“…For example,Harberger (1971),Willig (1976),Tirole (1988, pp. 7-13),Weitzman (1988),Vives (1999, chapter 3),Schlee (2013), andHayashi (2017).24 Unfortunately,Khan and Schlee (2017) omit explicit mention of this last assumption.25 The Finetti-Fenchel-Kannai examples highlight how a convex preference relation may not have a concave representation for precisely such plausible preference orderings; seeKannai (1977) for some examples and the antecedent background.26 If the consumer has expected utility preferences and we interpret u as a von Neumann-Morgenstern utility, this fact implies that aversion for wealth risk follows from aversion for consumption risk; seeKreps (2013, Proposition 6.16, p. 136).27 For two continuous real-valued functions f and g on a convex set, D ⊆ R n that represent the same binary relation , g is less concave than f if there is a convex, and strictly increasing, real-valued function T on Range( f ) that is convex with g = T • f .28 For what it is worth, it was reflection on Samuelson's assertions about local concavity of money metric that led us to think about what global properties of concave functions, if any, that it possesses, independently of differentiability and interiority assumptions, and led us thereby to saddlepoints.…”