This seems purely driven by profit. Retrofitting cell towers has a ROI that is inverse to density of customers willing to pay (broadcast distance is limited). W

This seems purely driven by profit. Retrofitting cell towers has a ROI that is inverse to density of customers willing to pay (broadcast distance is limited). Whereas if fiber has already been run and good Ethernet cables already laid, increasing total bandwidth for wired connections has a cost ~proportional to paying customers.