Cache-enabled heterogeneous networks (HetNets) have recently emerged as an attractive solution to meet the exponentially increasing demand on mobile data traffic. However, the power consumption and the backhaul limitation of small base stations (SBSs) have become bottlenecks to deploy HetNets. How to relieve the burden of backhauls via wireless caching and enable the HetNets to operate in an energy-efficient way are still open issues. Aiming to minimize the power consumption while guaranteeing QoS requirements of users, in this paper, we address the problem of joint user association (UA) and resource allocation (RA) for coded cache-enabled HetNets. First, based on the many-to-many matching game between the virtual SBSs and users (VSU), we propose a low-complexity joint UA and power allocation (PA) algorithm (JUPVA). Then, considering the unequal BA, we design a three-phase optimization algorithm (JURVA), which makes a joint decision on UA, PA, and BA iteratively. The simulation results demonstrate that the proposed algorithms yield significant performance improvement in terms of power consumption.