Agricultural insurance can be classified into two broad categories: Compulsory (government/public) and Voluntary (market/private) insurance. In practice, the vast majority of compulsory insurance is partially compulsory, where compulsory insurance provides only partial coverage, and it allows for supplemental voluntary purchases (mixed insurance). In this article, we use the Balloon Analogue Risk Task as the assessment of risk-taking and insurance context to conduct an online experiment. The main objective is to compare compulsory, voluntary and mixed insurance in terms of adverse selection and moral hazard. We find adverse selection in purely voluntary insurance, but advantageous selection in mixed insurance. Moral hazard exists in all three types of insurance, but it is smaller in mixed insurance. Our ancillary results suggest that under the combined effects of significant moral hazard and "no adverse selection" in purely compulsory insurance make it the insurance type with the lowest social earnings. Overall there is no crowding-out effect of the compulsory part on residual voluntary purchases in mixed insurance.

This article studies the stability of risk-preference during the COVID-19 pandemic. The results differ between risk-preference measurements and also men and women. We use March 13, 2020, when President Trump declared a national state of emergency as a time anchor to define the pre-pandemic and on-pandemic periods. The pre-pandemic experiment was conducted on February 21, 2020. There are three on-pandemic rounds conducted 10 days, 15 days, and 20 days after the COVID-19 emergency declaration. We include four different risk-preference measures. Men are more sensitive to the pandemic and become more risk-averse based on the Balloon Analogue Risk Task (BART). Women become more risk-averse in the Social and Experience Seeking domains based on the results from the Domain-Specific Risk-Taking (DOSPERT) and Sensation Seeking Scales (SSS). Both men's and women's risk-preference are stable during COVID-19 based on a Gamble Choice (GC) task. The results match our hypotheses which are based on the discussion about whether the psychological construct of risk-preference is general or domain-specific. The differential outcomes between incentivized behavioral and self-reported propensity measures of risk-preference in our experiment show the caveats for studies using a single measure to test risk-preference changes during COVID-19.

No abstract

No abstract

Women often respond less favorably to competition than men. In this paper, we test for the effects of social norms on willingness to compete. Subjects compete in two-person teams. In the treatment, one team member is randomly assigned the role of "breadwinner", and the other person is randomly assigned as the "supporter". There are no real differences between the roles in our experiment, except for the framing. These two roles have opposite social norms for competitiveness, reminiscent of gender roles in western society. In the baseline, subjects compete in two-person teams without role assignment. We find women's willingness to compete significantly increases when they are assigned as breadwinners compared to women in the baseline or female supporters. We also find that there is no gender gap in willingness to compete between female breadwinners and males in the baseline. The increase in willingness to compete is mainly contributed by high-ability women. Males are also affected by the role assignment; male supporters are less likely to enter the tournament than male breadwinners. We argue that the changes in willingness to compete are mainly driven by the social norms implied by the two roles.