The increasing penetration of intermittent renewable production requires more flexibility in the power system. We propose priority service and multilevel demand subscription as two alternative methods for the mobilization of residential demand response. Whereas priority service relies on the differentiation and nonlinear pricing of electricity according to reliability, multilevel demand subscription further differentiates electricity service according to duration. We design a priority service menu as the equilibrium solution to a Stackelberg game, which is modeled as a bilevel optimization problem involving a vertically integrated utility and consumers. To tackle the computational challenge brought about by introducing scenarios of renewable production, the model is decomposed by the alternating direction method of multipliers and solved on a high-performance computing infrastructure. We then extend the modeling approach to multilevel demand subscription and priority service is treated as a special case. We propose a framework based on stochastic programming for quantifying the trade-off between a complex but efficient multilevel demand subscription pricing menu and a simpler but less efficient priority service pricing menu. Case studies on the Belgian power system demonstrate the effectiveness of the proposed pricing schemes.