ELEMENTARY MY DEAR...

So I slept on it...and slept and slept.

Got a call from a buddy

"You were wandering in you thoughts."
"No *&^% Sherlock."
"It's elementary, really."
"Oh is it?"
"Yes of course it is. Your point is that if you leave it to machines and formulae you ignore the greatest human strength."
"Which is?"
"What is found only in humans but often lacking."
"Common sense?"
"Of course. As I said, elementary."
"*&^%!"

He was right of course. It matters little who or what is appointed or designated as the overseer of risk. It all comes down to the INDIVIDUALS performing the task and their experience, not only with the area, the products but with the people involved. Let me explain:

A baseball manager with a man on first and needing a run to tie the game may well leave it to the runner as to whether he attempts o steal second based on the runner's experience and past performance. A risk manager may well, within agreed maximum perimeters, allow considerable more leeway to one individual than to another. In a sense, the first law of banking, "know your customer," is applied to the most important level of action within the institution, the art of taking risk. BUT, this requires substantial experience and knowledge; the risk managers within an institution must be among the most professional and knowledgeable of all employees. I am afraid that in some cases that is not always the case or that risk management, while deemed important, is actually regarded and compensated as a lesser job where no rain makers are found. Profit and the making of the same is all important.

Another problem is easily found upon inspection. There is often an enormous amount of pressure to, in all cases, get the deal done. This leads to the close calls that most often result in the problems that rise up only at a later date. Most deal-doers are convinced that they can structure their way around a problem and will go to great lengths to convince risk management that the issues raised have been resolved but their structuring brilliance. In some cases this is true; in others the result is layer upon layer of complexity that in itself creates addition risk that should never be undertaken. Sometimes the ultimate reward is worth the additional risk. Sometimes. The problem, however, is that too often every risk is deemed the ultimate deal. They never are. Intelligence and EXPERIENCE is what makes a really good risk management team.

There is another problem that is not often seen or recognized. Risk managers are like all other humans; they like to be loved and wanted. Saying "no" to colleagues with whom one shares a corporate existence or setting parameters which limit the taking or risk (and hence, profit and pay) is a difficult existence. A simple example from my past. I was asked to approve a simple risk on an interest rate swap for a western hemisphere country with whom we had little business. It was a one-off transaction that had no business being done for a variety of reasons (not the least of which was that the country was on the wrong side of the swap given the direction of interest rates). I approved the transaction nevertheless primarily because the risk to the institution was small but if I were to be honest because I liked my colleagues and wanted to help. This happens all the time. The desire to be seen within the institution as being "constructive" is a huge motivator as well. At the end of the day, all went well except for the country who would have been better off by not doing anything but as a wise man once said, "do not mislead a client, but if he is determined to make an honest mistake on his own that benefits you, accept the gift." Make your own call on that little gem.

In any case I think the function of risk management in financial institution should be as divorced from the general management line as possible. Risk managers should be independent, reporting to, perhaps, a sub-committee of the board and directed by a separate line of management. I would not at all be opposed to external direction of some sort. In the good old days in London, the Bank of England used to assign a retired senior British banker to newly opened foreign institutions in London just to make sure things were done the "right" way. Not a bad idea. For those institutions deemed to present a "systemic" risk (you know who you are) a cadre of risk managers in specific areas might be assigned by the regulator from a pool of proven professionals trained in risk management either independently or at oversight institutions, and their reporting line would be to the institutions not the regulator. Their pay, given their environment would be commensurate with the private sector levels in the institutions to which they are assigned. Remember, you pay peanuts, you get monkeys. They could even be moved between institutions after a period of time to insure independence. On Monday, J.P Morgan could have a risk manager who worked for Citibank on the previous Friday. Now wouldn't that be a kick? Think about it.

Coming up tomorrow, a little discussion on our girl friend Ms. Bair who, moving with her usual grace undoubtably planted the story on Citibank management problems in the WSJ last week, and other Potomac follies to which we have alluded. In the mean time the Supremes allowed the stay in the Chrysler bankruptcy this afternoon. Could it be that Ruthie, a lovely Yiddisha girl with her female experience came to a better decision than Rahm, a nice Yiddisha boy? We shall see.

...AND TO CONTINUE

We talked last week about how it was that everything went so badly wrong and promised to discuss what might be done to prevent such a mess from occurring again. In regard to CDSs, the solution (if there is one) might be fairly simple and one without a good deal of disagreement. What set the dire events in motion was the total lack of knowledge on the part of the participants as to the volume of the market and the exposure of the individual participants, not only in aggregate but to one another. There were intelligent and in some cases accurate guesses to be sure, but real time information was woefully lacking due to the fact that there was no central body through which transactions were cleared; the business was conducted as they say in the trade in an over-the-counter market...companies delt directly with other companies directly on their own books with no one keeping score.

CDSs were considered by some (many?) to be just another derivative--a creation whose value is based on something else. Derivatives have been around for some time starting with simple interest rate swaps--you pay me a steam of money on a fixed rate basis and I pay you a stream...based on the same principal amount...on a floating rate basis and VOILA, we have created a fixed rate obligation out of a floating rate one thereby transforming the underlying debt into one or the other. Cleaver and useful, no? Now there is a chance that on side may get it wrong and pay a floating rate during a period of rapidly rising interest rate but the loss suffered will never be the principal amount; it will only be the difference in the cost of money during the period that the transaction is outstanding which in most cases will be only a small percentage of the nominal amount. Not so with Credit Default Swaps. That, my friends, can be a zero sum game for if the underlying obligor defaults, the loss can be 100% for the poor slob who wrote the contract.

AIG and others wrote A LOT of contracts about which no bod ee knew nuttin'. Think of the different result if all of there transactions would have been through a central clearing house. Three things would have been accomplished:

1. Counterparties would have been revealed
2. Outstanding exposures would have been revealed
3. There would have been a hell of a lot fewer contracts written due to the now-revealed exposure which would have tempered the zeal for dealing with companies that were perceived as becoming overexposed

I suspect there will shortly be a clearinghouse for transactions of this type, as there should be, for there is very little disagreement among regulators as to its need.

As to other forms of derivative transactions, there seems to be apparent agreement among regulators around the world that some additional forms of control are required. However, there is yet to be agreement as to who the regulators are to be and how the regulation is to be accomplished. If the institutions involved are banks, the control is fairly simple: bank regulators have long controlled the businesses in which banks participate through reserve requirements; take this risk and you have a reserve requirement of x; do that, x + y; do THAT, (x + y)2. But if they are not banks like AIG? Ah ha, a subject for another day as well as a discussion on rating agencies.