You might be able to squeeze some efficiency out of RC delegation pools, but note that the overall cost of performing RC tasks will rise to counter balance any increase in efficiency of use.

In other words, the reason why RC costs are so 'cheap' (other than the arbitrary multiplier constants being set that way) is that most of the larger stakeholders aren't efficiently using their RC. As RC pooling become more commonplace, costs for performing common tasks like posting and comment will likely also increase considerably.

At a higher level, I feel here you're offering a free market solution to a scaling problem that heavily depends on subsidizing. The free market is great at a lot of things but acting as a subsidy isn't really among them, so the vision of having projects buy or rent their RC as needed is likely not as scalable as what we'd ideally like.

However, certainly Inc will likely delegate their RC very generously for good causes, and that's definitely a benefit.

In theory, you'd probably want a set of rules that treat RC as a scarce public resource rather than a private transferable good. Ultimately, I feel true scaling solutions will have to come from technical improvements to the design and infrastructure of the protocol rather than managing bandwidth.