Where is the energy saved in cuprate superconductivity?

I first raise the question: "In what regions of wave vector and frequency does the saving (?) of Coulomb energy associated with the superconducting transition in the cuprates predominantly take place?", and point out that in principle, and probably in practice, it can be answered quite directly by differential EELS measurements. I then conjecture the answer "In the region of long wavelengths and mid-infrared frequencies", and show that, if true, this hypothesis explains, inter alia, the systematics of T_c in the cuprates as a function of the layering structure.