There's ample evidence that many users aren't aware of the costs of using NAT, or especially, weren't aware of those costs before they started using NAT - so their choices were poorly informed. So no, it's not reasonable to conclude that decisions to use NATs are justified by realistic cost-benefit estimations of doing so.
NATs have been around for quite a while. This might have been a convincing argument 5 years ago, but I don't find it very convincing now, particularly in view of the fact that some people who clearly understand the cost/benefits choose to use them.
NATs have been around for awhile but not everyone has been using them for 5 years. Some of those who thought NATs were a good idea are now looking for alternatives. For other people, NATs work just fine for their current needs, and this has never been disputed.
Note also that cost optimization by individual users (even ifOf course. But then you have to describe the negative externality.
well-informed) does not necessarily produce a cost-optimized result for
the overall community.
plenty of examples have already been given.
Keith