Date of Completion


Embargo Period



Induced Innovation, Environmental Stringency, Cap-and-Trade, EU ETS, Kyoto Protocol, WTP, Non-Market Valuation, Latent Class

Major Advisor

Prof. Kathleen Segerson

Associate Advisor

Prof. Michele Baggio

Associate Advisor

Prof. Ling Huang

Field of Study



Doctor of Philosophy

Open Access

Open Access


This dissertation research contributes to the areas of environmental economics and industrial organization. In the first essay, I set out to understand firms' induced innovation in response to environmental regulation with an emphasis on the temporal decision making of regulated firms. I find that firms anticipate future increases in environmental stringency and strive to have patent applications submitted three years prior to the increase in environmental stringency. I find evidence that increases in the level of environmental stringency spur GHG-related innovations. In my second essay, I use a latent class model to relax the assumption of homogeneity in preference for willingness to pay (WTP) for quality improvements to the Puerto Rican coral reefs. I determine distinct subclasses and their WTP for amenity improvements in the population of visitors to Puerto Rico. Determining different subclasses' WTP allows for government policies that price discriminate using entrance fees for beaches, entrance fees for the coral reefs, and/or excise taxes on goods and services. My results indicate the two groups are indirect (beach goers, fishermen, etc.) and direct (snorkelers, divers, etc.) coral reef users. I use the results to suggest an illustrative policy that could raise approximately $9.9 million annually with a minimal impact on tourism. In my third essay, I incorporate respondents’ uncertainty for visitors’ WTP for quality improvements to the Puerto Rican coral reefs and the Olympic Coast Marine Sanctuary. I use a novel methodological approach, the nested uncertainty measure (NUM), and three goodness of fit measures to make comparisons to existing approaches that incorporate uncertainty. I find that my NUM improves goodness of fit for almost every measure of goodness of fit across two datasets (visitors to the Puerto Rican coral reefs and visitor to the Olympic Coast Marine Sanctuary) and two identification strategies. However, it does put positive pressure on WTP estimates exacerbating hypothetical bias.