A lot of FSF advocates don't realize how toxic the GPL has become outside of the FSF bubble. GPLv3 software in particular is a now non-starter for a lot of companies and has undermined a lot the goodwill businesses had towards GPL software.
The shift towards BSD, Apache, and other open source licenses is a symptom of that and unless the FSF wises up, support for GPLv3 projects will dry up.
I would very much like to see such calculations so the market could evaluate if it is correct. Maybe see what a insurance company would ask in return for covering it, as that is their domain of expertise. So far there has not to my knowledge been a single lawsuit over GPLv3 specific requirements, which is something I think a insurance company would consider. Good money with a risk that has never ever occurred is something they might like in a deal.
Oh, I completely agree with that. I'll never forget a particularly frustrating conversation with a client who refused to allow access to their database, because we were using 'open source code that anybody could have changed'. That sort of thing – banning open-source software from your systems – is silly.
However, you've got to consider the default-conservative attitude of larger companies. In Apple's case for example, the breaking point seems to have been GPLv3 - this was when they started reimplementing things like Samba and removing copylefted software. And that's because the GPLv3 represents a real business risk – aside from the direct patent implications, litigation under it is sparse, and businesses don't like operating in unsure legal territory!