Does decision error decrease with risk aversion?

There is substantial evidence that risky decision-making involves a stochastic error process. The literature has adopted different approaches to address this issue, however, risk preferences are not uniquely identified by the most popular methods; decision error is not predicted to monotonically decrease with risk aversion. This paper reports the results of an experiment that elicits risk preferences to identify risk averse individuals and evaluates the frequency the stochastically dominant of two lotteries is chosen. Risk averse subjects exhibit a strong preference for dominant lotteries. More importantly, violations are consistent with stochastic decision error that decreases with risk aversion.     

Risk-taking for others under accountability

We let subjects take risky decisions that affect themselves and a passive recipient. Adding a requirement to justify their choices significantly reduces loss aversion. This indicates that such an accountability mechanism may be effective at debiasing loss aversion in agency relations.     

Risk breeds risk aversion

We examine whether exposure to a more or less risky environment affects people’s subsequent risk-taking behavior. In a laboratory setting, all subjects went through twelve rounds of multiple-price-list decisions between a risky alternative and a safe alternative. In the first six rounds, subjects were randomly assigned to a high-, moderate-, or low-risk environment, which differed in the variances of the lotteries they were exposed to. In the last six rounds, subjects in all treatments made decisions on an identical set of lotteries. We found that subjects who had experienced a riskier environment exhibited a higher degree of risk aversion. Our experimental design allows us to conclude that this effect is driven by the risk environment per se, rather than the realized outcomes of the risk. This finding has important theoretical and policy implications.     

Loss aversion

Loss aversion is traditionally defined in the context of lotteries over monetary payoffs. This paper extends the notion of loss aversion to a more general setup where outcomes (consequences) may not be measurable in monetary terms and people may have fuzzy preferences over lotteries, i.e., they may choose in a probabilistic manner. The implications of loss aversion are discussed for expected utility theory and rank-dependent utility theory as well as for popular models of probabilistic choice such as the constant error/tremble model and a strong utility model (that includes the Fechner model of random errors and Luce choice model as special cases).     

Higher-order risk preferences in social settings

We study prudence and temperance (next to risk aversion) in social settings. Previous experimental studies have shown that these higher-order risk preferences affect the choices of individuals deciding privately on lotteries that only affect their own payoff. Yet, many risky and financially relevant decisions are made in the social settings of households or organizations. We elicit higher-order risk preferences of individuals and systematically vary how an individual’s decision is made (alone or while communicating with a partner) and who is affected by the decision (only the individual or the partner as well). In doing so, we can isolate the effects of other-regarding concerns and communication on choices. Our results reveal that the majority of choices are risk averse, prudent, and temperate across social settings. We also observe that individuals are influenced significantly by the preferences of a partner when they are able to communicate and choices are payoff-relevant for both of them.     

Kindness,confusion, or … ambiguity?

Other-regarding preferences or decision errors are the main explanations put forward to justify contributions exceeding the non-cooperative optimum in VCM games. An alternative rationale relies on ambiguity aversion. Ambiguity aversion increases the perceived marginal benefit of own contributions, which in equilibrium will exceed the Nash level. We present a series of experiments testing this hypothesis. To control for other-regarding preferences, we run a two-player game in which a human player plays with a virtual agent. Players are assigned either to a risky setting (known probabilities of opponent’s choices) or to an ambiguity setting (probabilities of opponent’s contribution are vague). Results show that ambiguity affects contributions. However, attitude to ambiguity appears to be affected by the location of the aggregate Nash optimum inside the decision space.     

Ambiguity aversion among student subjects: the role of probability interval and emotional parameters

In the literature on decision-making under uncertainty, it has been shown that decision-makers tend to prefer taking gambles with known-risk probabilities (pure risk) over equivalent gambles with ambiguous probabilities. This article contributes to the ongoing discussion in the literature on cognitive and non-cognitive covariates of ambiguity aversion. Through a series of experiments, it finds that subjects are more ambiguity-averse to prospects with wide probability intervals than to an equivalent prospect with narrow intervals, and that subjects’ inherent trust, happiness and level of optimism affect the level of ambiguity aversion.     

Poverty and probability: aspiration and aversion to compound lotteries in El Salvador and India

Some experimental participants are averse to compound lotteries: they prefer simple lotteries that depend on only one random event, even when the simple lotteries offer lower expected value. This paper proposes that many behavioral “investments” represent more compound risk for poorer people—who often face multiple dimensions of deprivation—than for richer people. As a result, identical aversion to compound lotteries can prevent investment among poorer people, but have no effect on richer people. The paper reports five studies: two initial studies that document that aversion to compound lotteries operates as an economic preference, two “laboratory experiments in the field” in El Salvador, and one Internet survey experiment in India. Poorer Salvadoran women who choose a compound lottery are 27 percentage points more likely to have found formal employment than those who chose a simple lottery, but lottery choice is unrelated to employment for richer women. Poorer students at the national Salvadoran university choose more compound lotteries than richer students, on average, implying that aversion to compound lotteries screened out poorer aspirants but not richer ones. Poorer and lower-caste Indian participants who choose compound lotteries are more likely than those who choose simple lotteries to have a different occupation than their parents, which is not the case for better-off participants. These findings suggest that the consequences of aversion to compound lotteries are different in the context of poverty and disadvantage.     

EXPLORING THE CONSISTENCY OF HIGHER ORDER RISK PREFERENCES

This study measures higher order risk preferences and their consistency. We explore the role of country differences, the variation of stakes, and the framing of lotteries. We observe a robust dichotomous pattern of choice behavior in China, the United States, and Germany. The majority of choices are consistent with mixed risk aversion or mixed risk-loving behavior. We also find this pattern after a 10-fold increase in the stakes. Finally, our results reveal that this pattern is strengthened if the lotteries are displayed in compound instead of reduced form. In a follow-up study, we explore potential explanations for this framing effect.     

Entrepreneurial overconfidence and ambiguity aversion: dealing with the devil you know,than the devil you don't know

Various empirical studies find that entrepreneurs are systematically overconfident in their venture's probabilistic chances of success. Yet, entrepreneurs often face an ambiguous future that precludes them from making such probabilistic judgements. A theoretical framework based on ambiguity aversion is developed to explain an entrepreneur's overconfidence under complex and novel conditions of ambiguity. Unlike optimistic explanations, this ambiguity-averse form of overconfidence offers a non-probabilistic approach to entrepreneurial judgements of uncertainty.     

The effect of ambiguity aversion on reward scheme choice

We test the implications of ambiguity aversion in a principal–agent problem with multiple agents. Models of ambiguity aversion suggest that, under ambiguity, comparative compensation schemes may become more attractive than independent wage contracts. We test this by presenting agents with a choice between comparative reward schemes and independent contracts, which are designed such that under uncertainty about output distributions (that is, under ambiguity), ambiguity averse agents should typically prefer comparative reward schemes, independent of their degree of risk aversion. We indeed find that the share of agents who choose the comparative scheme is higher under ambiguity.     

Experimental evidence on personality traits and preferences

Using an experiment, we test the relation between personality traits and revealed risk and ambiguity preferences, and we consider the effects of personality traits prevalence in a group on the decision making of each group member. In the experiment, subjects reveal their risk and ambiguity preferences through lottery choices. They then participate in an unstructured group chat. Afterwards, they are given the chance to revise their initial lottery choices. Results show that personality traits affect ambiguity but not risk preferences before the chat. Specifically, agreeableness is negatively related to ambiguity aversion. We also show that the probability of changing decisions after the chat is affected by the individual's personality traits but not by the traits of the other group members. The latter only affects the direction and the degree of the change.     

Ambiguity Without a State Space

Many decisions involve both imprecise probabilities and intractable states of the world. Objective expected utility assumes unambiguous probabilities; subjective expected utility assumes a completely specified state space. This paper analyses a third domain of preference: sets of consequential lotteries. Using this domain, we develop a theory of objective ambiguity without explicit reference to any state space. We characterize a representation that integrates a non-linear transformation of first-order expected utility with respect to a second-order measure. The concavity of the transformation and the weighting of the measure capture ambiguity aversion. We propose a definition for comparative ambiguity aversion.     

Scientific Ambiguity and Climate Policy

Economic evaluation of climate policy traditionally treats uncertainty by appealing to expected utility theory. Yet our knowledge of the impacts of climate policy may not be of sufficient quality to be described by unique probabilistic beliefs. In such circumstances, it has been argued that the axioms of expected utility theory may not be the correct standard of rationality. By contrast, several axiomatic frameworks have recently been proposed that account for ambiguous knowledge. In this paper, we apply static and dynamic versions of a smooth ambiguity model to climate mitigation policy. We obtain a general result on the comparative statics of optimal abatement and ambiguity aversion, and then extend our analysis to a more realistic, dynamic setting, where we introduce scientific ambiguity into the well-known DICE model of the climate-economy system. For policy-relevant exogenous mitigation policies, we show that the value of emissions abatement increases as ambiguity aversion increases, and that this ‘ambiguity premium’ can in some plausible cases be very large. In these cases the effect of ambiguity aversion on welfare is comparable to that of other much studied welfare parameters. Thus ambiguity aversion may be an important neglected aspect of climate change economics, and seems likely to provide another argument for strong abatement policy.     

Stochastic Choice and the Allocation of Cognitive Effort

Data from a risky choice experiment are used to estimate a fully parametric stochastic model of risky choice. As is usual with such analyses, Expected Utility Theory is rejected in favour of a form of Rank Dependent Theory. Then an estimate of the risk aversion parameter is deduced for each subject, and this is used to construct a measure of the “closeness to indifference' of each subject in each choice problem. This measure is then used as an explanatory variable in a random effects model of decision time, with other explanatory variables being the complexity of the problem, the financial incentives, and the amount of experience accumulated at the time of performing the task. The most interesting finding is that significantly more effort is allocated to problems in which subjects are close to indifference. This presents us with another reason (in addition to statistical information considerations) why such tasks should play a prominent role in experiments. JEL Classification: C23, C91, D81     

UNCERTAIN LIFETIME AND INTERTEMPORAL CHOICE: RISK AVERSION AS A RATIONALE FOR TIME DISCOUNTING*

This article makes explicit the links between preferences over lotteries on length of life and intertemporal choice. It shows that the approach used by traditional life cycle models to account for uncertain survival corresponds to a strong assumption of risk neutrality with respect to length of life. Relaxing such an assumption leads us to develop a more general formulation of lifetime utility in which time discounting is directly related to preferences over length of life. Exponential and hyperbolic discounting are found to result, in a first approximation, from constant and hyperbolic risk aversion with respect to length of life.     

Irrational diversification in multiple decision problems

The paper deals with multiple decision problems, which are similar to the task of guessing the color outcomes of five independent spinnings of a roulette wheel, 60% of whose slots are red and 40% white. Each correct guess yields a prize of $1. The guess of 5 Reds clearly first order stochastic dominates any other strategy. In contrast, subjects diversify their choices when facing a multiple decision problem in which the choice is between lotteries with clear objective probabilities. The diversification is stronger when the subjects face uncertainty without objective probabilities and weaker when the choice problem involves real life actions.     

Anticipated regret as an explanation of uncertainty aversion

The paper provides a psychological explanation of uncertainty aversion based on the fear of regret. We capture an agent’s regret using a reference-dependent utility function in which the agent’s utility depends on the performance of his chosen option relative to the performance of the option that would have been best ex post. An uncertain option is represented as a compound lottery. The basic idea is that selecting a compound lottery reveals information, which alters the ex post assessment of what the best choice would have been, inducing regret. We provide sufficient conditions under which regret implies uncertainty aversion in the sense of quasi-concave preferences over compound lotteries.     

Ambiguity of scientific probability predictions and willingness-to-pay for climate change mitigation policies

We estimate a smooth ambiguity preference function, wherein an individual faces multiple probability predictions of policy outcomes, and then empirically measure their willingness-to-pay for the policy, ambiguity attitude, and ambiguity premium. Climate change mitigation policy is used as the example. The estimation results reveal that most people have ambiguity-seeking attitudes, but that these attitudes are heterogeneous across individuals. People who are older, are university graduates, have higher income, or trust more in science show stronger ambiguity-seeking attitudes. Their willingness-to-pay can be underestimated if ambiguity is not considered. Moreover, individuals with stronger ambiguity-seeking attitudes support aggressive mitigation policies more strongly. Our estimation strategy is generally applicable to policy evaluations wherein policy outcomes are ambiguous.     

Measuring Strategic Uncertainty in Coordination Games

This paper proposes a method to measure strategic uncertainty by eliciting certainty equivalents analogous to measuring risk attitudes in lotteries. We apply this method by conducting experiments on a class of one-shot coordination games with strategic complementarities and choices between simple lotteries and sure payoff alternatives, both framed in a similar way. Despite the multiplicity of equilibria in the coordination games, aggregate behaviour is fairly predictable. The pure or mixed Nash equilibria cannot describe subjects' behaviour. We present two global games with private information about monetary payoffs and about risk aversion. While previous literature treats the parameters of a global game as given, we estimate them and show that both models describe observed behaviour well. The global-game selection for vanishing noise of private signals offers a good recommendation for actual players, given the observed distribution of actions. We also deduce subjective beliefs and compare them with objective probabilities.