The impossibility of experimental elicitation of subjective probabilities
ABSTRACT We show that if decision makers may have stakes in certain events then the experimental elicitation of their subjective probabilities of these events is impossible.
- SourceAvailable from: au.dk[Show abstract] [Hide abstract]
ABSTRACT: If participation in the labour market helps to secure women's outside options in the case of divorce/separation, an increase in the perceived risk of marital dissolution may accelerate the increase in female labour supply. This simple prediction has been tested in the literature using time and/or spatial variation in divorce legislation (e.g., across US states), leading to mixed results. In this paper, we suggest testing this hypothesis by exploiting a more radical policy change, i.e., the legalization of divorce. In Ireland, the right to divorce was introduced in 1996, followed by an acceleration of marriage breakdown rates. We use this fundamental change in the Irish society as a natural experiment. We follow a difference-in-difference approach, using families for whom the dissolution risk is small as a control group. Our results suggest that the legalization of divorce contributed to a significant increase in female labour supply, mostly at the extensive margin. Results are not driven by selection and are robust to several specification checks, including the introduction of household fixed effects and an improved match between control and treatment groups using propensity score reweighting.Experimental Economics 01/2008; 13(3517). DOI:10.1007/s10683-010-9249-1 · 1.36 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: We introduce a simple method of constructing an incentive com-patible scoring rule to elicit an agent's belief about a random variable. The method does not depend on the exact form of the agent's utility function. Independent of her risk preference, it is optimal to report the value that minimizes the expected value of a loss function specified in the incentive scheme. The agent receives a fixed prize when her prediction error, defined by the relevant loss function, is smaller than a randomly generated value and earns a smaller prize or a penalty otherwise. Adjusting the loss function according to the belief elici-tation objective, the scoring rule can be used in a rich assortment of situations. Results from our probability elicitation experiment show that subjects' predictions are closer to the true probability under this scoring rule compared to the quadratic scoring rule.