Monday, September 14, 2009

Step 1: Rating Agency Reform

Anybody who’s seen the movie Charlie Wilson’s War will appreciate the fact that signs of economic improvement doesn’t mean we can take our foot off the peddle.

The film describes Charlie’s (ultimately successful) efforts in leading Congress to support Operation Cyclone, the largest-ever CIA covert operation, which supplied weapons to the Mujahideen during the Soviet war in Afghanistan. But the movie ends on a somber note, citing the U.S.’s premature exit in the epitaph to the film:
“These things happened. They were glorious and they changed the world ... and then we f----d up the endgame.” – Charlie Wilson
(The result: the Mujahideen eventually flowered into the Taliban and backed Osama bin Laden's war against the U.S.)

Our situation is hopefully not as drastic, but the point remains: even if we feel we have survived the crisis -- and that unemployment and default rates are leveling off -- we still need to implement the necessary reform measures to avoid the creation of a different monster in future.

With this in mind, we’ve put out our first paper on rating agency reform. The piece ends:
“The ultimate objective of this reform is to encourage financial market transparency and responsibility, from which liquidity will inevitably follow.”
Read it; criticize it; share your thoughts.

- PF2

1 comment:

Sane Person said...

Interesting article but it would be nice to hear your thoughts on potential solutions to these problems.

1) Decrease proliferation of ratings across our financial regulatory structure

Obviously, we've seen the negatives associated with this. But what is the alternative? You can't have a bank self-rate its positions for regulatory purposes. Are you suggesting the regulatory body institutes its own ratings agency? Or perhaps this is a reference to a previous EL article about "regulatory ratings" which seems to suggest that there should be some kind of set in stone guideline for rating assets that applies across all rating agencies. Unfortunately, such an idea marginalizes the ratings analyst to such an extent that it leads directly to a "race to the bottom" where a now totally commoditized product leads to price wars which leads to it no longer being cost effective to apply anyone with actual experience to the process.

2) Decrease dependance and over-reliance of market participants on ratings accuracy

Are you seriously espousing the virtues of a nanny state, here? While we're at it, why don't we create a government agency to monitor hedge fund trading floors? Before a trader decides to go long for a day trade, he has to give an oral presentation to a government committee describing the motivations and risk factors associated with his oncoming trade. If anything, this section should be about ramifications for market players when they fail. Why do people blow up funds/CDOs/whatever and get right back on the horse with new funding and new jobs? Is it about lack of investor education into the track record of the people they're investing with? Nanny states are very good at background checks - perhaps they should run one for everybody managing money.

3) Incentivize ratings accuracy and improved transparency of methodology

While incentives at ratings agencies are most probably out of whack and they've certainly done themselves no favors, ratings NEED to be looked at as a represenation of a probability distribution, not a prediction. Just because a Aaa defaults doesn't mean the Aaa rating was wrong, per se. Ratings are also not market barometers. Just because CCC loan prices are rising and therefore OC levels are rising doesn't mean that, suddenly, the agencies should rush back and re-rate every single CDO out there. It's silly. If (and, granted, it's a big if) the agency has done their job well, this recent rally in loan prices should have no effect on the vast majority of outstanding ratings. A probability distribution, not a prediction.

4) Create mechanism for methodology and quality oversight

While the goal of having high quality methodology and ratings is obvious, this section is somewhat vague beyond the example of massaging historical performance. Are you hoping for something like yearly auditors? The same ones who regularly catch Ponzi schemers three decades after they start, I hope.