All electric power plants (excluding hydro plants) have the problem that they cannot ramp power production up and down with demand over the 24 hour cycle. So, during the dead of night a lot of power goes to waste. If this power can be captured and stored for later use, it would help us conserve fuel and the environment. That would be the logic behind Laherrere's idea.
Making hydrogen from electricity is not particularly effective (30 - 35% efficient), but I guess 30% is better than 0%. Other ways to store the unused electricity including pumping water from a low reservoir to a high one during off peak hours and releasing it from the high reservoir to the low one (to generate power) during peak hours. You obviously need the right terrain for this to work and a substantial upfront investment.
Laherrere's other idea, to convert the hydrogen to conventional petrochemicals actually makes a lot of sense, in principle. I am amazed that the US Energy Department is seriously discussing hydrogen as a fuel of the future. In the post 9/11 world, why would you promote a high explosive as your fuel of choice?
If this hydrogen to petrochemicals process would use CO2 as a feedstock, you could be addressing global warming at the same time. If you cannot use CO2, you are back to using a fossil fuel, and what would be the point of that? In practice, however, I doubt there are any existing processes for combining hydrogen and CO2 to yield petrochemicals.
It looks as if one can produce methanol from CO2 and hydrogen, see
http://sschi.chtf.stuba.sk/konf2003%5Ca ... %5C117.pdf, although the process looks like it is still in the lab.
Perhaps a more elegant solution is the direct electric fixation of CO2, see
http://www.ikm.org.my/Journal_articles/ ... alimon.doc. If these guys are to be believed, you can convert CO2 to methane @ 80% efficiency! Now there is a way to store that unused electrical power...