managers face two countervailing incentives when fishers have access to private insurance. Numerical results show these behavioral changes are modest, and only subsidized insurance meaningfully alters optimal harvest control rules.
This dissertation demonstrates that index insurance has the potential to promote climate resilient fisheries when combined with effective management and cutting edge data science to build good contracts.
Education
MESM, Environmental Science & Management, UCSB Bren School
BS, Marine Science & Economics, Coastal Carolina University