Tuesday, April 7, 2009

20% Estimated Losses For The 2007 CMBS Vintage; Add 16% For Whole Loans

That is the result of the stress test in the just published S&P "Scenario Analysis: Standard & Poor's Expcets the Downgrade Risk To Be High For Recent-Vintage U.S. CMBS Transactions."

Here is what the rating agency had to say about the methodology and summary of their analysis:
The economic recession combined with the absence of readily accessible financing in the capital markets has, in our opinion, skewed the credit risks related to the performance of CMBS sharply to the downside, and far in excess of what we expected at origination or in our prior scenario analysis. As a result, our baseline (or expected) case now reflects a more severe recession, with a peak unemployment rate forecast of 9.8%.

Since September/October 2008, we've witnessed significant deterioration in the credit performance of the CMBS transactions we rate. Our scenario analysis tests the resilience of our ratings on a loan-by-loan, deal-by-deal basis while factoring in the current and expected economic environment.

Through our scenario analysis, we intend to provide guidance on our expected future rating transitions. However, due to the high number of estimates and assumptions involved and the presence of loan-specific issues such as, among other things, in-place leases and debt structure, we cannot guarantee that the losses we project represent a measure of future rating changes.
These are S&P's key results:
  • Overall, we expect wide dispersion in performance between different vintages of CMBS, and even between particular deals within those vintages.
  • We expect the worst results in the 2007 vintage, with an average expected loss in our base case of 10.0%, followed by the 2006 vintage (5.2%), and then 2005 (2.5%).
  • The range of expected losses in our base case for 2007 vintage transactions was 0.5%-32.6%.
  • In our "stress" case, the average expected loss for the 2007 vintage is approximately 20%. Coming in much lower are the 2006 and 2005 vintages, at 13.4% and 8.0%, respectively.
  • In our baseline (expected) case, the majority of the 20% credit-enhanced 'AAA' classes (AM classes) are likely not at risk for downgrade. However, some of these classes within what we believe will be the worst performing 2007 vintage deals appear to be potentially vulnerable over their 10-year lives.
  • The majority of all junior 'AAA' rated securities (AJ classes) are likely vulnerable to downgrade. Of the total number of these at-risk classes, approximately 50% are from 2007 vintage transactions. Transactions we expect to display the worst performance may eventually see AJ classes transition to speculative-grade ratings.
  • We expect most speculative-grade classes from the 2005-2007 vintages to take principal losses. We also expect many transactions from the 2006 and 2007 vintages to eventually suffer losses at the low-investment-grade rating categories.
The chart below summarizes the various scenario expected lifetime losses by vintage:

And here is the argument for the horrendous expectations for 2005-2007 vintages:
We expect performance to be much worse in the 2005-2007 vintages because the underlying transactions generally had more aggressive loan underwriting and lower credit enhancement than the 2000-2004 cohorts. Examples of more aggressive loan underwriting include employing higher leverage (measured by LTV), widespread inclusion of interest-only periods (many for the entire term of the loan), and lowered required reserve amounts. In addition, many of the transactions included loans with initial DSCs below 1.0x based on in-place cash flow, with the
expectation of future significant increases in cash flow after origination (pro-forma underwriting).

The rationale for the variation in results among the 2005-2007 vintages is that the underwriting standards, already looser than for past vintages, became progressively worse as the property cash flows and valuations rose quickly during those years. For example, most of the large pro-forma loans were underwritten during 2006 and especially 2007, and 2007 deals also claimed the highest average Standard & Poor's LTVs and lowest average Standard & Poor's DSCs (followed by 2006, then 2005). Thus, when commercial real estate prices and cash flows peaked in 2007, the most recent origination was by far the most vulnerable to cash flow declines--which is what we stressed in our term default scenarios.
S&P notes that "an overwhelming majority of the 30% (usually called A4 classes or super-duper senior classes) and 20% credit-enhanced 'AAA' classes (AM classes) are not currently at risk for
downgrade. However, within what we believe to be the worst-performing deals in the 2007 vintage, a handful of AM classes may become vulnerable to downgrades over time. During the near term, we expect roughly half of the universe of junior 'AAA' securities (AJ classes) may be susceptible to downgrade, and a small number of classes may eventually drop to speculative grade.

Again, keep in minds that this is an analysis for CMBS. For whole loans the cumulative losses will likely result in substantially worse outcomes due to a majority of the loans having shorter maturities (the refi market is closed), being Interest Only initially and soon converting to amortizing, and generally lower DSCR ratios (at or below 1.0x). The impact of term and maturity defaults will likely add another 16% in lifetime losses for the 2007 vintage, causing total combined losses to reach just under 40% and declining ratably by older vintage. Sphere: Related Content
Print this post


qadi said...

PRU getting TARP implies that the capital markets agree with your less than sanguine outlook for cmbx.

___ said...

This is a great blog. Very happy someone is writing intelligently about the credit markets out there.

Part of the problem is that the rest of this country, even people who are in the business, refuse to learn about the credit markets. It is all a bunch of equity heads.

How can one understand the bottom and riskiest part of the capital structure without understanding the senior layers above it?

Anonymous said...

Sounds like the guys who bought the Hancock Building in Boston got robbed.

Bruce said...

"Everything's priced in 'cents'. Alcoa today admitted as much.

Anonymous said...

with respect to s&p's comments, keep in mind they had an 89% market share in the 2007 vintage. the fact that they totally missed the boat at issuance makes it pretty likely that they're missing it now too.

Anonymous said...

title is missleading, but i guess it sells doom and gloom.

20% is not expected, nor is 20% "the result of the stress test", 20% is there expectation in there stress case scenario.
10% is their current baseline expected loss (based on 9.8% unemployment projection).

any info on their "stress case" assumptions?

since its doomy and gloomy, interesting how the rating agencies are suddenly smart again and worthy of their analysis.

sweet! said...

This is laughable. The rating agencies knew from day one that the underwriting assumptions for the loans were bogus. How? Simply take a look at any one of the presale documents from S&P / Moody’s / Fitch. S&P’s assumption of value vs. the loan underwriting “appraised” value had a 50-75% variation. Furthermore, as soon as these loans were transferred to the master servicer (CMBS) they were instantly put on watchlist due to a high probability of default.

The rating agencies stress test is absolute bullshit (my humble opinion), and losses associated around a refinance (interest-only loans, etc.) is only one part of the story. Sustained office space rents NEVER justified the value of the assets. Expect major losses once interest reserves are depleted.

Anonymous said...

rating aganecy valuations are a useless measure, always have been. the agencies underwrite the loans, and apply a fixed cap rate to the cash flow. doesnt tell you much, other than since agency LTV went up because cap rates went lower. if cap rates went to 10+%, then the agency LTV will be too low. The only people that care about these valuations are people who dont know what they are and think they are more than what they are.

loans were instantly put on watchlist because DSCRs below 1.2 are automatically put on WL. Some loans were originated below 1.20, and thus went straight to WL, not because anything with the loan had changed.

sweet! said...

exactly my point...when a loan has a DSCR of ohhh.... .65 based upon in place cashflows there might be a problem, especially if projected rents were based upon inflated numbers that everyone in the industry knew were not sustainable. These things are timebombs. What the rating agencies actually did do was apply a reasonalbe cap rate (not going in 3-cap). Once interest reserves are pissed through-- depending upon inplace NOI --they will default (mezz and positions obviously toast) and will revert to special servicing. Pick a top-10 loan in a CMBS offering , vintage 2006-2008 and tell me the assumptions made sense...not a chance.