We have analyzed the credit quality of the assets in this transaction through conducting loan-level analyses of the mortgage pools. For each loan in the pool, our analysis estimated the foreclosure frequency and the loss severity and, by multiplying the foreclosure frequency by the loss severity, the potential loss associated with each loan. To quantify the potential losses associated with the entire pool, we calculated a weighted-average foreclosure frequency (WAFF) and a weighted-average loss severity (WALS) at each rating level. The product of these two variables estimates the required loss protection, in the absence of any additional factors. We assume that the probability of foreclosure is a function of both borrower and loan characteristics, and to become more likely (and the realized loss on a loan more severe) as the economic environment deteriorates.
In performing the credit analysis on this pool, we adopted the methodology and assumptions described in the sections entitled “Foreclosure Frequency Assumptions” and “Loss Severity Assumptions” in our Spanish residential mortgage-backed securities (RMBS) criteria (see “Criteria for Rating Spanish Residential Mortgage-Backed Securities,” published on March 1, 2002, and “Update To The Criteria For Rating Spanish Residential Mortgage-Backed Securities,” published on Jan. 6, 2009), with the following adjustments for this transaction:
-- ‘AA ’ base foreclosure frequency: 9%;
-- ‘A ’ base foreclosure frequency: 7%;
-- ‘BBB ’ base foreclosure frequency: 5%;
-- ‘AA ’ market value decline: 40%;
-- ‘A ’ market value decline: 35%;
-- ‘BBB ’ market value decline: 30%;
-- ‘BB ’ market value decline: 25%;
-- Jumbo loan penalty: EUR500,000 in Dublin;
-- Jumbo valuation penalty: EUR625,000 in Dublin;
-- First-time buyer penalty: 10% addition to adjusted base foreclosure frequency;
-- Income multiple penalty: 20% addition to adjusted base foreclosure frequency;
-- Self-certified penalty: 25% addition to adjusted base foreclosure frequency;
-- No adjustment is made for loans with loan-to-value (LTV) ratios of less than 50%;
-- Geographic concentration penalty: 1% addition to adjusted base foreclosure frequency for all loans if the concentration is greater than 60% in Dublin and greater than 20% in any other county;
-- The fixed costs of foreclosure are assumed to be 4% of the loan balance; and
-- The foreclosure period is assumed to be 48 months for the reasons set out further below.
The criteria applicable to our cash flow analysis for this transaction are primarily our “Cash Flow Criteria for European RMBS Transactions,” published on Nov. 20, 2003, and “Methodology And Assumptions: Update To The Cash Flow Criteria For European RMBS Transactions,” published on Jan. 6, 2009.
Due to current forbearance measures and the legal uncertainty regarding the foreclosure process, repossessions have generally been limited in the Irish residential mortgage market. To address this risk, we have increased our foreclosure period in our analysis to 48 months. Additionally, we have assumed that all loans with arrears greater than nine monthly payments default on Day 1 in our cash flow analysis, with losses and recoveries being realized at the end of the 48 month foreclosure period.
The level of severe arrears in both transactions has been increasing steadily. In Celtic 9, 90+ day delinquencies have risen to 19.09% from 11.71% since our previous review in March 2011, while In Celtic 10, 90+ day delinquencies have increased to 17.85% from 12.42% over the same period.
The continued decline in Irish house prices has also pushed up the weighted-average indexed LTV ratios in both mortgage pools, with both transactions exhibiting higher proportions of borrowers in negative equity, compared with our previous review. This, combined with the increases in arrears, has resulted in an overall increase in the WAFF, WALS, and required loss protection estimates in our analysis.
The reserve fund for both transactions are dynamically sized (i.e., partially sized based on the balance of nonperforming loans) and are nonamortizing. The reserve fund for Celtic 10 is currently at the level required by the transaction documents. However, with reduced excess spread, the reserve fund in Celtic 9 is currently at 98.7% of the required level for the current interest payment date.
Each transaction also features a liquidity reserve fund, which is funded through principal receipts and can be replenished on each interest payment date up to 1% of the current collateral balance. Thus, in a situation where transaction performance is weak, the issuer can continue to borrow principal to meet timely payment of interest on the class B notes, at the expense of meeting ultimate payment of principal on the class A2 notes.
When taking into account recoveries, the class A2 notes in both transactions are fully collateralized. However, our assumptions of undercollateralization give rise to negative carry. This, combined with the fact that principal can be used to pay interest through the liquidity reserve fund, means that we observe significant principal shortfalls in our cash flow analysis of each of these transactions. Our analysis shows that the levels of credit enhancement in both transactions are insufficient for the notes to maintain our current rating levels. Consequently, we have lowered to ‘B+ (sf)’ from ‘A (sf)’ and ‘B (sf)’ from ‘BBB+ (sf)’ our ratings on Celtic 9’s class A2 notes and Celtic 10’s class A2 notes, respectively.
We have also lowered to ‘B- (sf)’ our ratings on the class B notes in both transactions. This reflects our view on credit enhancement erosion due to assumed undercollateralization, even when considering recoveries (assumed at 50%) and the available reserve fund level.
Celtic 9 and Celtic 10 are Irish RMBS transactions backed by mortgages originated by First Active--a subsidiary of Ulster Bank Ireland Ltd..
-- Counterparty Risk Framework Methodology And Assumptions, May 31, 2012
-- European Structured Finance Scenario And Sensitivity Analysis: The Effects Of The Top Five Macroeconomic Factors, March 14, 2012
-- Global Structured Finance Scenario And Sensitivity Analysis: The Effects Of The Top Five Macroeconomic Factors, Nov. 4, 2011
-- Principles Of Credit Ratings, Feb. 16, 2011
-- Update To The Criteria For Rating Spanish Residential Mortgage-Backed Securities, Jan. 6, 2009
-- Update To The Cash Flow Criteria For European RMBS Transactions, Jan. 6, 2009
-- European Legal Criteria For Structured Finance Transactions, Aug. 28, 2008
-- Cash Flow Criteria For European RMBS Transactions, Nov. 20, 2003
-- Criteria For Rating Spanish Residential Mortgage-Backed Securities, March 1, 2002
Celtic Residential Irish Mortgage Securitisation No. 9 PLC
EUR1.75 Billion Mortgage-Backed Floating-Rate Notes
A2 B+ (sf) A (sf)
B B- (sf) BB (sf)
Celtic Residential Irish Mortgage Securitisation No. 10 PLC
EUR1.79 Billion Residential Mortgage-Backed Floating-Rate Notes
A2 B (sf) BBB+ (sf)
B B- (sf) BB- (sf)