I agree with the synopsis of the presented theory. I also agree that it would be prejudicial to assume that attractive women are less intelligent and capable than their less attractive counterparts. However, I don't see how that assumption is baked into the theory.
Even if, on average, the more-attractive candidates were more intelligent and capable (purely for the sake of the argument), the theory presented seems to support the notion that, despite the hypothetical averages, there would still be a pool of less attractive, highly-qualified candidates held back. It seems to me that, if that were true, they would, under the presented theory, be replaced by attractive candidates who in fact do not qualify for upper management.
Maybe I'm missing something. I don't see why the use of this service would produce an increase in that phenomenon beyond what it is now. I tend to think this would merely swap encrypted versions of commonly-used passwords for the originals.