Say it's 2050, and we've built the first solar farm on the lunar pole. We build a transmitter on the moon, say 20-25 miles across, maybe 300 square miles of phased microwave arrays, made of tiled antenna.
We can build +99% of the array on the moon itself, though we'd have to ship up some advanced electronic components from Earth, mostly computer chips. Pretty reasonable though. 10 Starship launches, let's say.
This phased array pumps out around 400GW of microwaves to Earth, a fair amount of what is needed to power the US currently. (We'll likely lose 50-100GW as heat, so we need radiators or buried coolant lines to pump the heat into the lunar strata)
We'll have 100 rectenna on Earth, each picking up at most 4GW (roughly equal to a nuclear reactor output), each built next to a major city or industrial area.
At a 50 watts/m^2 average beam intensity at ground level, each rectenna will need to be about 15 miles in diameter, maybe 170 square miles. That's huge, but could be built over farmland and pasture.
I assume you know what a rectenna is, but for those who don't, it would be basically a grid of wires held off the ground, so the cropland underneath should be still useable. Think a few thousand poles holding metal wires 20 feet above a field. Really less noticeable than a modern solar farm.
From there it's just a matter of expansion, building more transmitters, more solar panels, and more rectenna. Eventually we'd have 100 transmitters, and 10,000 rectenna, powering every city on Earth from the Moon, even with losses.
This is all taking the Friis equation into account. It also doesn't require giant death rays of doom.
x 3 
