Plug-and-play femtocells will be an integrating part of future cellular networks. Resource management and interference mitigation become challenging, suffering from severely delayed network control in large-scale deployments. We propose a new game theoretic framework, where fast interference suppression is decoupled from the relatively slow frequency allocation process to tolerate the delayed control. The key idea is to cast femtocell clustering as an outer-loop evolutionary game coupled with bankruptcy channel allocation, which drives the cells to spontaneously switch to less interfered clusters. Within each cluster, we design an inner-loop non-cooperative power control game, such that the requirement of prompt control is eliminated. The two loops interact recursively with analytically confirmed stability. Simulations show that our framework can improve the throughput by 13.2% in a network of 200 cells, compared to the prior art. The gain grows further with the network size.