The article directly and elaborately addresses this point, advancing the argument that renewables are scarce vis-a-vis our urgent need for them and this isn't the best thing to use them on.
Your attempt at summary is selective; I'm glad I read the piece instead.
His estimation of how much energy is "per transaction" is flawed, due to complete disregard of available scaling solutions today.
He essentially equates a central bank transaction to a penny left in a tipping jar, that's the general standard of rigour in this article.
Bitcoin just demonstrated renewable energy cannot scale in practice.
If only 1% of the electricity demand cannot be sustainably and economically be supplied by renewables, some serious re-thinking about strategy is in order.