This paper addresses risk-averse stochastic hub location problems where the risk is measured using the conditional β -mean criterion. Three variants of the classical multiple allocation hub location problem, namely the p-hub median, the p-hub maximal covering, and the weighted p-hub center problems are studied under demand data uncertainty represented by a finite set of scenarios. Novel mixed-integer linear programming formulations are proposed for the problems and exact algorithms based on Benders decomposition are developed for solving large instances of the problems. A large set of computational tests are conducted so that the efficiency of the proposed algorithms is proved and the effect of various input parameters on the optimal solutions is analyzed.