Mortgage Securities: Inside the Rating Process

One question being asked about the rating agencies is why they were so late in downgrading CMBS and CRE issues that subsequently experienced problems. A recent report by Moody's Investors Service offers one explanation. Ratings to some extent reflect anticipated future market changes (rising or falling) that would affect a current offering.  Several examples of this are as follows:

Differentiation in property markets: Differentiation among the different property markets is more prevalent when an economy slips into a downturn because employment and space demand will then differentiate among property types.

Differentiation in leverage: Loans with loan-to-value ratios (LTVs) of 70 percent, 80 percent or 90 percent tend to perform about the same in a rising market, but in a falling market, the differences among them will increase.

Differentiation in losses: Losses on defaulted loans can be expected to be less in a rising market than in a falling market. Moody's offers the example of a defaulted loan heading toward liquidation with an expected 40 percent loss severity that might see the default drop to 30 percent if prices appreciate by 10 percent during the liquidation period, but by the same token rise to 50 percent if prices fall by 10 percent.

Loan structure matters: Whereas in a rising market, differences in loan terms may make little difference in performance, the case is different in a declining market where loans with higher amortization rates and more reserves will experience lower loss on default than will other loans.

Upgrades Outpace Downgrades

During the third quarter of 2007, Moody's rated 148 securitizations that consisted of 1,424 CMBS classes. Of these, 1,075 were affirmations and confirmations, while upgrades numbered 331 and downgrades were only 18. For investment-grade rated classes, upgrades exceeded downgrades by 310 to 7. Roughly 75 percent of all classes remained unchanged.

Impact on Holders

While the delayed downgrades by the rating agencies can be explained in part (as noted above), the consequences for firms with large holdings of CDOs and related instruments were nothing short of disastrous. This was due to their need to "mark to model" rather than to "market" (because markets were virtually nonexistent).

According to newspaper reports, major firms felt relatively secure in their holdings because they held the "super-senior" tranches of CDOs normally considered almost riskless because the lower tranches are expected to fully absorb any anticipated losses. But since the senior classes needed to be marked to model in order to be valued, the holders were forced to recognize very large "paper" losses when the ratings were slashed. Quoted in the New York Times, Darell Duffie, a finance professor at Stanford University's business school, said "In general, the industry standard model for pricing CDOs is not adequate in my view, which means that there's a lot of uncertainty about what they are worth .…They can get better models, but that's not something they can do overnight."