The problem boils down to "Really I am better at programming than this business stuff.".
Someone with his talent and expertise should have no problem with getting highly paid consulting gigs. Then he could continue working on GPG in his free time and even use the consulting income for hiring additional programmers to work on GPG. There are quite a few product-based businesses that could be built upon GPG as well (secure email, corporate communication tools, some kind of public-key-based social network come to mind ...). These could be used to support the continued development of GPG itself.
It's of course not as easy as it sounds. Not everybody wants to deal with 'all that business stuff' and that's fine but then by all means find someone who can help you with that part. If you want to change the world sometimes idealism alone just isn't enough. You also have to proactively deal with the everyday stuff like where the cashflow for paying the bills will come from next month.
There's also a problem with the purism put forth by some of the 'free as in freedom' enthusiasts, most notably Richard Stallman, who seem to gloss over the fact that coders have to make a living, too or who even frown upon making money with software altogether. Software eats the world but even RMS can't eat software.
How many successful larger companies come to mind whose business model is based upon open source? Red Hat, Ubuntu and that's about it. If we truly want to avoid dilemmas like this one we also need to think about how to successfully implement sustainable open source business models.