In March 2008, after conducting a year-long process of study, I put forward a series of comprehensive recommendations to modernise our regulatory architecture in the Treasury's Blueprint for a Modern Financial Regulatory Framework. The blueprint identified an optimal structure that was not designed to be accomplished overnight.
Note the word optimal. My theory is that we always have an optimal financial regulatory structure--for the previous cycle's financial system.
That is, think of financial markets as going through cycles of euphoria, crash, and recovery. Give each round of the cycle a number. For example, the 1930's might be cycle number n. Then the S&L crisis would be n+1. The current crisis would be n+2 (if you ignore other events, like the Penn Central commercial paper crisis of 1970 or the August 1987 stock market crash or Long Term Capital Management or the Dotcom bubble).
Note that after crisis n in the 1930's, we created an optimal mortgage finance system, in the sense that we no longer had short-term balloon mortgages. Instead, we created the S&L industry, with a mandate to issue thirty-year fixed rate mortgages. We created an optimal system to prevent bank runs, with deposit insurance, deposit interest rate ceilings, and other regulations.
Guess what crisis n+1 consisted of? S&L's going belly-up, because the fixed-rate loans were under water in a high-inflation environment. The deposit insurance system was exposed to be badly flawed, at great expense to taxpayers.
Crisis number n+1 was exacerbated because the mortgage loans were not marked to market and because capital requirements were not tied to risk. So we created an optimal regulatory system, consisting of risk-based capital regulations and market-value accounting.
Guess what crisis n+2 consisted of? "Toxic" assets created to satisfy risk-based capital regulations (the infamous AAA-rated securities were just what risk-based capital was supposed to encourage), and a vicious spiral caused by market-value accounting, which made everybody have to sell illiquid assets at once.
Suppose Henry Paulson and all of the other "experts" get their way, and we create another optimal regulatory structure that would have prevented crisis n+2. Can you guess what crisis n+3 will consist of?