Topic IIE: A New Approach to Loss AversionKoszegi & Rabin (2006):They develop a new and improved theory ofreference-dependent utility with loss aversion.Specifically, they address two major issues — two“loopholes” in the existing approach:(1) What determines the reference point?(2) When do people experience loss aversion, andwhat is the magnitude of this experience?They address these issues by incorporating two novelfeatures:(1) A person’s reference point is her recent beliefs orexpectations about outcomes.(2) Gain-loss utility is directly tied to the intrinsic utilityfrom consumption — so that a person experiences moregain-loss utility for goods that involve more consumptionutility.1
This preview has intentionally blurred sections.
Sign up to view the full version.
Model:Suppose there areNgoods:Person chooses a vector(x1, x2, ..., xN).Reference point is a vector(r1, r2, ..., rN).Preferences are:Total Utility≡NXi=1[wi(xi) +vi(xi|ri) ].•wi(xi)is intrinsic utility for goodi.•vi(xi|ri)is gain-loss utility for goodi.How to formalize that gain-loss utility is directly tied tointrinsic utility:Assume there exists a “universal gain-loss function”μ(z)such that for each goodi, gain-loss utility isvi(xi|ri) =μ(wi(xi)−wi(ri) ).In general,μ(z)takes form of the Kahneman-Tverskyvalue function.We’ll focus on the special case:μ(z) =(ηzifz≥0ηλzifz≤02