Higher-Order Risk Preferences – Consequences for Test and Treatment Thresholds and Optimal Cutoffs

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. The working papers published in the Series constitute work in progress circulated to stimulate discussion and critical comments. Views expressed represent exclusively the authors' own opinions and do not necessarily refl ect those of the editors. Terms of use: Documents in Ruhr Economic Papers #287

Exploring the Consistency of Higher Order Risk Preferences

“…In lifecycle savings models, for example, prudence and temperance determine how current savings are influenced by the riskiness of future income (Kimball, 1990(Kimball, , 1992. Other areas in which higher order risk preferences have been theoretically shown to impact behavior include auctions (Esö and White, 2004), bargaining games (White, 2008), research and development expenditures (Nocetti, 2015), prevention (Eeckhoudt and Gollier, 2005;Courbage andRey, 2006, 2016;Peter, 2017) and medical decision making (Eeckhoudt, 2002;Felder andMayrhofer, 2014, 2017). However, what still needs to be established is the degree to which the previous findings on individual differences in higher order risk preferences are robust and sufficiently general in different contexts.…”

Measuring multivariate risk preferences in the health domain

“…As a result, it is widely recognized that risk preferences are an important determinant of health-related decisions. The recent theoretical literature has shown that many health-related behaviors in unidimensional settings should not only be governed by risk aversion, but that the higher order risk preferences prudence and temperance complement risk aversion in important ways (Courbage and Rey, 2006;Felder and Mayrhofer, 2014;Pauker, 2014). 1 Intuitively, prudence governs how optimal behavior changes in the presence of a background risk, while temperance determines whether independent zero-mean risks should be aggregated or not (see Section 2 for details).…”

“…Eeckhoudt and Schlesinger, 2006;Gollier and Pratt, 1996;Kimball, 1990). In the case of health, Courbage and Rey (2006) observe that the level of prudence is a main determinant of the optimal level of prevention for health risks, while Felder and Mayrhofer (2014) show that prudence and temperance affect optimal medical test and treatment decisions. As a result, Pauker (2014) advocates that the study of higher order risk attitudes should receive priority on the research agenda of medical decision making.…”

Decision Thresholds and Changes in Risk for Preventive Treatment

“…This notion defines a probability of disease threshold above which the preferable course of action is to undergo treatment, otherwise to withhold treatment. This notion has been initially defined for curative treatment in the absence of precise diagnosis and in relation to individual preferences, in particular, risk attitudes (see also Eeckhoudt, ; Felder and Mayrhofer, ). The aim of this paper is to investigate the notion of treatment threshold in the case of preventive care with potential side effects as a tool to help a DM in deciding to administer or not a preventive treatment.…”

“…We show that utility of higher orders drives the decision to engage in prevention activities and in particular preferences that are commonly referred to as mixed risk averse (Caballé and Pomansky, ). We consider risk preferences of higher orders than the one of risk aversion, prudence, and temperance as previously used in the literature of treatment threshold (Felder & Mayrhofer, ). Although intuitively one may think that a riskier environment always increases the treatment threshold, this paper shows that this is not necessarily the case and that results strongly depend on the risky environment considered.…”