[Wonder Warthog]

"...since the light converted in orbit would have been intercepted by the Earth anyway -- and actually cooling -- because the best available photovoltaics are only 30 per cent efficient."

Not quite true. Cooling would only occur if the "sunsats" were stationed between the earth and the sun. More than likely, they will not be so stationed, so the net effect WOULD be some slight heating of the planet, as the total energy would be that normally received from sunlight plus the added energy from the sunsats.

[SunkenCiv]

Yeah, I suppose that could be true -- because they would be in geostationary orbit, I didn't think about how they could be turned to face the Sun near dusk and dawn, and after sunset but before sunrise. Even though 100 per cent of the eclipsed sunlight wouldn't reach the Earth (in the shadow of the device tracing a path across the surface), and only 30 per cent (at best; minus the transmission losses which would occur) would arrive as beamed electricity after the shadow had crossed the surface, there could be no net effect, or there could be slight heating.

OTOH, if it were possible to place a single large satellite on a path where it would receive sunlight all the time, or nearly so (polar orbit?), it would be necessary to build a series of the receiving dishes (six miles across, each) at various places to service various markets around the world. Problems abound, even without considering the maintenance of the satellite components in the steady rain of miles-per-second grains of sand. :')

Energy Savers: Solar Power Satellites (DOE)
Solar Power Satellites: Environment And Health ~1749k PDF
Appendix D: Environment And Health~255K PDF