Bitcoin will have no problem scaling. Sure, the supernodes may end up being run by VISA, WesternUnion, AliPay, etc., but they won't have governmental monopoly and will compete on an even playing field without TBTF bailouts. The Paracoin idea seems similar to the idea of creating a new crypto that allows legacy transfers from Bitcoin at a 1:1 ratio. The legacy coins would always be supported by the backbone supernodes and exchanges, but would dwindle at the consumer level as the new crypto takes hold. There has been discussion about that contingency, but no real urgency to pursue it. Again, Bitcoin will have no problem scaling to VISA levels.
Paracoin would view Bitcoin mostly as a natural resource existing in the environment which might be exploited. The fattest and juiciest one at this point.
This is kind of like a tic on the back of a wild cow. If the cow gets domesticated and subsidized by a farmer (like Bitcoin to corp/gov) the tic does not care...the cow's blood is that much richer for it. Nor does the cow particularly notice or care much about the tic either since he is so tiny. Of course the farmer could fear bubonic plague and fumigate at great cost and risk. The tic can pack up and leave in that case.
The goal of 'paracoin' in my conception is just to explore a different block-chain path. By happenstance it would layer on all kinds of niceties which any Bitcoin users who wish to are welcome to use (anonymity being one.)
If the cow died or the farmer injected her with so many hormones that the user's didn't like the taste of her milk, Paracoin could split and probably take a reasonable amount of hashing power with it. Of course at that time hashing power will likely be highly monopolized by corporations. I would be always looking for ways to either use hashing power asymmetrically, or use a different 'proof of work' altogether. That part of the system would be designed to be 'plug-able' as much as possible.
As long as Paracoin is 'riding on the back of' Bitcoin, the whole world of economic balances and incentives is out of whack. This is regrettable, but if one is interested in that for the academic aspects of that, there are plenty of places to study it. The real-world impacts of various decisions regarding management of the solution would have to be dealt with using intuition and considered low precision...just as is/was the case with Bitcoin.