Fitch: U.S. RMBS Criteria Changes for Newly Issued Transactions

NEW YORK--(BUSINESS WIRE)--Effective today, Fitch Ratings has announced several criteria revisions to ResiLogic, Fitch's mortgage default and loss model for U.S. RMBS prime, Alt-A and subprime transactions. While ResiLogic has proven to be an effective tool in measuring mortgage default and loss risk, Fitch believes the updated criteria captures the credit risk in the rapidly changing mortgage markets by incorporating new assumptions for falling home prices, poor performance of loans with certain characteristics and substantial changes in mortgage originations.

The revisions to Fitch's ResiLogic model are:

--Greater emphasis on regional economic risk. The ResiLogic model will give greater weight to the University Financial Associates (UFA) default multiplier component of the model. The UFA multipliers, which are updated quarterly, raise or lower the ResiLogic expected default probability as a function of a state-level forecast of default risk. This revision affects all mortgage products. Expected default rates will increase on average approximately 20%, depending on the geographic distribution of the mortgaged properties.

--Increased default expectations for short-term and hybrid adjustable-rate mortgage (ARM) products. Expected default rate multipliers for 2/28 ARMs are increased by 22% and by 12% for 3/27 ARMs. Expected default rate multipliers for 5/25 ARMs are raised by less than 5%. Expected default rates for ARMS with adjustment terms below two years are increased by 30%. Option ARMs are subject to an 8% higher multiplier.

--The introduction of a new risk category for borrower income and asset documentation. ResiLogic now categorizes certain loans as 'Low' documentation, in addition to the existing 'Full', 'Reduced' and 'None' categories. Expected default rates for 'Low' documentation loans are 43% greater than for 'Full' documentation loans.

--Default expectations will be determined through analysis of 'back-end' debt-to-income (DTI) ratios, rather than 'front-end' DTI ratios as had been implemented in ResiLogic. If a debt-to-income ratio is not provided for a subprime loan, the default DTI is assumed to be 50%. The impact of this revision is generally minor, although subprime data files lacking in DTI information will be noticeably affected.

--The removal of benefits for short-term loan seasoning. The ResiLogic model reduces default expectations for performing loans as a function of loan age at the time of securitization. This feature of ResiLogic has been removed.

Fitch has previously commented on the potential risks to subprime RMBS structures associated with changes in loan modification policies. Fitch will separately provide further commentary on those risks. Apart from consideration of loan modifications, Fitch does not plan any changes to its cash flow modeling methodology as described in Fitch's Feb. 6, 2007 criteria report, 'U.S. RMBS Cash Flow Modeling Criteria: Updated'.

Fitch introduced its ResiLogic default and loss model for establishing expected losses and loss coverage levels for RMBS in its Oct. 4, 2006 report, 'ResiLogic: U.S. Residential Mortgage Loss Model'.

Rationale for Criteria Changes

1) UFA weighting: Fitch believes that the greatest risk to new U.S. RMBS is the continued deterioration of home prices. To better address that risk, Fitch has chosen to make its loss expectations more dynamic in response to changes in regional housing markets. The UFA default multipliers that Fitch has employed since the introduction of ResiLogic have proven to be effective indicators of the default risk trend in the various regions of the U.S. The greater weighting to UFA multipliers introduced today will have the effect of raising default expectations by approximately 20%, reflecting the adverse trends in the states that contribute the majority of mortgages to RMBS, most notably California. This substantial increase reflects both the greater weighting of the UFA multipliers, and the negative trend in the multipliers in recent quarters. The UFA model contains a home price forecast for each state, which currently indicates an expected national decline in home prices of approximately 5% cumulatively over the next five years on a nominal basis. The UFA multiplier is applied in the analysis of prime, Alt-A, and subprime mortgages. While Fitch has seen little evidence of deteriorating prime credit performance, the enhancement to ResiLogic recognized the risk that prime borrowers are not immune to weaker home prices.

The UFA multipliers are updated quarterly. Fitch will publish the updated multipliers and indicate the impact on expected loss each quarter. Continued deterioration in home prices or regional economies could cause ResiLogic expected losses to rise noticeably from quarter to quarter.

2) ARM Default Risk Adjustment: With the recent decision by many large originators to cease origination of 2/28 hybrid ARM mortgages, Fitch anticipates that refinancing options for existing 2-year hybrid ARM borrowers will be sharply curtailed. The absence of a comparable product is likely to expose more borrowers to payment shock risk and increase defaults at the rate reset. This concern has been highlighted in Fitch's forecasted defaults and losses for 2005-2006 vintage subprime RMBS (see Fitch's Aug. 1, 2007 press release 'Fitch affirms $20B & Downgrades $2.4B of U.S. Subprime RMBS; New 2005-2006 Surveillance Criteria'). Therefore, Fitch believes that a more conservative approach to recently originated ARM products with two- to three-year fixed rate periods is warranted, and has increased default expectations to reflect the likelihood of substantial payment shock for these loans.

The ARM risk adjustment also affects Option ARMs. Fitch has been active in describing the risks of these products, see Fitch's report 'Option ARM Risks and Criteria' dated Oct. 4, 2006 and has rated very few senior/subordinate RMBS backed by Option ARMs. Fitch's loss coverage levels will continue to reflect Fitch's opinion of the substantial potential payment shock in these negatively amortizing products, particularly in an environment of continuing lower home prices and tighter mortgage credit.

3) New Documentation Category: ResiLogic initially provided three categories for documentation, reflecting the categorization of loans in the model development sample. However, after evaluating lender programs and the recent performance of subprime and Alt-A mortgages, Fitch's analysts felt that greater differentiation was called for. This has resulted in the introduction of a new 'Low' documentation category, which indicates default risk which is greater than the existing 'Reduced' category but less than the existing 'None' category. Fitch will publish a research report the week of Aug. 6, 2007 describing the mapping between lenders documentation programs and the new categories.

4) Change in DTI Ratio: Back-end DTI ratio data is generally more available than front-end data in the data files provided by mortgage issuers. Therefore Fitch has modified ResiLogic to utilize back-end DTI. However, Fitch continues to be concerned by the prevalence of missing DTI data. Fitch will use a default assumption for missing subprime DTI of 50%.

5) Removal of 'pre-seasoning' benefit: Fitch's regression analysis in the development of ResiLogic indicated that even a few months of performance prior to securitization reduced the risk of mortgage default. However, given the poor performance of recent originations and the expected performance pressure on new originations, Fitch has decided to disable this feature of ResiLogic.

Additional Criteria Changes Under Development

Fitch is currently developing two additional criteria changes which will further enhance ResiLogic's forward-looking approach to regional economic risk. First, Fitch has contracted with UFA to provide a series of MSA-level multipliers in addition to the existing state multipliers. This will enable enhanced differentiation among regions within states and will raise sensitivity to concentrations in high-risk regions. Second, Fitch is analyzing the application of UFA forecasts in determining expected case loss severity. While Fitch stress-case loss severities reflect the historical experience in housing downturns, adding a UFA component to the expected case loss severity analysis will create greater sensitivity to geographic risk in the base case. Fitch anticipates that these enhancements will be implemented into ResiLogic over the next 60 days.

Fitch will discuss its new ResiLogic and surveillance criteria as well as its most recent subprime rating actions during a teleconference on Tuesday, August 7th at 11am EST. Teleconference details are covered in a separate release. All criteria, rating actions and special comments are available at www.fitchratings.com/subprime.

Fitch's rating definitions and the terms of use of such ratings are available on the agency's public site, www.fitchratings.com. Published ratings, criteria and methodologies are available from this site, at all times. Fitch's code of conduct, confidentiality, conflicts of interest, affiliate firewall, compliance and other relevant policies and procedures are also available from the 'Code of Conduct' section of this site.