In chapter 1, I provide a rationale for establishing a new subclass of impure public goods—group goods—and model group goods markets, deriving unexpected welfare results that compellingly motivate a more nuanced antitrust policy and suggest a novel way to achieve environmental sustainability.
Abstract: Scientists in various disciplines have routinely used private goods, club goods, common-property resources, and public goods to categorize and explain diverse consumption possibilities. Economists differentiate the four classes of goods via characteristics of rivalrousness and excludability. Even though these classes of goods exhaust all configurations of rivalrousness and excludability, the taxonomy of goods is incomplete. Here I introduce the concept of the group good—a good whose level of provision does not affect the size of a voluntary group of individuals who share its financing, ownership, and consumption—to capture the essence of an important class of consumption possibilities that the established classes of goods cannot properly describe and explain. Formal modeling of markets for group goods yields intriguing patterns of consumption, demand, and welfare that are consequential for policymaking. For example, policies that seek to increase the size of consumer groups may achieve both gains in economic welfare and lower equilibrium quantities of group goods. Subsequent reductions in carbon emissions associated with lower industrial output levels may result in improved environmental conditions and help attain a broader social goal of sustainability. The same properties inform caution with respect to policies that seek to change the consumer group size in a repressive fashion (such as the One-Child Policy in China), as these policies may decrease economic welfare.
In chapter 2 of my dissertation, I introduce the concept of bias interaction in contingent valuation studies and compare the voting referendum and dichotomous choice elicitation format, concluding that within the latter stated preferences are more reliable due to bias cancellation.
Abstract: Few studies investigate the link between the degree to which contingent valuation (CV) survey biases manifest themselves and modes of administering such surveys. Herein, I model a scenario which illustrates that the severity and direction of biases is conditional upon the choice of a format used to elicit values that people assign to provision of nonmarket public goods. I look at the interaction of two types of biases (the prestige bias and the free rider bias) within two elicitation formats (the voting referendum elicitation format (VREF) and the dichotomous choice elicitation format (DCEF)).
First, I provide an alternative answer to the question of why the results of the CV lab experiments, administered under the VREF, are more prone to a higher proportion of people donating toward a public good compared to the DCEF. Due to the presence of prestige bias and the fact that the majority decision binds within VREF, non-decisive voters have a dominant strategy to donate, which is not the case within DCEF. Second, I show that the DCEF is not necessarily a more accurate elicitation format, as it is susceptible to the free rider problem, unlike the VREF. Finally, I demonstrate that prestige bias can be a desirable feature in the context of the DCEF, because it mitigates the consequences of the free rider problem and makes dichotomous choice a more reliable elicitation format than the VREF.