Journal of Risk and Uncertainty最新文献

Abstract

Does exposure to cognitive load affect key properties of economic behavior? In this experiment, subjects face a series of simple binary decision tasks between prospects, testing for monotonicity in monetary payments, consistency with (first-order) stochastic dominance, reduction of compound lotteries, risk attitudes, and ambiguity attitudes. Cognitive load is manipulated via simultaneous memory tasks. Our data show treatment differences resulting from cognitive load for decision tasks with risky outcomes. However, cognitive load has no impact on monotonicity and ambiguity attitudes. Under a dual-process view of human decision-making, our findings suggest that ambiguity attitudes and preferences for “more certain money” are intuitive, not reasoned.

Do physicians behave rationally when facing a new disease? This study assesses physicians’ ambiguity attitudes towards the future severity of the COVID-19 pandemic in its early stages and the financial market in the US using an incentive-compatible online experiment. Our findings indicate that physicians demonstrate significant deviations from expected utility, characterized by a modest degree of ambiguity aversion and pronounced levels of likelihood insensitivity. While physicians generally show less insensitivity to uncertainty compared to the general public, both groups exhibited similar levels of irrationality when dealing with the ambiguity surrounding the COVID-19 severity. These results underscore the necessity for debiasing strategies among medical professionals, especially in managing real-world uncertainties, with a specific focus on mitigating likelihood insensitivity.

We test for a novel pattern of menu-dependent risk attitudes that forms the basis of recent theories of risky choice: Does expanding the range of potential prizes from lotteries in a choice set lead people to overweight those prizes and make riskier choices? Contrary to our hypothesis, we find no evidence of such a menu effect. Varying the potential prize offered by an actuarially unfavorable, high-risk lottery does not affect the likelihood of choosing a different, moderate-risk gamble in favor of a safer alternative. Our well-powered null results cast doubt on prominent theories of menu-dependent risk preferences.

We assess whether the COVID-19 vaccine induces COVID-19 risky behavior (e.g., going to bars and restaurants) and thus reduces vaccine efficacy. A key empirical challenge is the endogeneity bias when comparing risk-taking by vaccination status since people choose whether to get vaccinated. To address this bias, we exploit rich survey panel data on individuals followed before and after vaccine availability over fourteen months in an event study fixed effects model with individual, time, sector, and county-by-time fixed effects and inverse propensity weights. We find evidence that vaccinated persons, regardless of the timing of vaccination, increase their risk-taking activities. The evidence is consistent with the “lulling effect”. While vaccine availability may reduce the risk of contracting COVID-19, it also contributes to further spread of the virus by incentivizing risk-taking in the short term.

Abstract

As climate variability is increasing, extreme events such as temperature fluctuations are expected to become more frequent. Low- and middle-income countries (LMICs) are especially vulnerable to heat-related variability and its ensuing impacts on mortality. Therefore, there is an urgent need to understand how citizens in LMICs trade-off climate-related mortality risks with other risks such as traffic accidents, and what values they place on reducing such risks. As populations in LMICs are income-constrained, we adopt a non-monetary, risk-risk trade-off (RRTO) valuation method instead of the standard willingness-to-pay stated preference-based approach. We estimate the resulting risk premium for heatwave-related mortality risks through an adapted double-bounded, dichotomous choice approach to establish whether, on average, people value avoiding these risks more compared to reducing traffic risks. Using a sample of over 2,300 individuals from across seven states in India, a country with one of the highest heat-related mortality globally, we estimate the heatwave risk mortality premium to be between 2.2–2.9, indicating that on average, individuals weigh reducing heatwave-related mortality risks more than two times that of reducing traffic accident mortality risks. Based on a standard benefit transfer methodology for LMICs, this premium translates to a Value of Statistical Life (VSL) of USD 0.37–2.61 million for India.

In this study, we compare ambiguity attitudes—ambiguity aversion and ambiguity-generated insensitivity (a-insensitivity)—toward natural and artificial sources of ambiguity in gain and loss domains with the participation of individuals with various attributes. In our experiment, we use precipitation during the rainy season as a natural source of ambiguity and the Ellsberg-type box as an artificial source. We find that people are more a-insensitive toward the natural source than the artificial source, even though the outcomes are identical. Additionally, people with low cognitive reflection ability are more a-insensitive than those with high cognitive reflection ability. Thus, people with low cognitive reflection ability have more difficulty in identifying likelihood under ambiguity and tend to view the likelihood of all uncertain events to be equal. Furthermore, we examine the relationships between ambiguity attitudes and real-world behaviors with regard to flood preparedness. In the group with high cognitive reflection ability, people with higher a-insensitivity are less likely to adopt flood preparedness behaviors in the gain domain of the natural source. However, we do not find any relationship between ambiguity attitudes and flood preparedness behaviors in the artificial source. Thus, applying ambiguity attitudes toward natural sources is worth considering when explaining real-world behaviors based on ambiguity attitudes.

This paper studies the policy implications of correlation between preferences and internalities in the context of tobacco products. Using novel survey data, I show that cigarette smokers who misperceive the relative health harms of cigarettes and e-cigarettes—and thus for whom internalities associated with imperfect information are potentially large—are also significantly less likely to respond to changes in relative prices. I build this heterogeneity into a model of cigarette and e-cigarette taxation to show that the relationship between the optimal e-cigarette tax and the mean elasticity of substitution is relatively flat. This is policy relevant because evidence of substitution is thought to suggest low (or even negative) e-cigarette taxes. Even at implausibly large degrees of substitution, simulated optimal e-cigarette taxes are positive and large.

This paper builds on the data from a published paper on behaviour under ambiguity (Conte & Hey, 2013)—henceforth C&H—to explore the determinants of decision time. C&H categorized individual subjects as being of one of four types (of decision-maker)—Expected Utility, Smooth Ambiguity, Rank Dependent and Alpha Expected Utility—by using the decisions of the subjects, but did not look at the decision times of the different types. We take as given the categorization identified by C&H, and explore whether the classification can explain the decision times of the subjects. We investigate whether and why different types take a different amount of time to decide. We explore the effects of various features related to (mainly psychological) theories of the process of decision-making—i.e., experience with the task, complexity, closeness to indifference and similarity of the options. Our results show that different types take a similar time to make their decisions on average, but decision times of different types are explained by different features of the decision task. This paper is the first investigating the heterogeneity of decision times based on a classification of subjects into different types in an ambiguous (rather than risky) decision context.