Subprime RMBS Ratings Tightened

Moody’s Investors Service is seeking comments on proposed changes to its ratings approach for analyzing delinquent loans in performing subprime RMBS.

The approach has been refined to evaluate the payment status of mortgage loans as of the closing date for securitizations backed by recently-originated subprime mortgages.

“Historically, transactions backed by recently originated collateral generally had de minimis amounts of delinquent loans as of the closing date,” Moody’s explained in the report. “In the recent past, however, Moody’s has observed a higher proportion of delinquent loans as of the closing date.”

“The purpose of this paper is to provide more transparency on our refined approach and to offer the market the opportunity to comment,” the ratings agency said in its report, Call for Comment: Update to Moody’s Approach to Analyzing Delinquent Loans Included in Performing Subprime RMBS.

When rating RMBS that have delinquent loans, Moody’s adjusts its loss estimates to account for the greater likelihood of default of delinquent loans. The proposed revision for assessing the rating impact of delinquent loans in securitizations is broadly driven by the following elements: pool delinquency levels compared to base assumptions; the reporting lag between the delinquency determination date and the closing date; and structural provisions that may shift the securitization classes’ exposure to delinquent loans, Moody’s said.

Moody’s has evaluated historical payment and delinquency patterns to establish base case expectations and assumptions to be applied for missing or untimely information. Moody’s is also placing greater emphasis on the reporting lag, or the difference between the reported “as of” date for payment status and the closing date.

“The smaller the reporting lag, the lower the volatility in the actual closing date delinquency status and therefore the lower our loss expectations,” Moody’s wrote.

The methodology is implemented by determining the base case pool delinquency levels based on the deal closing date and assumed payment rates; determining either the actual level of delinquencies as of the closing date or the presumed level of closing date delinquencies based on the reported delinquency status and the reporting lag; and adjusting loss expectations based on the actual or presumed level of delinquencies in the pool as of the closing date relative to Moody’s base delinquency assumptions, the report read.

For example, for a subprime first lien transaction that closes on March 30, Moody’s base assumption is that as of the closing date, 5.80% of the loans will not have made the payment that was due on the 1st of that month. The increase in expected losses on the 29th day since the loans are due is 26.4% If actual delinquencies are higher than the base assumptions, loss expectations increase and the inverse would happen if actual delinquencies were lower.

So, if on the closing date of the 30th, 7.80% of the loans have actually not made the payment that was due on the 1st, which is 2% higher than Moody’s base delinquency assumption, the expected loss level increases to 6.00% and the loss level would be adjusted by 0.53 percent — calculated by multiplying 2% by 26.4%.

“Therefore, if base expected loss levels are 6.00% assuming 7.80% of the pool has not made the March 1st payment by the closing date, our delinquency-adjusted loss levels would increase by 0.53% to 6.03%.

Assuming 29-day reporting lag between a transaction’s delinquency reporting date and the transaction’s closing date, Moody’s anticipates that 13.83% of a pool will be 29 days actually past due as of the closing date, as compared to 5.80% with zero-day reporting lag.

With a 59-day reporting lag, the ratings agency anticipates that 59.60% of a pool will be 29 days actually past due as of the closing date.

The report highlights four different scenarios for such results.

A similar approach to that applied to first lien RMBS is applied to those with second-liens. The two major differences are that, after evaluating historical default rates, the percentage increase in expected loss depending on the number of days delinquent are more stressful than for first liens and the payment rate assumptions would be lower than for first lien loans to reflect the weaker payment behavior associated with second lien loans.

Comments are good through June 15. Moody’s expects to publish a final methodology at the end of the month to be effective July 1.