New Report: Write-Downs Unlikely To Impact RMBS Ratings

New Report: Write-Downs Unlikely To Impact RMBS Ratings Principal reductions are unlikely to change long-term ratings for U.S. residential mortgage-backed securities (RMBS), according to a new report from Fitch Ratings that also determined most ratings for transactions with the largest potential for reductions are already distressed.

‘The ratings impact will ultimately hinge on the volume of strategic defaulters, the amount of the reduction offered, and re-default rates post-modification,’ says Fitch Ratings, adding that it has ‘conducted sensitivity analyses and found that RMBS losses would be lower than currently expected if principal forgiveness is offered to delinquent borrowers – even if half of them re-defaulted post-modification. However, this could increase the likelihood of losses associated with strategic defaults, which would offset this benefit, particularly when larger amounts were assumed to be forgiven.’

Fitch Ratings adds that its analysis of different scenarios indicates that principal reductions resulting from the $25 billion foreclosure settlement announced in February would likely have ‘only a small impact’ on private-label RMBS.

‘This is based on the view that the settlement will likely only reach a small portion of underwater borrowers; servicers will likely take a balanced approach in applying reductions to portfolio and securitized loans; and current ratings for most potentially impacted transactions are already distressed,’ Fitch Ratings says.

However, the impact on expected losses will ultimately depend on the volume of strategic defaulters, the amount of the reduction, and level of re-default rates in the post-modification period. But even if servicers attempted to reach the principal forgiveness targets solely with RMBS collateral, the settlement amount would only address roughly 10% of underwater principal. Fitch Ratings estimates approximately $203 billion of negative equity exists in private-label RMBS pools alone.


Please enter your comment!
Please enter your name here