Monte Carlo Simulation of CDOs (Part 1)

Most people have unanswered questions on their minds. Not me. Right now I have two answered questions on my mind.

1. Who's rating the rating companies?
2. Who's the ultimate guarantor of CDOs when the derivative boom collapses?

The answer in both cases is no one (or if you prefer, Madame Meriweather's Mudhut in Malaysia) as I proposed in An email from Bernanke. If you haven't read about Madame Merriweather yet, you might want to click on that link for a good laugh.

To address the first question please consider the Bloomberg article CDO Boom Masks Subprime Losses by Richard Tomlinson and David Evans. What follows next is a long but important summary.

Sales of CDOs worldwide have soared since 2004, reaching $503 billion last year, a fivefold increase in three years, according to data compiled by Morgan Stanley.

CDO holdings have already declined in value between $18 billion and $25 billion because of falling repayment rates by subprime U.S. mortgage holders, Lehman Brothers Holdings Inc. estimated on April 13. In many cases, investors don't even know that values have dropped. In this secretive market, there is no easy way for them to find out what their CDOs are worth.

The three leading rating companies, all based in New York, say that policing CDOs isn't their job. S&P, which controls 40 percent [of the market], asks investors in its published CDO ratings not to base any investment decision on its analyses. Fitch, which has 16 percent of the worldwide credit rating field, says its analyses are just opinions and investors shouldn't rely on them.

When it comes to CDOs, rating companies actually do much more than evaluate them and give them letter grades. The raters play an integral role in putting the CDOs together in the first place.

Bankers call the bottom sections of a CDO -- the ones that are the most vulnerable to subprime and other junk -- the equity tranches. They also have another, more-emotive phrase for them: toxic waste. Investors have little idea how toxic some of these CDOs are, Drexel's Mason says.

"We compose CDOs with a bunch of this stuff," he says. "Now we just jack up the risk, jack up the misunderstanding. We're throwing our money to the wind. We now know the defaults are in the mortgage pools and it's only a matter of time before they accumulate to levels that will threaten the CDO market."

"They don't blow up," Brian McManus, head of CDO research at Charlotte, North Carolina-based Wachovia Corp. says of CDOs. "They just kind of melt."

Why buy a corporate bond yielding 5 percent when you can invest in a CDO with the same credit rating and the promise of a return twice as high?

There are two caveats: It's nearly impossible to find out exactly what's in a CDO, and CDOs aren't regulated.

Almost all CDOs are sold in private placements, and their current values aren't posted anywhere. "There is absolutely no transparency," Das says. "It's difficult to get current values or information about the underlying assets in the CDO."

Financial regulators have effectively outsourced the monitoring of CDOs to the rating companies.

Bonanza for Raters

"CDOs are the cash cow for rating agencies," says Frank Partnoy, a former bond trader, now a University of San Diego law professor and author of `Infectious Greed: How Deceit and Risk Corrupted the Financial Markets'. "They're clearly a gold mine."

Corporate bonds rated Baa, the lowest Moody's investment rating, had an average 2.2 percent default rate over five-year periods from 1983 to 2005, according to Moody's. From 1993 to 2005, CDOs with the same Baa grade suffered five-year default rates of 24 percent, Moody's found.

Monte Carlo Simulation

Because there are so many moving parts to a CDO, rating companies have to assess not only the chance that something may go wrong with one piece but also the possibility that multiple combinations of things could falter. To do that, S&P, Moody's and Fitch use a mathematical technique called Monte Carlo simulation, named after the Mediterranean gambling city.

The rating companies take all the data they have on a CDO, such as information about specific bonds and securitizations and the remaining types of loans to be purchased for the package.

The firm enters data into a software program, which calculates the probability that a CDO's assets will default in hypothetical situations of financial and commercial stress. The program effectively rolls the dice more than 100,000 times by running the information randomly.

False Sense of Security

The rating companies base their simulations and ratings of each tranche on assumptions about default and recovery rates that may be incorrect, Cifuentes says.

"The danger with Monte Carlo is that it gives you a false sense of security," he says. "If the input data that you use is a little bit uncertain, your numbers are going to be trash, but they will look convincing."

Credit rating companies may have miscalculated the potential toxicity of securities backed by subprime loans, McManus says. "With CDOs, they underestimated the volatility of the subprime asset class in determining how much leverage was OK," he says.

The banks and rating companies have stretched the frontiers of CDOs with products known as CDO squareds and CDO cubeds. As the names suggest, a CDO squared is formed by bundling together a bunch of CDOs, and a CDO cubed, which can contain thousands of different securities, is formed by lashing together a bunch of CDO squareds.

The grades that rating companies give CDO squareds and cubeds are worthless, says Janet Tavakoli, founder of Chicago-based consulting firm Tavakoli Structured Finance Inc., which advises investors on CDO purchases. "Ratings on these products are based on smoke and mirrors," Tavakoli says.

The inner workings of CDOs are normally invisible to the public. The demise of the $340.7 million CDO Credit Suisse sold in December 2000 was documented in a 38-page report dated March 26 that Moody's stamped as confidential.

Former banker Das wonders why few people are probing the potential dangers for CDO investors. "I think the regulators seem to be fairly sanguine about all of this," he says. He says subprime mortgage defaults have just started to soar. "The fuse has been lit," Das says. "Somebody should be trying to find where this wire is running to."

If that is not an eye opening piece what is? I condensed the original as much as I could but there was a ton of information there.

Short Summary

• Investors are buying bundles of CDOs without having a clue as to what is in them
• Rating CDOs is a cash cow for the rating companies
• There is a clear conflict of interest between the rating companies and the deals they are involved in
• Ratings programs are based on Monte Carlo Simulation but the initial input assumptions about defaults and recovery rates may be far off the mark
• After the initial rating, CDOs are not marked to market thus no one really knows what anything is worth
• No one is rating the rating companies or their models
• No one is even concerned about this state of affairs

Based on the above it is highly likely that pension plans holding some of the CDO toxic waste are assuming big returns when in fact they are holding worthless paper. There is easily a potential of a waterfall event as I talked about in Liquidity Risk & Effective Leverage.

Long Term Capital Management blew up (see When Genius Failed) when models written by Nobel prize winning economists Myron Scholes and Robert Merton failed.

Are the Monte Carlo models (or more importantly the assumptions going into the Monte Carlo models) any better? How can anyone even know if the only "mark to market process" is based on what the model says?!

The scariest thing is that we are in a monetary environment that is unprecedented in history. M3 is soaring in nearly every major country on the globe. Asset prices have never been more correlated than now. The amount of derivatives in play is in the hundreds of trillions of dollars all bet on Monte Carlo Simulation. Wow!

Single tranche CDO pricer

Those wishing to delve into the CDO market can even purchase their own Single Tranche CDO Pricer complete with a Monte Carlo simulation method that employs a factor copula approach for handling exotic variations, a fast, specialized semi-analytic method and the Large Pool Gaussian Copula Model (Large Pool Model) approach. The Large Pool Model is a quick, user-friendly model that requires a minimal amount of data.

Who Owns What?

Initial pricing is one thing, subsequent pricing is another but compounding the problem is that no one even knows with certainty who owns what in which tranch. Please consider Paper Chase.

You're in luck. Your mortgage lender has flipped, sliced and diced your loan--and now no one knows who holds it. You can't be foreclosed on if no one knows who holds the paper or the paper can't be found. Assuming that people can figure out who owns what, the next question is....

Who is the guarantor? (if and when a credit event happens in the higher rated tranches)

The guarantor (and how well financed the guarantor is) will be addressed in Part 2 of Monte Carlo Simulation, tomorrow.

Mish notes:
This post originally appeared in Minyanvile.
I will post something on bond markets worldwide later this evening.

Mike Shedlock / Mish
http://globaleconomicanalysis.blogspot.com/