Running a power cable across the Pacific is going to be rife with expensive engineering problems, but sure, you could send electrons from Three Gorges to Google at some cost. I wouldn't call it cheaper, though.
Unless there are massive improvements in superconductive materials, of course.
If saharan electricity is the cheapest then nobody else can compete in hydrogen production.
[1] https://en.wikipedia.org/wiki/Electric_power_transmission
[2] https://www.wolframalpha.com/input/?i=7079+MWh+%2F+265Wh%2Fk...
It's energy intensive, and uneconomic - which is why people are stringing HVDC links across bodies of water - Denmark/Sweden, Australia/Tasmania, NZ North/South Islands instead.
https://en.wikipedia.org/wiki/Fischer%E2%80%93Tropsch_proces...
Converting it back-and-forth to electricity isn't easy nor very efficient, though. Hydrogen is a bit easier, although still lossy. I'd say it's doable, and not as bad as it sounds, if you really have excess production.
inb4 compressed air supertankers.
That said, it will in most cases be less expensive to use it as close to the originating point as possible if you can find a need for it.
Not to mention global annual battery production amounts to less than 1 hour of the USA's national electricity usage.
(I would be surprised if this was viable even with my suggested measure, no argument there!)
I absolutely accept that this is a poor solution, and that I am quibbling over a factor of perhaps 500 at most when it’s short by a factor of ~250,000.
But what about electrifying a small collection of households by batteries that are replaced from a truck every N weeks, instead of building many miles of power lines through (perhaps flammable) wilderness?