Perhaps metawatts is the correct term; it was late at night and the subconscious could have had the upper hand:). I don't know if we actually need a 10 megawatt electrical generator on the Moon, in the immediate future, the way I see it, that would produce a ton of rocket fuel per hour, plus leave lots of power for other things. But if a 10 megawatt generator weighs is only 20 tons, then this would be useful number to help determine lunar economics. If the total cost is say 1 billion dollars- to make generator and deliver to lunar surface, in 6 months of operation it could produce about 36,000 megawatts, so that means that for around $30 a kilowatt, you pay for most of your capital investment, and within about a year you are near to turning the corner in profitable- assuming you have enough demand at $30 per kwh. $30 per kwh is a good price for electrical power for anywhere in Space. But what is more likely is smaller generators- probably super light weight solar panels, and with a starting price of around $50-100 per kwh with rocket fuel production somewhere around a ton per day. Once demand is is firmly established, then shifting to nuclear and/or solar panel production on the Moon would make more sense. I think that if you reach a point where power was selling for about $10 per kwh, you would reach a tipping point, and within a decade or so you would have power being produced so cheaply that it would be shipped to Earth.