I think the implied comparison was: do we launch panels to the moon or build them in-situ? The increased solar exposure should boost both of these cases by the same percentage, but if the Earth-manufactured panel is (e.g.) 3x as efficient, it might still make more sense to ship them.