Cartesian Residential Mortgages 1 S.A.

Transcription

1 Presale: Cartesian Residential Mortgages 1 S.A. Primary Credit Analyst: Annabelle C Teo, London (44) ; annabelle.teo@standardandpoors.com Secondary Contact: Neil Monro, London (44) ; neil.monro@standardandpoors.com Table Of Contents Million Floating-Rate Mortgage-Backed Notes (Including 47.2 Million Unrated Notes) Transaction Summary Strengths, Concerns, And Mitigating Factors Transaction Structure Origination And Underwriting Note Terms And Conditions Collateral Description Counterparty Risk Credit Structure Hedging Risk Payment Structure And Transaction Features Scenario Analysis Sector Credit Highlights FEBRUARY 26,

2 Table Of Contents (cont.) Surveillance Standard & Poor's 17g-7 Disclosure Report Related Criteria And Research FEBRUARY 26,

3 Presale: Cartesian Residential Mortgages 1 S.A Million Floating-Rate Mortgage-Backed Notes (Including 47.2 Million Unrated Notes) This presale report is based on information as of Feb. 26, The ratings shown are preliminary. This report does not constitute a recommendation to buy, hold, or sell securities. Subsequent information may result in the assignment of final ratings that differ from the preliminary ratings. Class Preliminary rating* Preliminary amount (mil. ) Available credit enhancement (%) Interest Step-up interest A AAA (sf) Three-month EURIBOR plus a margin B AA (sf) Three-month EURIBOR plus a margin C A+ (sf) Three-month EURIBOR plus a margin D NR Three-month EURIBOR plus a margin Three-month EURIBOR plus a margin Three-month EURIBOR plus a margin Three-month EURIBOR plus a margin Three-month EURIBOR plus a margin Optional call date Legal final maturity April 2019 April 2044 April 2019 April 2044 April 2019 April 2044 April 2019 April 2044 E NR N/A N/A April 2019 April 2044 S NR N/A N/A N/A N/A *Our rating on each class of notes is preliminary as of Feb. 26, 2014, and subject to change at any time. We expect to assign final credit ratings on the closing date, subject to a satisfactory review of the transaction documents and legal opinions. Standard & Poor's ratings address timely interest and ultimate principal for the class A notes. We treated the class B and C notes as deferrable-interest notes in our analysis. Our preliminary ratings on the class B and C notes only address the ultimate payment of principal and the ultimate payment of interest. Available credit enhancement for the class A notes will consist of subordination and a cash reserve. The class S notes' issuance proceeds will fund the cash reserve at closing. NR--Not rated. EURIBOR--Euro Interbank Offered Rate. N/A--Not applicable. Transaction Participants Originator Quion 10 B.V., Ember Hypotheken 1 B.V., Ember Hypotheken 2 B.V. Arranger The Royal Bank of Scotland PLC Seller Ember VRM S.a.r.l. Issuer advisor Venn Partners Mortgage administrator/servicer Quion Hypotheekbemiddeling B.V. Security trustee Stichting Security Trustee Cartesian Residential Mortgages 1 Issuer administrator Intertrust (Luxembourg) S.à r.l. Interest swap counterparty BNP Paribas Issuer bank account provider BNP Paribas Securities Services (Luxembourg Branch) Foundation collection account provider ABN AMRO Bank N.V. FEBRUARY 26,

4 Supporting Ratings Institution/role BNP Paribas as interest swap counterparty BNP Paribas Securities Services (Luxembourg Branch) as issuer bank account provider Rating A+/Negative/A-1 A+/Negative/A-1 Transaction Key Features Expected closing date March 2014 Collateral Dutch prime first-ranking mortgages Outstanding net principal of the portfolio ( )* 481,519,988 Country of origination The Netherlands Highest concentration Noord Holland (40.08%) Property occupancy 100% owner-occupied Weighted-average loan-to-value ratio (%) Weighted-average loan-to-value ratio in the absence of data adjustment for audit results (%) Average net loan size balance ( ) 233,621 Largest loan size 2,414,158 Weighted-average seasoning (months) Arrears (%) 0.97 Modelled arrears (%) 2.71 Redemption profile Cash reserve Mortgage priority Principal deficiency ledger Of the pool, 62.43% comprises interest-only loans and 1.37% comprises savings loans The reserve will amount to 2.30% of the initial pool and can build-up through excess spread up to 4.00% of the initial pool's balance First-ranking or first- and sequentially-lower-ranking Loss-based Proportion of jumbo valuation loans (more than 500,000) 30.33% Revolving period None, as the pool is static *Based on the provisional portfolio as of Nov. 30, Based on indexation in line with Standard & Poor's methodology. Total borrower exposure. Current arrears and projection combined. Transaction Summary Standard & Poor's Ratings Services has assigned preliminary credit ratings to Cartesian Residential Mortgages 1 S.A.'s residential mortgage-backed floating-rate class A, B, and C notes. At closing, Cartesian Residential Mortgages 1 will also issue unrated class D, E, and S notes. Cartesian Residential Mortgages 1 securitizes a pool of performing loans secured on first- and consecutive-ranking Dutch mortgages. The originators Quion 10 B.V., Ember Hypotheken 1 B.V., and Ember Hypotheken 2 B.V. were previously owned by GE Artesia Bank. On Dec. 13, 2013, Ember VRM S.a.r.l. (the seller) purchased the mortgage receivables from the originators, having acquired the originators from GE Artesia Bank immediately beforehand. It intends to establish a new mortgage platform in the Netherlands. It then sold the mortgages to the issuer on Dec. 19, The issuer financed the purchase by issuing unrated notes to GIFS Capital Company LLC and subordinated notes to the seller. At closing, FEBRUARY 26,

5 Cartesian Residential Mortgages 1 will issue the rated and unrated notes to repay this bridge financing (interim borrowing). As the seller is a special-purpose entity (SPE), it has limited resources to meet its obligations. In our view, this limits the value of the seller's representations and warranties to the issuer. We consider the representations and warranties package to be nonstandard. We have increased our weighted-average foreclosure frequency (WAFF) estimates to address this risk. The class A to D notes' interest rate is equal to three-month Euro Interbank Offered Rate (EURIBOR) plus a class-specific margin. The issuer will pay interest according to the interest priority of payments. We treated the class B and C notes as deferrable-interest notes in our analysis. Under the transaction documentation, the issuer can defer interest payments on the class B and C notes. Consequently, any deferral of interest would not constitute an event of default. While our preliminary 'AAA (sf)' rating on the class A notes addresses the timely payment of interest and the ultimate payment of principal, our preliminary ratings on the class B and C notes only address the ultimate payment of principal and the ultimate payment of interest. The notes' interest rate will be based on three-month EURIBOR, while the pool contains loans linked to a standard variable rate (SVR) and EURIBOR, as well as fixed-rate loans. Basis and interest-rate mismatch risk therefore arise in the transaction. For the proportion of the pool that contains SVR and fixed-rate loans, the issuer will enter into a swap agreement with BNP Paribas to hedge the interest rate risk. However, the swap will not address basis risk for the EURIBOR-linked loans. We have addressed the transaction's unhedged exposure to basis risk by stressing the cash flow from the assets. We consider the pool to be well-seasoned, with a weighted-average seasoning of 105 months. However, as GE Artesia Bank's clientele consists of high net worth individuals, the average loan amounts and property valuations are greater than other comparable Dutch residential mortgage-backed securities (RMBS) pools. The pool's weighted-average loan-to-value (LTV) ratio is of 95%. Of the pool, 0.96% comprises loans with arrears either greater than, or equal to one month. Our preliminary ratings reflect our assessment of the transaction's payment structure, cash flow mechanics, and the results of our cash flow analysis to assess whether the notes would be repaid under stress test scenarios. The transaction's structure relies on a combination of subordination, excess spread, and a reserve fund to mitigate credit losses and income shortfalls. Taking these factors into account, we consider the available credit enhancement to the rated notes to be commensurate with the preliminary ratings that we have assigned. Strengths, Concerns, And Mitigating Factors Strengths The preliminary capital structure provides 14.8% of credit enhancement to the class A notes through subordination and a specially-dedicated nonamortizing reserve fund, which replenishes after the class A notes' principal deficiency ledger (PDL) in the revenue priority of payments. The class S notes' proceeds will fund this reserve up to 2.3% (the reserve fund first target level) of the notes' closing balance. FEBRUARY 26,

6 Credit enhancement for the junior rated class B and C notes is available through subordination and the reserve fund. Any available excess spread will be used to increase the size of the initial reserve fund up to 4% of the notes' closing balance. The reserve's top-up to this level will occur after the class C notes' PDL mechanism is triggered. This part of the reserve will therefore be available for the class B and C notes. A PDL mechanism permits excess spread to be trapped in the priority of payments. The PDL will be debited for any losses on the portfolio, and for any use of principal to mitigate interest shortfalls. The capital structure is sequential, with no pro rata amortization. Credit enhancement can therefore build-up meaningfully over time for the rated notes, which will enable the structure to withstand performance shocks. There is a total-return swap in place to mitigate interest rate risk arising from the fixed-rate and SVR rate assets in the pool. This accounts for 73% of the preliminary pool. The portfolio is geographically diverse and predominantly exposed to urban areas, which tend to carry a lower risk of default in our analysis. We consider the pool to be well-seasoned, with a weighted-average seasoning of 105 months. A collection foundation account structure will be established at closing, which mitigates commingling risk: Borrowers make mortgage payments to a bank account held with the collection foundation (a bankruptcy-remote entity). At closing, we understand the documentation will contain appropriate downgrade language for the account provider. Concerns and mitigating factors GE Artesia Bank targets high net worth individuals. The average loan amount is greater than other comparable Dutch RMBS pools. Likewise, the properties' value tends to be greater than the average Dutch property, which may affect their liquidity upon a forced sale. We have addressed this risk in our credit analysis. In our opinion, the portfolio's weighted-average indexed loan-to-value (LTV) ratio is relatively high, at 95%. However, the audit for this transaction did not cover the valuation date, a key field that we rely upon in our analysis. As a result, we assumed all valuation dates to be as of August 2008, the height of the Dutch housing market. Consequently, the weighted-average indexed LTV ratio that we used in our credit analysis was 105%. Additionally, the scope of this transaction's audit did not cover the income data. Consequently, we did not give credit to income data in our analysis, as we did not consider this data to be reliable. Of the pool, 0.96% comprises loans with arrears either greater than, or equal to one month. We view these types of borrowers as having a higher risk of default. In line with our Dutch RMBS criteria, we have increased our WAFF accordingly to address this increased risk (see "Dutch RMBS Market Overview And Criteria," published on Dec. 16, 2005). The Dutch economy remains weak, in our view. Due to high unemployment, we expect arrears in Dutch RMBS transactions to increase. We also performed an arrears analysis using historical information from the originator's mortgage book. Overall, we projected 1.74% total arrears in our credit analysis to mitigate this risk. In total, we have therefore modeled 2.71% of arrears. If a mortgage borrower's saving policy provider becomes insolvent, borrowers may be able to set-off any loss resulting from their savings policy repayments against their mortgage loans. In most cases, we are assured that the sub-participation agreement between the savings provider and the issuer mitigates this set-off risk. In this transaction, we have been informed that some of the savings providers may not enter into a sub-participation agreement before closing. However, in order to mitigate set-off risk, a reserve fund will be funded at closing equal to the original balance of these savings loans in the pool. As the seller is a SPE, it has limited resources to meet its obligations. In our view, this gives limited value to the seller's representations and warranties to the issuer. We have therefore increased our WAFF assumptions to address this risk. At present, none of the pool's mortgage receivables are shared with other claims that the originator may have FEBRUARY 26,

7 against the borrower. However, under the transaction documents, other claims or further advances may be originated in the future. The risk of co-sharing security and the increased potential credit risk arising from further advances is mitigated by the fact that the seller will repurchase these loans from the pool. Given the seller is a SPE and its resources are limited to fulfill its obligations, the seller will purchase these loans from the pool before any other claims or further advances are originated. Of the pool, 26% includes loans that are linked to three-month EURIBOR. Although the notes are also linked to three-month EURIBOR, the index on the loans reset at a different date to the notes. This leads to a basis mismatch risk and the swap will not cover these loans. We have addressed the transaction's unhedged exposure to basis risk by stressing the cash flow from the assets. We note that the seller may be subject to a VAT liability of 350,000 due to previous VAT payments that should have been charged by the Dutch tax authorities, which could heighten the risk of seller insolvency. However, we do not view this as a material risk, as we assume as part of our analysis that the seller is insolvent on Day 1. We also understand that the VAT liability will not fall to the issuer. Transaction Structure On Dec. 13, 2013, the seller (Ember VRM) purchased the mortgage receivables from the originators, having acquired the originators from GE Artesia Bank immediately beforehand. The issuer then purchased these mortgages on Dec. 19, The issuer financed the purchase by issuing unrated notes to GIFS Capital Company and subordinated notes to Ember VRM. FEBRUARY 26,

8 At closing, the issuer will use the class A to E notes' issuance proceeds to repay the bridge financing from GIFS Capital Company and Ember VRM. The class S notes' issuance proceeds will fund the cash reserve. We have received legal opinion confirming that the mortgages receivables were transferred to the issuer by a silent assignment and that if GE Artesia Bank or Ember VRM were to become insolvent, the mortgages would not become part of their respective insolvency estates. Representations and warranties The seller provides representations and warranties in the mortgage sale agreement, which we consider to be standard for a Dutch RMBS transaction. If there is a breach in the representations and warranties, then the seller will repurchase the loan from the pool, if the breach cannot be remedied within 14 days. As the seller is a SPE, it has limited resources to meet its obligations. In our view, this limits the value of the seller's representations and warranties to the issuer. FEBRUARY 26,

9 We have considered the fact that the pool is highly seasoned and the availability of a 3.5 million reserve account to mitigate the risk of the seller's limited resources to repurchase loans from the pool. Furthermore, Ember VRM received representations and warranties on the portfolio from GE Artesia Bank on the first leg of the mortgage portfolio sale. This gives the issuer the ability to seek recourse from GE Artesia Bank if Ember VRM is unable to fulfil a repurchase. Moreover, Ember VRM's rights under GE Artesia Bank's representations and warranties will be pledged from Ember VRM to the issuer. Consequently, if the seller becomes insolvent, the issuer will have direct recourse to GE Artesia Bank. However, we note that Ember VRM can only seek a portion of the notional amount of the rated transaction from GE Artesia Bank. Additionally, its ability to seek recourse from GE Artesia Bank would expire before the first optional call date, and the seller's reserve is not pledged to the issuer. We have increased our WAFF assumptions to address the risk of the nonstandard representations and warranties package. Further advances and other claims At present, none of the pool's mortgage receivables are shared with other claims that the originator may have against the borrower. However, under the transaction documents, other claims or further advances may be originated in the future. The fact that these loans will be bought out of the pool mitigates the risk of co-sharing security and the increased potential credit risk arising from further advances. Given that the seller is an SPE, with limited resources to meet its obligations, the seller will purchase these loans from the pool before any other claims or further advances are originated. Porting The transaction documents allow porting, in which the originator may agree with the borrower to replace its existing mortgage loan with a new one (with the same mortgage conditions), except with a different asset. This could lead to an increase in the pool's credit risk, particularly if the new property results in a higher LTV or loan-to-income ratio. We have been informed that a request to port would have to fall within the applicable mortgage conditions. The LTV ratio on the new mortgage must meet the agreed upon conditions of the underwriting criteria, and the LTV ratio of the new mortgage may not exceed the maximum LTV ratio as stated in the then current underwriting criteria. Similar to other claims and further advances, we would expect the seller to buy these loans out of the pool before porting occurs, if the original loan to original foreclosure value of the new mortgages loan is more than the existing loan. Savings mortgages set-off risk If a mortgage borrower's savings policy provider becomes insolvent, they may be able to set-off any loss resulting from their savings policy repayments against their mortgage loans. In most cases, we are assured that this set-off risk is mitigated due to the sub-participation agreement between the savings provider and the issuer. This is because the borrowers pay savings premiums to the savings provider, which are in turn paid to the issuer in return for a savings participation in the relevant loan. In this transaction, we have been informed that some of the savings providers may not enter into a sub-participation agreement before closing. However, in order to mitigate set-off risk, a reserve fund FEBRUARY 26,

10 will be funded at closing, equal to the original balance of the pool's savings loans, where the savings policy provider hasn't entered into a sub-participation agreement. Origination And Underwriting GE Artesia Bank is a Dutch financial institution based in Rotterdam and Amsterdam. It was founded in 1863 as the Nederlandsche Credit and Deposit bank. In 1998, following the formation of Artesia Banking Corp., the name changed to Banque Artesia Nederland N.V. In 2006, GE Commercial Finance took over the bank and the bank now operates as GE Artesia Bank. A third of the mortgage loans were originated through the Quion "Generic Funding Model", a well-known platform in the Dutch market. Intermediaries apply for mortgage loans and then receive bids from a competing panel of lenders (one of which is GE Artesia Bank). Quion administers the platform and underwrites the applications based on criteria that is pre-agreed with the lenders. GE Artesia Bank funds the approved loans through a wholly-owned SPE (Quion 10). GE Artesia Bank originated two-thirds of the portfolio's loans, most of which were originated via intermediaries, who had direct contact with borrowers. The majority of borrowers are high net-worth individuals with their own businesses. As a result, the average loans size ( 250,000) is higher than the market average. Furthermore, about 20% of borrowers are considered to be self-employed. In 2008, GE Artesia Bank stopped originating new mortgages. Quion continued to service the existing mortgages, except special servicing for the mortgages on the Ember Hypotheken 1 B.V., Ember Hypotheken 2 B.V. books (which a small team at GE Artesia Bank carried out). Following the sale of the mortgages to Cartesian, Quion (being sub-contracted by Ember) has undertaken all servicing with Ember Hypotheken 1 B.V. and Ember Hypotheken 2 B.V.'s special-servicing team from GE Artesia Bank moving to Quion. Quion was initially incorporated in 2002 as an independent mortgage servicer in Holland. Quion has organized its primary servicing operation into three separate teams. Each team is dedicated to individual clients and services mortgages through their entire life cycle. Quion has multi-skilled staff across the full range of primary servicing activities. This provides Quion with greater scalability, as the company can move staff from one team to another, depending on workload. Staff manage between 2,000 and 3,000 mortgages; a high number, in our opinion. However, according to Quion, the volume is manageable thanks to the automation of their servicing platform. The special servicing department is split into five teams that carry out the full range of special servicing activities from early arrears to possession, property sale, and where appropriate, shortfall recovery. Quion focuses on treating customers fairly and encourages its staff to ascertain the reasons for a customer's financial problems to achieve an amicable and quick resolution. This could result in the sale of the property if there is no other outcome. As with other Dutch servicers, the emphasis is on assisted rather than forced sale, as this generally improves the return. Foreclosed mortgages were around 5% of all sales (at the time of our servicer visit), which we view as positive. We are satisfied with Quion's capacity to perform its functions in the transaction. FEBRUARY 26,

11 Note Terms And Conditions The issuer will pay interest quarterly in arrears on the payment dates in January, April, July, and October of each year, beginning in July The notes pay interest at three-month EURIBOR plus a margin. All of the classes of notes have an April 2044 legal final maturity date. The class A to D notes' interest rate is equal to three-month EURIBOR plus a class-specific margin. The issuer will pay interest according to the interest priority of payments. We treated the class B and C notes as deferrable-interest notes in our analysis. Under the transaction documentation, the issuer can defer interest payments on the class B and C notes. Consequently, any deferral of interest would not constitute an event of default. While our preliminary 'AAA (sf)' rating on the class A notes addresses the timely payment of interest and the ultimate payment of principal, our preliminary ratings on the class B and C notes only address the ultimate payment of principal and the ultimate payment of interest. Security for the notes The noteholders will benefit from the security granted in favor of the security trustee, Stichting Security Trustee Cartesian Residential Mortgages 1. Three pledges to the security trustee from the issuer will indirectly secure the notes: The first includes all rights to the portfolio's mortgage loans. The second includes all issuer rights over the swap rights, issuer advisory rights, mortgage receivables purchase agreement rights, administration rights, servicing rights, insurance savings participation rights, and collection foundation rights. The third includes the issuer's claims on the issuer accounts (the issuer transaction account, the reserve account, the construction deposit account, the swap collateral account, and the cash advance facility stand-by drawing account). To pledge these security rights under Dutch law, the issuer will pay to the security trustee (through parallel debt) an amount equal to the aggregate of all its undertakings, liabilities, and obligations to the security beneficiaries under the relevant transaction documents. Additionally, there is a first-ranking right of pledge over the accounts receivables on the collection foundation account, which the collection foundation will pledge to the security trustee. There is also a first-ranking disclosed right of pledge from the seller to the issuer over the seller's rights to GE Artesia Bank's representations and warranties. Optional redemption/mandatory redemption On each interest payment date (IPD), the issuer will use principal collections to redeem the class A to E notes. The issuer will redeem the class S notes using excess spread. The class S notes' principal is subordinated to the replenishment of the cash reserve. If the class A to E notes are fully redeemed, any amounts in the cash reserve would be available in the interest priority of payments to redeem the class S notes. The issuer can fully redeem the class A to E notes on the first optional redemption date in April 2019, or quarterly thereafter, at their outstanding principal amounts plus accrued interest. FEBRUARY 26,

12 The issuer can also redeem the rated notes if the seller buys back the mortgage loans on any quarterly payment date when the mortgage balance is lower than 10% of the closing mortgage loan balance. Collateral Description The provisional collateral portfolio consists of loans secured on first-ranking (or first- and consecutive-ranking) mortgages over residential properties in the Netherlands. The mortgage loans are of various product types: Interest-only, bank savings, savings, annuity, life, investment, linear, switch, and combinations of these (see below). Chart 2 The provisional portfolio has an outstanding principal balance of 481,519,988, which comprises 2,055 loans as of Nov. 30, The provisional portfolio We consider the preliminary pool to be well-seasoned, with a weighted-average seasoning of 105 months. Our weighted-average indexed current LTV ratio for the portfolio is %, which includes recent house price declines and our assumption of valuation dates (see below). The transaction's audit did not cover the valuation date, a FEBRUARY 26,

13 key field that we rely upon in our analysis. As a result, we assumed the valuation dates for all loans to be in August 2008, the height of the Dutch housing market. Consequently, we used an indexed weighted-average LTV ratio of 105% in our credit analysis. The weighted-average LTV ratio for the portfolio without our data adjustment would be 95%. Chart 3 Of the principal balance outstanding, 62.43% represents interest-only and life mortgage loans. The portfolio is geographically diverse and predominantly exposed to urban areas, which tend to carry a lower risk of default. The largest geographic concentration is in Noord-Holland, representing 40.08% of loans by outstanding principal balance (see below). FEBRUARY 26,

14 Chart 4 A large proportion of GE Artesia Bank's clients comprise high net-worth individuals. The average loan is greater than in other comparable Dutch RMBS pools. Likewise, the property values tend to be greater than the average Dutch property, which may affect their liquidity upon a forced sale. We have addressed this risk in our credit analysis by adjusting upwards our WAFF and WALS estimates for loans with a balance greater than 1 million. The average outstanding balance is 233,621 and the maximum balance is 2,414,153 (total borrower exposure). Of the preliminary pool, 0.96% comprises loans with arrears greater than or equal to one month. We view these types of borrowers as having a higher risk of default. In line with our Dutch RMBS criteria, we have increased our WAFF accordingly to address this increased risk. Due to the high unemployment rate and weak Dutch economy, we expect the arrears levels in Dutch transactions to increase. We also performed an arrears analysis using historical information from the originator's mortgage book. Overall, we projected 1.74% total arrears in our credit analysis to mitigate this risk. In total, we have therefore modeled arrears of 2.71%. FEBRUARY 26,

15 Counterparty Risk Foundation collection account At closing, a collection foundation, a bankruptcy remote entity, will be established. Collections will be paid into the foundation account held in the collection foundation's name and maintained with ABN AMRO Bank N.V. These collections are then transferred every five days into the issuer transaction account, which is held in the issuer's name. We expect that the transaction documents will specify that the issuer must take remedial actions, including the replacement of ABN AMRO Bank as the foundation collection account provider with a suitably rated financial institution, if: Our long-term issuer credit rating (ICR) on ABN AMRO Bank falls below 'BBB', where the short-term rating is at least 'A-2'; or Our long-term ICR on ABN AMRO Bank falls below 'BBB+', if it does not have a short-term rating. Issuer transaction account The issuer will open the issuer transaction account with BNP Paribas Securities Services (Luxembourg Branch). We understand that the transaction documents will specify that the issuer must take remedial actions, including the replacement of BNP Paribas Securities Services (Luxembourg Branch) as the bank account provider with a suitably rated financial institution, if: At any time, our long-term ICR on the bank account provider falls below 'A', where the short-term rating is at least 'A-1'; or Our long-term ICR on the bank account provider falls below 'A+', if it does not have a short-term rating. Swap The issuer will enter into a swap agreement with BNP Paribas in order to hedge the interest-rate risk due to a proportion of the pool paying fixed and SVR rates and the notes paying EURIBOR. We expect that the transaction documents will specify that replacement option 3 under our current counterparty criteria will be effective at closing (see "Counterparty Risk Framework Methodology And Assumptions," published on June 25, 2013). In addition, in line with our criteria, the swap counterparty must take remedial actions. These actions include the posting of collateral and the replacement of BNP Paribas as the swap provider with a suitably rated financial institution, if our long-term ICR on the swap provider falls below 'A' (where the short-term rating is at least 'A-1'). Credit Structure A combination of subordination, the cash reserve, and excess spread on the mortgages will provide credit enhancement to the notes. Table 1 Credit Enhancement For The Notes Class Rating Size of class (%) Mil. (equivalent) Credit enhancement (%)* A AAA (sf) FEBRUARY 26,

16 Table 1 Credit Enhancement For The Notes (cont.) B AA (sf) C A+ (sf) D NR E NR S NR *Reflects subordination and funded cash reserve at closing. NR--Not rated. N/A--Not applicable. Cash reserve At closing, the issuer will fund the cash reserve using the class S notes' proceeds to 2.3% of the initial pool balance. Under the transaction documents, the reserve fund's replenishment to the reserve fund first target level will rank above the class B to E notes' interest. This level is 2.3% of the initial pool. This isolates the use of the reserve specifically for the class A notes. The reserve fund's replenishment to the reserve account second target level will rank subordinated to the class C notes' PDL. This level is 4% of the initial pool balance and will be topped-up using any available excess spread. The reserve fund is nonamortizing. This part of the reserve will therefore be available for the class B and C notes, as well as the class A notes. The balance of this reserve account is available on any IPD to pay senior fees, expenses, and interest, and to reduce any debit balances on the notes' respective PDLs. If the issuer uses the cash reserve, it is replenished up to the required amount from available excess spread, in accordance with the priority of payments. Liquidity support If the general reserve fund is insufficient to remedy any shortfall in senior fees or in the class A notes' interest, the issuer can use principal receipts to mitigate this. Principal deficiency ledger The cash administrator will establish and maintain a PDL, comprising five subledgers, one for each class of notes. PDL amounts will first be recorded in the class E notes' PDL up to the class E notes' collateralized outstanding amount. They will then be debited sequentially upward. The PDL is loss-based. It is recorded if any losses occur on a mortgage loan in the pool and the amount of principal used to remedy senior fee and/or interest shortfalls. Priority of payments The issuer will apply interest available amounts in the following order: Senior fees; Servicer, special servicer, issuer advisory servicer, guaranteed investment contract, and paying agent fees; Amounts due to swap provider; The class A notes' interest; The class A notes' PDL; The reserve amount's top-up to the reserve fund first target level (2.3%); The class B notes' interest; FEBRUARY 26,

17 The class B notes' PDL; The class C notes' interest; The class C notes' PDL; The class D notes' interest; The issuer advisor's subordinated fee amount; The reserve amount's top-up to the reserve account second target level (4%); The class D notes' PDL; The class E notes' PDL; Swap subordinated amount; Principal (after step-up date, allocated to class A, B,C, D, and E notes in sequential order); Redemption of the class S notes' principal; and The class E notes' interest. The issuer will apply principal available amounts in the following order: The class A notes' principal; The class B notes' principal; The class C notes' principal; The class D notes' principal; and The class E notes' principal. Hedging Risk Hedged proportion of the pool: Fixed-rate and SVR loans The notes' interest rate will be based on an index of three-month EURIBOR, while, of the pool: 11.50% contains loans linked to SVR; 26.96% to EURIBOR; and 62.54% comprises fixed-rate loans that will revert to either SVR, or another fixed rate at the reset date. This leads to both basis and interest-rate mismatch risk. For the proportion of the pool that consist of SVR and fixed-rate loans, the issuer will enter into a swap agreement with BNP Paribas to hedge the interest rate risk between the rate of interest that the issuer receives on the mortgage loans and the interest-rate that it pays on the notes. Under the swap agreement, the issuer will pay to the swap counterparty the scheduled interest and any prepayment penalties, less senior fees and expenses the issuer owes (capped at 0.27% of the swap notional), less excess spread of 0.70% per year. In turn, BNP Paribas will pay to the issuer the weighted-average interest rate of the class A to D notes' outstanding principal balance, excluding any PDLs. This is all paid on the notional of the performing balance (loans that are not in arrears for more than 30 days) of the pool's SVR and fixed-rate assets. Unhedged proportion of the pool The three-month EURIBOR-linked loans, which account for 26.96% of the pool are not included in the notional of the interest rate swap. Although the notes are also linked to three-month EURIBOR, a basis risk mismatch arises here, as FEBRUARY 26,

18 the three-month EURIBOR reset date on the notes differs from those on the loans. As the transaction's exposure to basis risk has not been hedged, we have addressed this through stressing the cash flows from the assets. We stressed this by taking the historical difference between the index paid on the assets and the liabilities. Our analysis then takes the 95th, 90th, and 65th percentiles of the resulting distribution for the 'AAA', 'AA', and 'A' rating levels respectively. Payment Structure And Transaction Features We stressed the transaction's cash flow to test the credit and liquidity support provided by the assets, subordinated tranches, cash reserve, and any external sources (such as a cash advance facility). We implemented these stresses in our cash flow analysis at all relevant rating levels. Amount of defaults and recoveries For each loan in the portfolio, we estimated the likelihood that the borrower will default on its mortgage payments (the foreclosure frequency), and the amount of loss upon the subsequent sale of the property (the loss severity, expressed as a percentage of the outstanding loan). We assume the total mortgage balance to default. We determine the total amount of this defaulted balance that is not recovered for the entire portfolio by calculating our WAFF and WALS. Our WAFF and WALS estimates increase in tandem with the respective rating levels, because the higher the rating on the notes, the higher the level of mortgage default and loss severity they should be capable of withstanding. We based our credit analysis on the loans' characteristics and the associated borrowers. We have applied market-specific criteria in our assessment of the WAFF and WALS for this portfolio (see below). Table 2 Portfolio WAFF And WALS Rating level WAFF (%) WALS (%) Credit coverage (%) AAA AA A BBB Default timings Our WAFF assumptions at each rating level quantify the mortgages' total balance assumed to default over the transaction's life. We assume that these defaults occur over a three-year recession. Furthermore, we assess the effect of the recession's timing on the ability to repay the liabilities, and choose the recession start period based on this assessment. Although the recession normally starts in the first month of the transaction, the 'AAA' recession is usually delayed by 12 months. We applied the WAFF to the principal balance outstanding at the start of the recession (e.g., in a 'AAA' scenario, we applied the WAFF to the balance at the beginning of month 13). We assume defaults occur periodically in amounts calculated as a percentage of the WAFF. The timing of defaults generally follows two paths, referred to here as "fast" and "slow" defaults (see below). FEBRUARY 26,

19 Table 3 Default Timings For Fast And Slow Default Curves Recession month Fast default (percentage of WAFF; %) Slow default (percentage of WAFF; %) Recovery timings We assumed that the issuer would regain any recoveries 18 months after a payment default. The value of recoveries at the 'AAA' level will be 100% minus the WALS given above. We base the WALS figures in our cash flow model on principal loss estimates, including costs. We assumed no recovery of any interest accrued on the mortgages during the foreclosure period. After we apply the WAFF to the balance of the mortgages, the asset balance is likely to be lower than that on the liabilities (a notable exception is when a transaction relies on overcollateralization). We consider that the interest reduction arising from the defaulted mortgages during the foreclosure period could be mitigated by other structural mechanisms in the transaction (e.g., the cash advance facility and the cash reserve). Delinquencies We model the liquidity stress arising from short-term delinquencies, i.e., those mortgages that cease to pay for a period of time but then recover and become current on both interest and principal payments. To simulate the effect of delinquencies, we assume a proportion of interest receipts equal to one-third of the WAFF to be delayed. We apply this in each month of the recession and assume that full recovery of delinquent interest will occur 18 months after it is removed from the transaction. Therefore, if in month five of the recession, the total collateral interest expected to be received is 1 million and the WAFF is 30%, 100,000 of interest (one-third of the WAFF) will be delayed until month 23. Interest and prepayment rates We model three different interest rate scenarios rising, falling, and stable using both high and low prepayment assumptions. Interest rates were 0.25% at the time of modeling and we modeled them to rise or fall by 2.0% each month to a high of 12.0% or a low of 0.0%. For stable interest rates, we keep the interest rate at the current rate throughout the transaction's life. In the 'AAA' scenario, we modeled the interest rate increase to begin in month 13. We assume prepayment rates to be static throughout the transaction's life, and apply them monthly to the decreasing mortgage balance. We reserve the right to increase the high prepayment assumption if historical prepayment rates are at high levels, or if the transaction is particularly sensitive to high prepayments (e.g., the transaction relies heavily on excess spread). In this transaction, 62.54% of the pool contains fixed-rate loans that will revert to either a SVR, another fixed rate, or refinance at the reset date. We have observed that the historical prepayment rates on the reset date are very high, due FEBRUARY 26,

20 to the high interest rates on these products, which are typically 1-1.5% higher than the market. We have therefore also run sensitivity tests with higher prepayment assumptions at the reset date for the fixed loans. The combination of default timings, interest rates, and prepayment rates described above result in 12 different scenarios (see below). Our preliminary ratings address timely interest and ultimate principal under each of the 12 scenarios at the proposed rating levels for the class A notes. Our preliminary ratings on the class B and C notes only address the ultimate payment of principal and the ultimate payment of interest. Table 4 RMBS Stress Scenarios Scenario Prepayment rate Interest rate Default timing 1 Low Flat Fast 2 Low Up Fast 3 Low Down Fast 4 Low Flat Slow 5 Low Up Slow 6 Low Down Slow 7 High Flat Fast 8 High Up Fast 9 High Down Fast 10 High Flat Slow 11 High Up Slow 12 High Down Slow Scenario Analysis Various factors could cause downgrades of rated RMBS notes, such as increasing foreclosure rates in the securitized portfolios, house price declines, and changes in the portfolio composition. We analyzed the effect of house price declines by testing the sensitivity of the transaction to two different movement levels. House price decline analysis Declining house prices generally lead to higher loan-to-foreclosure-value (LTFV) ratios and more borrowers entering negative equity. This may increase the default probability of a securitized portfolio and its associated loss severity. Consequently, depending on its effect, declining house prices could be a contributing factor in the downgrade of rated notes. In our analysis, assumptions for house price declines will affect the calculation of both the WAFF and WALS. The house price decline analysis assumes house price declines that are relevant and specific to a jurisdiction rather than being uniform across all European transactions. This does not reflect any views on whether such house price declines will materialize in the future; the levels merely represent scenarios relevant to the specific market. So, for example, the assumed (for the purposes of the house price decline analysis) house price decline levels for a country that has experienced significant house price growth over the past few years may be different from the ones assumed for a FEBRUARY 26,

21 country that has experienced stable house prices. We perform our analysis on a loan-by-loan basis. Hence, the effect of applying different levels of house price declines differs between transactions, given the different concentrations in LTFV ratio bands. Note that even in these house price decline scenarios, structural features in securitizations might mitigate these declines. Further house price declines of 5% and 10% In the first scenario, in addition to the 'AAA' stress assumption, we apply an additional 5% decrease in house prices. We have applied the same 5% house price decline to the 'AA' and 'A+' stress scenarios to test the effect on the ratings on each class of notes. In the second scenario, we apply a 10% decrease in house prices. Table 5 Results Of Our House Price Decline Scenario Analysis House price environment WAFF (%) WALS (%) Rating level --Class A notes in a 'AAA' Scenario -- 'AAA' initial assumptions AAA Additional 5% house price decline AA+ Additional 10% house price decline AA+ --Class B notes in a 'AA' scenario -- 'AA' initial assumptions AA Additional 5% house price decline AA Additional 10% house price decline AA --Class C notes in a 'A' scenario -- 'A' initial assumptions A+ Additional 5% house price decline A+ Additional 10% house price decline A Under the first scenario, the ratings on the notes in the transaction would not suffer a ratings transition of more than one rating category. For example, the 'AAA (sf)' rated notes would achieve a rating of at least 'AA' under our credit stability criteria (see "Methodology: Credit Stability Criteria," published on May 3, 2010). Under the second scenario, the ratings on the notes would not suffer a rating transition of more than three rating categories (for example, the 'AAA (sf)' rated notes would achieve a rating of at least 'BBB'). We based our scenario analysis on simplified assumptions, with these assumptions materializing immediately on the day after closing. In reality, these are likely to occur over a period of time. Therefore, other factors, such as seasoning or repayments of some loans, could partially mitigate the effect of deteriorating performance of other loans. Sector Credit Highlights In the fourth quarter of 2013, 90+ day delinquencies increased slightly in the transactions included in our Dutch RMBS index. The small increase of four basis points (bps) came as the most recent macroeconomic indicators point to improving economic conditions, greater fiscal policy certainty, and increased affordability of housing, which seem to FEBRUARY 26,

22 have halted the slump in The Netherlands' housing market. Although we saw a promising fall in unemployment in October and November, it returned to Q3's rate of 7.0% in December. The increase in the unemployment rate during 2013 was one of the highest in Europe. On a positive note, consumer confidence returned to the highest level seen since early We do not think that the Q rise in house prices heralds a strong recovery of the housing market. Fiscal modification should make housing a less attractive investment for households, the number of house sales remained high at 212,000 in December 2013 compared with 121,000 in September 2008, and credit constraints and banking sector reform will also likely limit the market's recovery. We expect house prices to be flat in 2014, followed by growth of 2% in 2015 (see "Europe's Housing Markets May Be On A Slow Path To Recovery," published on Jan. 22, 2014). The economy returned to growth in the third quarter, even if growth was marginal at 0.2% quarter-on-quarter, following an increase from Statistics Netherlands' initial estimate of 0.1%. But, in November, we lowered our long-term unsolicited sovereign credit ratings on The Netherlands (see "Long-Term Ratings On The Netherlands Lowered To 'AA+' On Weak Growth Prospects; Outlook Stable," published on Nov. 29, 2013). In 2014, we expect GDP to grow by 0.2%, followed by 1.1% in 2015, which is lower than our previous forecasts. Despite the significant increase in the Dutch unemployment rate and an overall contraction in economic output in 2013, 90+ day delinquencies increased by only 5 bps in 2013, making the Dutch RMBS market one of the best performing jurisdictions in Europe. Meanwhile, prepayments increased to 4.19% from 3.71%. Surveillance The key performance indicators in the surveillance of this transaction will be: Total and 90-day delinquencies; Cumulative realized losses; LTFV ratios; Constant prepayment rates; and Increases in credit enhancement for the notes. Standard & Poor's 17g-7 Disclosure Report SEC Rule 17g-7 requires an NRSRO, for any report accompanying a credit rating relating to an asset-backed security as defined in the Rule, to include a description of the representations, warranties and enforcement mechanisms available to investors and a description of how they differ from the representations, warranties and enforcement mechanisms in issuances of similar securities. The Standard & Poor's 17g-7 Disclosure Report included in this credit rating report is available at FEBRUARY 26,