Norris Warns

new_york_times:http://www.nytimes.com/2008/11/07/business/07norris.html

By FLOYD NORRIS
Published: November 6, 2008
Some things are worth doing quickly, even if you might have done a better job if you had taken more time. Others must be done well, even if that means delay.
Putting together the financial bailout plan fell squarely into the first group. By the time it was enacted into law, the Treasury Department had changed its plans almost beyond recognition. What was a plan to buy dicey assets turned into a plan to invest directly in banks. Congress discovered it had approved something it did not know it had considered.
Fixing the financial and regulatory systems falls into the second category, and it is something that will be left to the Obama administration, as well as the next Congress. An effective plan will also include more international cooperation and agreement than has ever been seen before. We have seen the havoc that internationally linked markets and institutions can cause. Whether or not we wish they would vanish, they will not.
The details are not crucial now, but the basic approach should be. If Barack Obama has strong views on basic issues, he should make them clear to people being considered for high economic posts in his administration. If not, he should understand the views of those he is appointing, and make sure that he is comfortable with them and that his appointees are in basic agreement with each other.
The issue is far more complex than the ritual denunciation of deregulation that the president-elect employed on the campaign trail. Contrary to that rhetoric, there may well have been too much regulation of some markets and some activities of financial institutions — regulation that encouraged and accelerated the growth of a parallel unregulated financial system.
That system was allowed to grow with almost no oversight. Hedge funds could invest huge sums, and borrow even larger ones, with no disclosure of what they were doing. Credit-default swaps — which should have been treated as insurance contracts — could grow with no required reserves and no regulatory attention to assure the sellers could meet their obligations.
One principle of the new regulatory system must be that regulations are based on what people do, not on who they are. If two products accomplish the same objective, they should be regulated the same way. Hedge funds that raise money from pension funds should not be allowed to wave a magic wand and be treated far differently from other institutional investors.
There should not be one regulatory system for financial products regulated by the Securities and Exchange Commission, another for similar products under the jurisdiction of the Commodity Futures Trading Commission and a third for products provided by companies that have the word “insurance” in their names.
It may be tempting to seek a return to the separation of investment banking from commercial banking. But it is too late for that. Integrated companies are here, and must be dealt with.
Perhaps most important, regulators must take responsibility for doing their job, rather than subcontract it to the regulated. As new financial products were developed, the regulators threw up their hands and allowed the banks to apply their own (supposedly) sophisticated risk models, or to rely on bond rating agencies. We saw what happened in mortgage-backed securities.
One of the big problems with those securities could be called the “Russian roulette” issue. The probability of a disastrous outcome appeared to be so low that it was ignored in the models used by the issuers and raters, and the investors were happy to rely on them. But even a low probability event may represent an unacceptable risk. Few of us would play Russian roulette, even if the odds were wildly in our favor, because it is a game no one can lose twice.
A new division at Standard & Poor’s, the market, credit and risk strategies group, has begun analyzing the products that S.& P. used to rate highly, and has issued a report that illustrates one part of the dilemma. The study takes a more or less typical mortgage securitization sold at the peak of the market, and applies different, but still reasonable, assumptions about the future.
“It is amazing how fast you can go from having something that has a very high value to one with a very low value,” said Michael G. Thompson, the group’s managing director.
Mr. Thompson pointed to a security, originally rated AA-plus, that would now be worth 76 cents on the dollar if you assume that 20 percent of the underlying mortgages will default and that losses on those mortgages will be 35 percent of the amount owed. Raise each of those figures by just five percentage points, and the security’s estimated value falls to 9 cents. But the value of another security, this one previously rated AAA, from the same securitization would fall only to 85 cents from 86 cents under the same set of assumptions. And the first security’s value would rise by only a penny if you applied more optimistic assumptions.
That is because these securitizations were structured so that one security would suffer no losses until all those below it were wiped out. At that point, even a few more losses can be devastating.
Those numbers are crucial in understanding the current battle over mark-to-market accounting, which the banks blame for much of their problem. They argue that even when current market prices are available, they are obviously unreasonable because they are so far below the values produced by the bank’s own models.
As it is, banks now are using models to measure about $600 billion of assets, Mr. Thompson said, adding that better disclosure is needed to allow investors to assess what assets the banks hold and whether the values shown are based on optimistic or pessimistic assumptions.
There are other regulatory principles that should be adopted. One is that regulators should review the structure of new products before they are widely distributed and before they become a significant part of the portfolios of regulated institutions.
Such a review should focus on the risks that need to be disclosed and understood, and on the risks that the new securities will create. Mortgage securitizations led to a system where the lender thought it need not care if the loan was repaid, and to a situation where it was much harder to restructure a troubled loan, given that such a restructuring could be advantageous to one part of the securitization and not to another. If the private securitization market ever does come back, steps should be taken to deal with those problems.
But the most important decisions will come after the new structure is adopted. The regulators must be given enough resources to do their jobs, and those appointed to do them must be people who believe in the regulatory system. Former investment bankers may not be the best choices for hiring as watchdogs.
The regulatory system has never been a top priority of a new president, or at least not since Franklin D. Roosevelt had to invent a system during the Great Depression. Mr. Obama’s choices, like those of Mr. Roosevelt, may be with us long after he is gone.
Floyd Norris’s blog on financeand economics is at nytimes.com/norris.