"Don't say 'If only,' say 'next time,'" goes the familiar saying. But there's a problem in learning from mistakes. If we're not extremely careful, the lessons we draw may create even greater problems. The historian William H. McNeill, one of my graduate teachers, coined a brilliant phrase for the phenomenon, one unfortunately even more apt now than when I cited it in Why Things Bite Back: the Law of the Conservation of Catastrophe. It suggests that "every gain in precision in the coordination of human activity and every heightening of efficiency in production [is] matched by a new vulnerability to breakdown." Safety and security measures, for example against forest fires and terrorism, may succeed at first but also set natural and human processes in motion that may lead to even greater disasters in the future. Building dams and levees produces more serious flooding; suppressing small forest fires helps accumulate fuel for greater infernos.

In finance, too, safeguards can over time become dangerous. For many of Bernard Madoff's victims it might have been better to have no Securities and Exchange Commission than one that could not discover the fraud even after persistent complaints. While the full story may take years to tell, it's likely that Madoff used his good standing with the agency to make the Ponzi scheme harder to detect. The journalism professor and Ponzi expert Mitchell Zuckoff has even speculated that tax regulations requiring a five percent payout of foundation assets attracted philanthropists to the steady returns that Madoff appeared to offer, another unintended consequence of apparently progressive rules.

So is "a centralized, unitary financial-intelligence apparatus in government that would have complete and continuous access to the books of all financial institutions"  the long-term answer? Possibly, but even if such a monitoring system would work and could be built at an acceptable cost, its software would instantly become a target of espionage and terrorism, needing constant upgrades and fixes, each of which could introduce new vulnerability. And even without malicious attacks, the search for honest anomalies and loopholes would have unforeseen results, just as Congress's 1993 cap of $1,000,000 for tax deductions of executive salaries produced even more excessive, and dangerous, options-based compensation.

The Conservation of Catastrophe is only a tendency, not destiny. But it does suggest any reform of complex systems will take exceptional vigilance. Or as Lewis Carroll's Red Queen put it in another great phrase I quoted, "it takes all the running you can do, to keep in the same place."