The expansion of distributed solar necessitates additional research into the impacts on both utilities and their customers. In this paper we use New Jersey solar data, PJM market data, and demand profiles from a PJM utility to investigate rate and bill impacts of large-scale solar penetration. In addition to the subsidization of solar adopters by non-participants, we highlight the channels through which cross-subsidization of rate classes can arise in practice. The results of our study indicate that the fear of a utility "death spiral" may be exaggerated. Significant solar can be incorporated with only a 2% increase in non-participant bills. At high levels of penetration, distributed solar has the potential to alter the system peak hour which affects the allocation of costs across rate-classes. As the peak hour shifts to the evening when solar production diminishes, residential customers face higher distribution costs. Policy makers and utilities need to be aware of these challenges in designing the next generation of rates that are better aligned with cost causality.