First, Paul Romer confirms some of my views when I wrote "The Chess Game of Financial Regulation." That is, eventually, all fixed rules will be gamed. He calls this "Myron's law" in deference to Myron Scholes.
Second, Sheila Bair says that big banks are not more valuable franchises. She points out that, if anything, there are diseconomies of scope. The idea that a "financial supermarket" is a great thing was promoted by Citigroup for years. But Bair points out that the more valuable franchises are focused shops.
In the most extreme scenario, where a temporary liquidity shortfall at a central counterparty has the potential to cause systemic disruption or even threaten the solvency of a central counterparty, it is likely that central banks in major jurisdictions will stand ready to give whatever support is necessary (and recent regulatory proposals suggest that the Fed and ECB will do so). However, such an arrangement creates moral hazard - and a roundabout way for derivatives risk to be picked up by taxpayers.
Pointer from Mark Thoma. Read the whole thing. There are folks whose claim to expertise in finance is that they advocate a centralized exchange for derivatives. Long-time readers of this blog know that I think that it is a futile idea. Centralized exchanges deal in options that are close to being at-the-money. Derivatives are deep out-of-the-money options. These behave quite differently from market-traded options, and the fact that so many self-proclaimed expert regulators do not understand that is frightening.