Subjective beliefs and economic preferences during the COVID-19 pandemic.

The COVID-19 pandemic presents a remarkable opportunity to put to work all of the research that has been undertaken in past decades on the elicitation and structural estimation of subjective belief distributions as well as preferences over atemporal risk, patience, and intertemporal risk. As contributors to elements of that research in laboratories and the field, we drew together those methods and applied them to an online, incentivized experiment in the United States. We have two major findings. First, the atemporal risk premium during the COVID-19 pandemic appeared to change significantly compared to before the pandemic, consistent with theoretical results of the effect of increased background risk on foreground risk attitudes. Second, subjective beliefs about the cumulative level of deaths evolved dramatically over the period between May and November 2020, a volatile one in terms of the background evolution of the pandemic. Supplementary Information: The online version contains supplementary material available at 10.1007/s10683-021-09738-3.

Eliciting Beliefs about COVID-19 Prevalence and Mortality: Epidemiological Models Compared with The Street.

Subjective belief elicitation about uncertain events has a long lineage in the economics and statistics literatures. Recent developments in the experimental elicitation and statistical estimation of subjective belief distributions allow inferences about whether these beliefs are biased relative to expert opinion, and the confidence with which they are held. Beliefs about COVID-19 prevalence and mortality interact with risk management efforts, so it is important to understand relationships between these beliefs and publicly disseminated statistics, particularly those based on evolving epidemiological models. The pandemic provides a unique setting over which to bracket the range of possible COVID-19 prevalence and mortality outcomes given the proliferation of estimates from epidemiological models. We rely on the epidemiological model produced by the Institute for Health Metrics and Evaluation together with the set of epidemiological models summarised by FiveThirtyEight to bound prevalence and mortality outcomes for one-month, and December 1, 2020 time horizons. We develop a new method to partition these bounds into intervals, and ask subjects to place bets on these intervals, thereby revealing their beliefs. The intervals are constructed such that if beliefs are consistent with epidemiological models, subjects are best off betting the same amount on every interval. We use an incentivised experiment to elicit beliefs about COVID-19 prevalence and mortality from 598 students at Georgia State University, using six temporally-spaced waves between May and November 2020. We find that beliefs differ markedly from epidemiological models, which has implications for public health communication about the risks posed by the virus.

The Risk of Gambling Problems in the General Population: A Reconsideration.

We examine the manner in which the population prevalence of disordered gambling has usually been estimated, on the basis of surveys that suffer from a potential sample selection bias. General population surveys screen respondents using seemingly innocuous "trigger," "gateway" or "diagnostic stem" questions, applied before they ask the actual questions about gambling behavior and attitudes. Modeling the latent sample selection behavior generated by these trigger questions using up-to-date econometrics for sample selection bias correction leads to dramatically different inferences about population prevalence and comorbidities with other psychiatric disorders. The population prevalence of problem or pathological gambling in the United States is inferred to be 7.7%, rather than 1.3% when this behavioral response is ignored. Comorbidities are inferred to be much smaller than the received wisdom, particularly when considering the marginal association with other mental health problems rather than the total association. The issues identified here apply, in principle, to every psychiatric disorder covered by standard mental health surveys, and not just gambling disorder. We discuss ways in which these behavioral biases can be mitigated in future surveys.

Disordered Gambling Prevalence: Methodological Innovations in a General Danish Population Survey.

We study Danish adult gambling behavior with an emphasis on discovering patterns relevant to public health forecasting and economic welfare assessment of policy. Methodological innovations include measurement of formative in addition to reflective constructs, estimation of prospective risk for developing gambling disorder rather than risk of being falsely negatively diagnosed, analysis with attention to sample weights and correction for sample selection bias, estimation of the impact of trigger questions on prevalence estimates and sample characteristics, and distinguishing between total and marginal effects of risk-indicating factors. The most significant novelty in our design is that nobody was excluded on the basis of their response to a 'trigger' or 'gateway' question about previous gambling history. Our sample consists of 8405 adult Danes. We administered the Focal Adult Gambling Screen to all subjects and estimate prospective risk for disordered gambling. We find that 87.6% of the population is indicated for no detectable risk, 5.4% is indicated for early risk, 1.7% is indicated for intermediate risk, 2.6% is indicated for advanced risk, and 2.6% is indicated for disordered gambling. Correcting for sample weights and controlling for sample selection has a significant effect on prevalence rates. Although these estimates of the 'at risk' fraction of the population are significantly higher than conventionally reported, we infer a significant decrease in overall prevalence rates of detectable risk with these corrections, since gambling behavior is positively correlated with the decision to participate in gambling surveys. We also find that imposing a threshold gambling history leads to underestimation of the prevalence of gambling problems.

Behavioral responses to surveys about nicotine dependence.

Behavioral responses to surveys can significantly affect inferences about population prevalence unless correctly modeled statistically. An important case study is the prevalence of nicotine dependence, a formal psychiatric disorder satisfying clinical criteria. Data from the National Epidemiologic Survey on Alcohol and Related Conditions in the United States are used, along with a flexible semi-nonparametric sample selection model. Corrections for sample selection responses to "gateway" survey questions lead to significantly higher estimates of the prevalence of nicotine dependence among current daily smokers. These corrections also imply even higher levels of the decades-long and lifetime-long persistence of nicotine dependence after the onset of smoking.

Estimating the subjective risks of driving simulator accidents.

We examine the subjective risks of driving behavior using a controlled virtual reality experiment. Use of a driving simulator allows us to observe choices over risky alternatives that are presented to the individual in a naturalistic manner, with many of the cues one would find in the field. However, the use of a simulator allows us the type of controls one expects from a laboratory environment. The subject was tasked with making a left-hand turn into incoming traffic, and the experimenter controlled the headways of oncoming traffic. Subjects were rewarded for making a successful turn, and lost income if they crashed. The experimental design provided opportunities for subjects to develop subjective beliefs about when it would be safe to turn, and it also elicited their attitudes towards risk. A simple structural model explains behavior, and showed evidence of heterogeneity in both the subjective beliefs that subjects formed and their risk attitudes. We find that subjective beliefs change with experience in the task and the driver's skill. A significant difference was observed in the perceived probability to successfully turn among the inexperienced drivers who did and did not crash even though there was no significant difference in drivers' risk attitudes among the two groups. We use experimental economics to design controlled, incentive compatible tasks that provide an opportunity to evaluate the impact on driver safety of subject's subjective beliefs about when it would be safe to turn as well as their attitudes towards risk. This method could be used to help insurance companies determine risk premia associated with risk attitudes or beliefs of crashing, to better incentivize safe driving.

Individual discount rates and smoking: evidence from a field experiment in Denmark.

We elicit measures of individual discount rates from a representative sample of the Danish population and test two substantive hypotheses. The first hypothesis is that smokers have higher individual discount rates than non-smokers. The second hypothesis is that smokers are more likely to have time inconsistent preferences than non-smokers, where time inconsistency is indicated by a hyperbolic discounting function. We control for the concavity of the utility function in our estimates of individual discount rates and find that male smokers have significantly higher discount rates than male non-smokers. However, smoking has no significant association with discount rates among women. This result is robust across exponential and hyperbolic discounting functions. We consider the sensitivity of our conclusions to a statistical specification that allows each observation to potentially be generated by more than one latent data-generating process.

Evaluating the tobacco settlement damage awards: too much or not enough?

OBJECTIVES: This study compared the present economic value of the 1998 tobacco settlement with the present economic value of the damages attributable to tobacco. METHODS: The 1987 National Medical Expenditure Survey was used to estimate the smoking attributable fraction (SAF) of medical expenditures. SAFs were then applied to Medicaid and other expenditures. RESULTS: Settlement payments covered only 40% of Medicaid treatment costs already incurred and only 30% of past and projected future Medicaid costs. Excess medical expenditures for all other payment sources were roughly comparable to those incurred by Medicaid. CONCLUSIONS: Although the settlement may reduce future smoking prevalence rates by limiting the ability of tobacco companies to promote smoking and by raising cigarette prices, euphoria over the huge settlement funds should be balanced by a sober comparison with the even larger damage amounts.