Insurance decisions under nonperformance risk and ambiguity

An important societal problem is that people underinsure against risks that are unlikely or occur in the far future, such as natural disasters and long-term care needs. One explanation is that uncertainty about the risk of non-reimbursement induces ambiguity averse and risk prudent decision makers to take out less insurance. We set up an insurance experiment to test this explanation. Consistent with the theoretical predictions, we find that the demand for insurance is lower when the nonperformance risk is ambiguous than when it is known and when decision makers are risk prudent. We cannot attribute the lower take-up of insurance to our measure of ambiguity aversion, probably because ambiguity attitudes are richer than aversion alone.

Effortful Bayesian updating: A pupil-dilation study

When confronted with new information, rational decision makers should update their beliefs through Bayes’ rule. In economics, however, new information often includes win-loss feedback (profits vs. losses, success vs. failure, upticks vs. downticks). Previous research using a well-established belief-updating paradigm shows that, in this case, reinforcement learning (focusing on past performance) creates high error rates, and increasing monetary incentives fails to elicit higher performance. But do incentives fail to increase effort, or rather does effort fail to increase performance? We use pupil dilation to show that higher incentives do result in increased cognitive effort, but the latter fails to translate into increased performance in this paradigm. The failure amounts to a “reinforcement paradox:” increasing incentives makes win-loss cues more salient, and hence effort is often misallocated in the form of an increased reliance on reinforcement processes. Our study also serves as an example of how pupil-dilation measurements can inform economics.

Crowded out: Heterogeneity in risk attitudes among poor households in the US

Not much is known about the heterogeneity of risk attitudes among poor households in rich countries. This paper provides estimates from a unique data set collected among the urban poor in Atlanta, Georgia. The data set includes lab-in-the-field experiments on the relationship between risk attitudes and several household characteristics. Apart from looking at income, wealth, and education, we are particularly interested in household composition as it captures the number and kind of people who are dependant on the income of the household head. Heads of households who are less risk averse may be willing to take on the extra risk from smaller resource margins resulting from additional dependants, implying a negative relationship between household size and risk aversion. However, if the size of the household is a result of exogenous forces some heads of households may become more risk averse with more dependants. Household size can also reflect a risk management choice that involves adding non-dependant members who can provide resources and risk sharing. However, this possibility is limited to homes that are not already too crowded. We find that household size correlates positively with the risk aversion of the head, but with a large proportion of children the correlation is strongly dampened. However, this negative effect of children is conditional on the home not already being crowded. These heterogeneous findings have implications for the design of new insurance, savings, and credit programs where risk attitudes are important to the decisions to adopt.

Quantifying loss aversion: Evidence from a UK population survey

We quantify differences in attitudes to loss from individuals with different demographic, personal and socio-economic characteristics. Our data are based on responses from an online survey of a representative sample of over 4000 UK residents and allow us to produce the most comprehensive analysis of the heterogeneity of loss aversion measures to date. Using the canonical model proposed by Tversky and Kahneman (1992), we show that responses for the population as a whole differ substantially from those typically provided by students (who form the basis of many existing studies of loss aversion). The average aversion to a loss of £500 relative to a gain of the same amount is 2.41, but loss aversion correlates significantly with characteristics such as gender, age, education, financial knowledge, social class, employment status, management responsibility, income, savings and home ownership. Other related factors include marital status, number of children, ease of savings, rainy day fund, personality type, emotional state, newspaper and political party. However, once we condition on all the profiling characteristics of the respondents, some factors, in particular gender, cease to be significant, suggesting that gender differences in risk and loss attitudes might be due to other factors, such as income differences.

The value of statistical life in the context of road safety: new evidence on the contingent valuation/standard gamble chained approach

This study estimates the value of statistical life (VSL) on a road traffic accident using the Contingent Valuation/Standard Gamble chained approach. A large representative sample (n = 2020) is used to calculate a VSL for use in the evaluation of road safety programmes in Spain. The paper also makes some methodological contributions, by providing new evidence about the consistency of the chained method. Our main results are: (1) A range from 1.3 million euro to 1.7 million euro is obtained for the VSL in Spain in the context of road accidents. This range is in line with the values used in the same context in other European countries, although it is lower than those obtained in different contexts and with other methods. (2) The method performs much better in terms of scope sensitivity than the traditional contingent valuation method, which asks subjects about their willingness to pay for very small reductions in the risk of death. (3) We introduce a new ‘indirect’ chaining approach which reduces (but does not remove) the disparity between direct and indirect chaining approaches. More extreme VSL estimates are still obtained with this indirect method than with the direct one. (4) VSL estimates depend on the injury used. More specifically, we obtained a lower VSL when a more severe injury is used. (5) Framing the risk of death in the modified standard gamble question as “10n in 10,000” instead of “n in 1000” influences the value of VSL. We attribute this effect to the Ratio Bias.