This special case of the "Kreps-Porteus utility"aggregates the role of the conditional distribution of future consumption through an "expected utility of next period's utility."If h and J are concave and increasing functions, then U is concave and increasing.If h(v)=v and if f(r,y)=u(x)+By for some u:R+>R and constant B>0,then (for Vr+1=0)we recover the special case of additive utility given by U(c)=E "Non-expected-utility"aggregation of future consumption utility can be based,for example,upon the local-expected-utility model of Machina [1982 and the betweenness-certainty-equivalent model of Chew [1983],Chew [1989], Dekel [1989],and Gul and Lantto [1990].With recursive utility,as opposed to additive utility,it need not be the case that the degree of risk aversion is completely determined by the elasticity of intertemporal substitution. For the special case(7)of expected-utility aggregation,with differentia- bility throughout,we have the utility gradient representation T=fi(c,Eh(V+1)f五(cs,E,[h(V+1)E,N(W+i小， s<t where fi denotes the partial derivative of f with respect to its i-th argument. Recursive utility allows for preference over early or late resolution of un- certainty (which have no impact on additive utility).This is relevant for asset prices,as for example in the context of remarks by Ross [1989],and as shown by Skiadas 1998 and Duffie,Schroder,and Skiadas 1997.Grant, Kajii,and Polak [2000]have more to say on preferences for the resolution of information. The equilibrium state-price density associated with recursive utility is computed in a Markovian setting by Kan [1995].6 For further justification and properties of recursive utility,see Chew and Epstein [1991],Skiadas [1998,and Skiadas [1997.For further implications for asset pricing,see Epstein 1988,Epstein 1992,Epstein and Zin [1999,and Giovannini and Weil [1989] 6Kan 1993]further explored the utility gradient representation of recursive utility in this setting. 11This special case of the “Kreps-Porteus utility” aggregates the role of the conditional distribution of future consumption through an “expected utility of next period’s utility.” If h and J are concave and increasing functions, then U is concave and increasing. If h(v) = v and if f(x, y) = u(x) + βy for some u : R+ → R and constant β > 0, then (for VT +1 = 0) we recover the special case of additive utility given by U(c) = E " X t βt u(ct) # . “Non-expected-utility” aggregation of future consumption utility can be based, for example, upon the local-expected-utility model of Machina [1982] and the betweenness-certainty-equivalent model of Chew [1983], Chew [1989], Dekel [1989], and Gul and Lantto [1990]. With recursive utility, as opposed to additive utility, it need not be the case that the degree of risk aversion is completely determined by the elasticity of intertemporal substitution. For the special case (7) of expected-utility aggregation, with differentia￾bility throughout, we have the utility gradient representation πt = f1 (ct, Et[h(Vt+1)])Y s<t f2 (cs, Es[h(Vs+1)]) Es[h0 (Vs+1)], where fi denotes the partial derivative of f with respect to its i-th argument. Recursive utility allows for preference over early or late resolution of un￾certainty (which have no impact on additive utility). This is relevant for asset prices, as for example in the context of remarks by Ross [1989], and as shown by Skiadas [1998] and Duffie, Schroder, and Skiadas [1997]. Grant, Kajii, and Polak [2000] have more to say on preferences for the resolution of information. The equilibrium state-price density associated with recursive utility is computed in a Markovian setting by Kan [1995].6 For further justification and properties of recursive utility, see Chew and Epstein [1991], Skiadas [1998], and Skiadas [1997]. For further implications for asset pricing, see Epstein [1988], Epstein [1992], Epstein and Zin [1999], and Giovannini and Weil [1989]. 6Kan [1993] further explored the utility gradient representation of recursive utility in this setting. 11