Structured Finance
Special Report
UK Non-Conforming RMBS: Catching a Cold?
Introduction The non-conforming RMBS market in the UK faces a number of challenges in the upcoming months. While some of these including affordability pressures faced by mortgage borrowers on the back of recent Bank of England base rate (BBR) increases have been on the radar screen for several months, others have been exacerbated by the recent crisis in the US sub-prime market. The fallout from this downturn has resulted in liquidity constraints for UK mortgage lenders - particularly those which lend to subprime borrowers - which predominantly rely on capital markets for funding needs. Lack of liquidity in the capital markets has not only resulted in market lending rates increasing but also a repricing of lenders’ product offerings - some of which have already been scaled back in scope. In this report, Fitch Ratings compares a number of elements in the UK non-conforming RMBS market with corresponding ones in the US sub-prime market and provide Outlooks on the back of what has transpired in the global markets over the past few months. Our main conclusions are: Macroeconomic Factors: current UK macroeconomic conditions do not differ significantly from those which existed in the US in 2006 on the eve of sub-prime problems emerging. The main difference was the cyclical position of the housing market and, in particular, house prices which declined (by some measures) through the course of 2006 in the US (whereas they recovered in the UK after a slowdown in 2005). UK house prices are expected to slow quite sharply from mid-2007 and remain very high relative to traditional valuation metrics. However one important mitigating difference with the US is that the UK did not experience the negative real policy interest rates witnessed in the US over 2003 to 2006 which may have contributed to excess credit in the US. Geographic Concentration: according to Fitch estimates, volumes of sub-prime mortgages are evenly distributed throughout the UK, whereas in the US there is heavy concentration in California, which has seen relatively higher house price declines. The high concentration of BTL in some UK regions - particularly London - is a concern, however, given that this market has never been tested in a market downturn. Origination and Servicing: statutory mortgage market regulation in the UK, whereby all major players to a mortgage transaction are regulated - including brokers - is a relative positive from an origination and underwriting perspective. Collateral Type: UK product characteristics is generally similar to the US - although the UK has not experienced the type of negative amortisation products which emerged in the US - while very high LTV positions in UK non-conforming remain minimal in comparison.
Analysts
Analysts
Gregg Kohansky +44 20 7862 4091 gregg.kohansky@fitchratings.com Ketan Thaker +44 20 7862 4124 ketan.thaker@fitchratings.com Stuart Jennings +44 20 7417 6271 stuart.jennings@fitchratings.com
Performance Analytics
Alison Ho +44 20 7862 4065 alison.ho@fitchratings.com
Origination and Servicing
Robbie Sargent +44 20 7862 4165 robbie.sargent@fitchratings.com Mark Wilder +44 20 7862 4165 mark.wilder@fitchratings.com
Quantitative Financial Research
Atanasios Mitropolous +44 20 7417 4317 atanasios.mitropoulos@fitchratings.com
Sovereign
Brian Coulton +44 20 7862 4097 brian.coulton@fitchratings.com Related Research • “Interest Rate Issues in European RMBS”, published 5 September 2007;
•
"UK Non-Conforming RMBS – Performance Review Q207", published 30 August 2007; "ResiLogic: U.S. Residential Mortgage Loss Model - Amended", published 14 August 2007; “House Prices and Household Debt Where are the Risks?”, published 29 July 2007; “UK Residential Mortgage Default Model Criteria”, published 5 February 2007.
•
•
•
2 October 2007
www.fitchratings.com
Structured Finance
RMBS Performance: while the asset performance Outlook is negative for UK non-conforming RMBS, the agency’s Outlook for transaction performance is negative only for recent vintages of UK nonconforming RMBS (where asset characteristics, such as LTV and DTI, are relatively worse and for those transactions with higher proportions of loans to borrowers with heavy adverse credit). The Macroeconomic Perspective Table 1: US and UK Economic Indicators
GDP Inflation (Dec on Dec) Policy interest rates Real policy interest rates Mortgage rates Unemployment Real household income Consumers expenditure Saving ratio House prices (Y on Y)
Source: Fitch
real house prices on a scale which exceeded that seen in earlier housing market booms (see chart 1). The UK also witnessed high rates of real house price inflation from the late 1990s - at a rate far exceeding that seen in the latest US boom - but this peaked in early 2003 with rates declining sharply in 2005 before recovering in 2006. Divergent movements in housing market conditions have been reflected in the pattern of mortgage credit growth. This declined sharply in the UK through 2005 before recovering last year, whereas it remained strong until Q1 2006 in the US before declining to 9% in Q4 2006 (see chart 2). The pace with which house price inflation has adjusted downwards in the US has contributed to recent problems in the housing sector, in particular creating negative equity problems on some loans made in 2005 and 2006. While the UK housing market cycle looks to be at a different stage - with the household sector having weathered the 2005 slowdown relatively well - it would be foolish to conclude that house price adjustments as seen in the US could not be repeated in the UK. This belief is grounded in a number of factors: firstly recent increases in GBP LIBOR have resulted in a de facto monetary policy tightening (though BBR is expected to have reached its peak), in contrast to the US where central bank rates have recently been lowered. More importantly, real house price inflation has been much more volatile in the UK than in the US historically and, with the exception of Q2 2005 to Q1 2006, it was above that in the US in every quarter after Q2 1996 (chart 1). Rapid house price growth has seen the ratio of house prices to household income reach unprecedented levels in the UK, currently over 40% higher than the average between 1967-2006 (chart 3) and significantly above the peak seen before the late 1980s crash. The house-
US 2006 2.9 2.5 5.0 1.7 6.6 4.6 2.6 3.1 0.4 -2.8
UK 2007q2 3.1 1.8 5.4 2.8 7.3 5.4 1.9 2.8 3.1 10.7
As the table shows there is relatively little to choose between current UK macroeconomic performance (Q2 2007) and that in the US in 2006 when subprime problems emerged. Both economies grew at or slightly faster than trend, unemployment was low by historical standards and interest rates were around 5 to 5.5%. US household saving was much lower at 0.4% of disposable income compared to 5% in the UK. The most striking contrast in economic performance was with regard to house prices. These grew 9.3% in the UK (according to the Nationwide Building Society) in the year to Q2 2007 but in the US the NAR index declined by 2.8% in the year to Q4 2006. The downturn in US house prices in 2007 followed five years of strong sustained growth in
Chart 1: House Prices Real
US Real (%) 40 30 20 10 0 -10 -20 -30 Q167 Q369 Q172 Q374 Q177 Q379 Q182 Q384
UK Real
Q187
Q389
Q192
Q394
Q197
Q399
Q102
Q304
Q107
Source: National Association of Realtors, Nationwide, Datastream
UK Non-Conforming RMBS: Catching a Cold?: October 2007 2
Structured Finance
Chart 2: Mortgage Loans Outstanding
Annual Growth
(%) 25 20 15 10 5 0 Q188 Q189 Q190 Q191 Q192 Q193 Q194 Q195 Q196 Q197 Q198 Q199 Q100 Q101 Q102 Q103 Q104 Q105 Q106 Q107 Source: ONS, Fed, Datastream UK mortgages US Mortgages
price income ratio also reached high levels in the US in 2006 but it was only 20% above its long run average, in line with its previous peak in 1979. Please see “House Prices and Household Debt Where are the Risks?”, published 29 July 2007 on www.fitchratings.com for more information as to which economies look most vulnerable to the combination of weakening property prices and rising interest rates. It is far from clear that the finances of UK households in aggregate leave them better placed to cope than US households with any downward adjustment in house prices. Household debt is significantly higher as a share of income in the UK at 146%. Net financial assets (ie excluding real estate and other tangible assets) were 298% of income in the US compared to 283% in the UK (see table 2). With regard to the “flow” burden of debt service, Federal Reserve data show the ratio of debt service payments to disposable income for the US household sector in aggregate to have reached 14.5% in 2006, a record high. It is hard to find comparable figures for
the UK. The figure of 11.3% (which relates to 2006) in table 2 below combines interest payments with regular repayments on repayment mortgages, but excludes repayments on unsecured consumer debt (which are incorporated in the US data) and “investment” payments on endowment mortgages. Given the sharp rise in unsecured debt in recent years it is likely that the UK debt service burden is somewhat higher than in the US - for example if it is assumed that the average maturity of unsecured credit is five years, this would add around five percentage points to the debt service ratio in the UK. Nevertheless, the UK aggregate household debt service burden would still appear to remain lower than that seen in the early 1990s, in contrast to the situation in the US.
Chart 3: House Prices/Income Per Head
US (%) 150 140 130 120 110 100 90 80 70 60 50 Q267 Q270 Q273 Q276 Q279 Q282 Q285 Q288 Q291 Q294 Q297 Q200 Q203 Q206 UK
Source: Nationawide, NAR, ONS, Datastream, Fitch
UK Non-Conforming RMBS: Catching a Cold?: October 2007 3
Structured Finance
Chart 4: Real Interest Rates
UK Policy (%) 12 10 8 6 4 2 0 -2 -4 Jan 89 Nov 90 Sep 92 Jul 94 May 96 Mar 98 Jan 00 Nov 01 Sep 03 Jul 05 Jun 07 US Policy
Source: Datastream, Fitch
Table 2: Household Balance Sheets Compared 2006
Disposable income (%) Debt Net worth Net financial assets Debt service US 132.4 574.6 297.7 14.5 UK 146.3 786.8a 282.5 11.3a
a Fitch estimate Source: ONS, Bank of England, Federal Reserve, Datastream, Fitch
account deficit to 6.1% of GDP in 2006 - compared to 3.7% in the UK - with the national savingsinvestment shortfall exacerbated by the collapse in the household saving ratio. This suggests that macroeconomic policy may have played a more significant role in fuelling excess credit and asset price growth in the US than in the UK over the last few years. Affordability Spotlight Over the past decade UK house price growth has strongly outpaced average income growth, stretching the affordability of mortgage borrowers, particularly first time buyers.
UK House Price/Income
However, there are some grounds for believing that macroeconomic conditions lessen the chance of housing sector problems emerging in the UK, especially when UK monetary policy is compared to the volatility of US monetary policy in recent years and the associated imbalances that have emerged in the economy. In particular, the Fed’s aggressive rate cuts in 2001, sustained until mid-2004, resulted in a three year period of negative real policy interest rates. This had not been seen since the days of high and volatile inflation in the mid-1970s and was in sharp contrast to the UK, where real policy rates have fluctuated in a narrow range of 2% to 4% since mid2001 (chart 4). Loose monetary policy has contributed to the widening of the US current
Chart 5: Nationwide FTB Affordability Index
(1985 = 100) 160 140 120 100 80 60 40 20 0 Q183 Q384 Q186 Q387 Q189 Q390 Q192 Q393
The average UK house price was more than GBP218,000 in July 2007 (source: ODPM) compared to a price of around GBP100,000 in 2000. Hence, over this period as a whole, house price growth has averaged nearly 12% per annum, is significantly higher than average earnings growth of under 4% per annum (source: BoE labour market data).
Q195
Q396
Q198
Q399
Q101
Q302
Q104
Q305
Q107
Source: Datastream, Fitch
UK Non-Conforming RMBS: Catching a Cold?: October 2007 4
Structured Finance
Such a growth in house prices has led to an increase in the amounts buyers need to borrow relative to their income. Around a decade ago, a majority of buyers borrowed less than three times their income. Today, the situation has changed significantly with a majority of borrowers having an income multiple of greater than three, including a sizeable number of borrowers who are borrowing over four times their annual income. An important factor behind borrowers being able to borrow more for their income has been the relatively stable level of interest rates compared with earlier decades. Since 2000, BBR has varied between 3.5% and 6.0% with an average of 4.7%. This compares to a range of 5.0% to 15.0% and an average BBR of 7.9% in the 90s. The Bank of England increased rates from 4.50% to 5.75% between August 2006 and July 2007 in order to combat inflationary pressures within the UK. In addition to these rate rises, non-conforming lenders are also likely to pass on the increased cost in the GBP LIBOR market to borrowers. As can be seen in chart 5 (source: CML), these rate rises have stretched the affordability of first time buyers almost to the point seen at the time of the last housing market downturn in the early 1990s. Since borrowers now have larger mortgages, any further rate rises will significantly affect their affordability. Please see “Interest Rate Issues and European RMBS”, published 5 September 2007 on www.fitchratings.com for more information as to potential payment shock to UK non-conforming mortgage borrowers. An increasing proportion of mortgage loans originated in the UK over the past few years have had an initial two or three year teaser fixed/discount rate (from approximately 30% in 2004 to approx 80% in Q2 2007 according to CML data). Fitch has estimated that nearly 45,000 sub-prime borrowers have mortgages which roll off to variable rates in Q4 2007 (in addition to 35,386 in Q1 2008 and 39,715 in Q2 2008). These borrowers are most exposed to mortgage payment shock. Even if these borrowers manage to find a new mortgage deal with teaser rates, the rates are likely to be much higher than the rate they were paying till now. BoE base rate is now 1.25% higher compared to when these products were originated. Additionally, almost all sub-prime lenders are likely to pass on their increased cost of borrowing, due to capital market turbulence, to the borrowers. Fitch's review of the current product range offered by the top UK non-conforming lenders shows that on average a sub-prime borrower opting for a new two year fixed
UK Non-Conforming RMBS: Catching a Cold?: October 2007 5
deal at the end of their teaser period on an existing mortgage is likely to pay around 1.75% to 2.00% higher than the teaser rate offered at the time of their previous mortgage.
Chart 6: Quarterly Sub-Prime Originations in Fitch Rated Pools
Discount (LHS) Fixed (LHS) Variable (LHS) 90%+LTV Loans (RHS) Heavy/Unlimited Adverse Loans (RHS)
70 Initial Rate Type (%) 60 50 40 30 20 10 0 Q305 Source: Fitch Q405 Q106 Q206 Q306 Q406 0 10 5 20 15
Of most concern is the likelihood that, due to the recent tightening in criteria for sub-prime lending, borrowers with poor credit histories might not be able to find another mortgage deal and would be stuck with a high rate on their existing loan when it reverts. In the past, owing to strong house price appreciation and easy availability of credit, even borrowers with severe credit problems (and even in arrears) have had little difficulty in being able to remortgage. However, due to recent market turmoil there is evidence that a number of lenders are repricing their products and tightening their lending criteria.
Chart 7: Number of N/C Lenders Offering Product Pre Summer Post Summer
(No. of Lenders) 12 10 8 6 4 2 0 85% LTV 85% LTV 90% LTV 90% LTV 90% LTV 95% LTV Near Light Light S/C Near Heavy Medium Prime Prime S/C S/C Source: Fitch
Proportion of Loans Originated (%)
80
25
Structured Finance
Also as can be seen in chart 7 a number of lenders have withdraw from the riskier segments of the market. All these factors are likely to make it increasingly difficult for sub-prime borrowers to be able to re-mortgage as they have been doing in the past.
Chart 8: Average Reversionary Margin Change Pre Summer Post Summer
(%) 6 5 4 3 2 1 0 85% LTV 85% LTV 90% LTV 90% LTV 90% LTV 95% LTV Near Light Light S/C Near Heavy Medium Prime Prime S/C S/C Source: Fitch
performance of UK non-conforming RMBS. Geographic Concentration Analysis
Regional Repossession Study
Across Fitch rated pools, over 78,000 sub-prime loans originated in Q4 2005 and Q1 2006 paid on an average an initial teaser rate of 6.05%. Over the next two quarters if these borrowers are unable to remortgage to a new teaser rate deal the interest rate on an average across these loans is likely to increase by a steep 2.75%. On a typical repayment mortgage these could increase the monthly mortgage payment by over 27%. This payment shock could result in many sub-prime borrowers being pushed into arrears or defaults. Chart 9 shows the DTI impact where a borrower mortgages to a higher fixed rate deal or the DTI impact where the borrower is unable to remortgage to a new teaser rate deal. Please see RMBS Performance below for details regarding how such payment shocks can impact
Chart 9: Impact on DTI
(%) 50 45 40 35 30 25 Nov Dec Jan Feb Mar 05 05 06 06 06 Source: Fitch Apr May Jun 06 06 06 Jul 06 Fixed Rate Loan: From 6% TO Libor+2.75% 6% Fixed remortgaging to 7.75% Fixed
While US loan performance has deteriorated dramatically, starting with the most risky market segment but also filtering through to less risky segments over time via home price depreciation, performance of loans in the UK has already been on the decline for quite some time. An inspection of dynamic repossession order rates on a regional level shows that repossession orders made have been rising for more than three years, with London rates having bottomed as early as 2001. More recently, however, the increase in repossession orders granted has accelerated significantly. Based on the most recent figures published by the Ministry of Justice the agency estimates for some regions the rate to have risen beyond 1% of the mortgaged housing stock. These are levels last experienced only during the recession starting in 1989/1990. While there still remains a significant gap between repossession orders granted and actual repossessions executed, such that actual repossessions remain relatively lower, in the event of a downturn these two data sets are more likely to start to converge as other options which may prevent actual repossession occurring (for example, reaching an arrears repayment plan with the borrower or voluntary sales) may begin to become more difficult to achieve. While the current trend points towards levels approaching those experienced during the last recession, attributes of the housing market are different now to those at the time of the last recession. Indeed, the current repossessions in the UK comprise a substantial portion attributable to the non-conforming sector. This therefore appears more closely related to the current situation in the US rather than conditions during the late 1980s in the
Discount Loan: Libor+2.5% with 2% teaser discount
End of teaser period for Fixed & Discount loans Aug Sep Oct Nov Dec Jan Feb Mar 06 06 06 06 06 07 07 07 Apr May Jun 07 07 07 Jul 07 Aug Sep Oct 07 07 07 (E) (E)
UK Non-Conforming RMBS: Catching a Cold?: October 2007 6
Structured Finance
Chart 10: Rate of Repossession Orders Granted
(%) 2.0 East Midlands North West South West East North East West Midlands London South East Yorkshire and the Humber
1.5
1.0
0.5
0.0 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 H107e Source: Datastream, Fitch
UK when the non-conforming sector did not exist. Correspondingly, an overall increase in repossession orders cannot be seen as a surprise. Purely looking at trends in repossessions does not tell the complete story about the state of, and the prospects for, the housing market. Factors such as house price appreciation and geographic concentration of riskier mortgages must be considered as well.
House Price Appreciation
The development of house prices is a crucial factor for the projection of loan performance. It is not only recovery rates that reduce as soon as house prices turn downwards. A reduction in house prices also reduces the effective equity of the borrower within the property, thereby adversely affecting the borrower’s refinancing opportunities as soon as he or she comes into financial distress (and reducing the borrower’s incentive to continue payments on the loan). One of the regions that has been hit particularly severely by the current market downturn in the US is California. Comparing the house price appreciation
on a vintage basis with that experienced by London during the last recession shows remarkable similarities. The timing of the downturn coincides pretty well, though the deterioration of London at the time appears to have been somewhat more severe for the 1988 and 1989 vintages than is the case for the Californian 2005 and 2006 vintages . The London graph uses the regional HPI according to the Nationwide Building Society to calculate the regional price inflation by origination quarter, while the US graph uses the Case-Shiller-Weiss data.
Chart 11: California Home Price Inflation by Origination Quarter
Q105 (%) 20 15 10 5 0 -5 3 Source: Fitch, CSW 6 9 12 Months 15 18 21 24 Q205 Q305 Q405 Q106 Q206 Q306
UK Non-Conforming RMBS: Catching a Cold?: October 2007 7
Structured Finance
Chart 12: London House Price Inflation by Origination Date
Q188 (%) 20 15 10 5 0 -5 -10 -15 3 Source: Fitch 6 9 12 15 Months 18 21 24 27 Q288 Q388 Q488 Q189 Q289 Q389 Q489
In order to capture early signs of house price deterioration Fitch reviewed the latest house price information on a very granular basis as provided by the Financial Times House Price Index, compiled by Acadametrics. However, while most indices indicate a flattening environment on the aggregate (see chart 13) an inspection of indices on a granular basis reveals that the picture is rather more diverse, with some regions already experiencing house price declines while others continue to rise (see graph on Greenwich, currently the worst performing London Borough, and Kensington and Chelsea, the best performing). Overall, according to this data, UK house prices are not showing any signs, to date, of dramatic deterioration. However, it should be noted that all indices are only updated until August, prior to the liquidity crisis of that month and the impact on public confidence that this may have had. The impact of the August crisis on house price indices therefore still remains to be seen - particularly since non-conforming lenders have tightened lending criteria, eg by limiting LTVs for adverse credit borrowers.
Another influential factor in the development of house prices is the supply of housing. While in the US house price appreciation has caused a rapid surge in residential construction, supply of housing has always been slow in responding to increased demand in the UK. Housing starts over the past six to seven years have increased at an annual rate of approximately 3% to 4%, roughly matching demographic developments. In particular, completions in London have matched fairly well net inward migration. This compares to an annual increase in housing starts of approximately 8% in the US before they plummeted in early 2006 (according to figures by the National Association of Realtors). This is providing a mild mitigant against the prospect of reducing demand.
Chart 13: UK HPI Indices
Q105 (%) 20 15 10 5 0 -5 3 Source: Fitch 6 9 12 15 Months 18 21 24 27 Q205 Q305 Q405 Q106 Q206 Q306 Q406
UK Non-Conforming RMBS: Catching a Cold?: October 2007 8
Structured Finance
Chart 14: Greenwich House Price Inflation by Origination Date
Q105 (%) 15 Q205 Q305 Q405 Q106 Q206 Q306 Q406
10
5
0
-5 3 Source: Fitch 6 9 12 15 Months 18 21 24 27
Regional Risks In order to get to an informed opinion about underlying risks in mortgage portfolios it is essential to study risk concentrations in pools. Fitch conducts a review of any securitised pool via its default model prior to assigning any rating which will assign default loadings to specific loans based on characteristics. The agency will also perform detailed reviews of portfolio performance, which will become more intensive and detailed if asset performance appears to deteriorate on an aggregate level. This further review reports on an additional analysis conducted an on additional review of a pool of more recent transactions, providing more market-wide insight into originations across the country. The study attempts to arrive at estimates for buy-to-let and non-prime lending concentrations on a regional and sub-regional level, including determination of key risk-characteristics such as average loan-to-value ratios, debt-to-income ratios and proportion of self-
certification.
Methodological Remarks
The following figures are based on a pool of more than three million loans compiled from origination information from 13 prime and 48 non-prime transactions. Among the prime transactions, most of the observations stem from three large Master Trust programmes, while the remainder is from four pure buy-to-let transactions and two mixed transactions. The non-prime transactions are each significantly smaller in size and pool buy-to-let/owner occupied and near-prime/sub-prime loans. When calculating estimated concentrations across regions and subregions, the agency makes use of aggregate information provided by the CML, the BoE and Datamonitor regarding estimated distributions across portfolio segments. For a more detailed account of the methodology used please refer to Appendix 2. The compiled pool is selected such that it comprises a representative sample of recent securitisations. It is cautioned not to over-interpret the numbers as the securitised sample is not a complete representation
Chart 15: Kensington and Chelsea House Price Inflation by Origination Date
Q105 (%) 50 40 30 20 10 0 -10 3 Source: Fitch 6 9 12 15 Months 18 21 24 27 Q205 Q305 Q405 Q106 Q206 Q306 Q406
UK Non-Conforming RMBS: Catching a Cold?: October 2007 9
Structured Finance
of the entire mortgage market and selection bias can skew the results. Equally, no ‘hard and fast’ figures exist regarding market statistics from third-party sources concerning percentages of the mortgage market attributable to non-conforming and buy-to-let are estimates However, the agency tries to address this shortcoming in a number of ways: firstly Northern Ireland and Scotland have been excluded from the analysis as some lenders are not originating in these areas. Secondly, all regional aggregates for which the number of observations does not exceed 500 are also excluded. Finally, figures for the subregional level are interpreted as indicative evidence only and should not be viewed as a definitive statement for the entire UK mortgage market.
Buy-to-Let
often let by the owner without notifying the bank of this change in tenure in order not to be charged a potentially higher interest rate. This is often referred to as “disguised buy-to-let”. Equally risky, if not more so, are those loans where the borrower would not meet underwriting criteria on affordability for an owner-occupied mortgage and takes out a buy-to-let mortgage which allows him or her to meet eligibility requirements on interest coverage without actually intending to let the property. (Conversely to the above, these loans might be referred to as “disguised owner-occupied”.) While the following paragraph is not looking for evidence of a deterioration in buy-to-let performance, it tries to determine those regions that are most exposed to concentration risk from the buy-to-let segment. For this purpose, buy-to-let lending is assumed to be 11.37% of overall mortgage lending, as derived from data from the CML and the Bank of England for 2006 and H1 2007. This figure is used to scale the regional proportion of buy-to-let originations to the national level, but otherwise does not affect the relative scaling of concentrations between regions. For a detailed account of the methodology employed please refer to Appendix 2. Table 3: Estimated Buy-to-Let Concentration Among all Originations by Region
Rank 1 2 3 4 5 6 7 8 9 10 11 Region London Other South East North West Outer Metro South West East Anglia East Midlands Yorks and Humber Wales West Midlands North % BtL among all 20.4 12.1 10.6 10.4 10.0 9.9 9.5 9.2 9.2 8.9 8.0
Over the past two to three years, performance of buy-to-let mortgages has been remarkably good, often outperforming owner occupied borrowing even after correcting for typically lower loan-to-value ratios and stricter affordability requirements. This has led a growing number of market participants to claim that buy-to-let mortgages are generally of better quality and should be expected to continuously show fewer delinquencies and defaults than owner occupied mortgages. This view has been supported by the mainly theoretical argument that most buy-tolet investors are older, more financially experienced and view their asset as a long-term investment. This view has been criticised by some economic analysts who point out that the rental return of buyto-let assets is already much lower than alternative investments, meaning that such an investment can only be sustainable under expectations of strong house price growth, and that rental income streams are much more volatile and sensitive to general economic prospects. In particular, the growth in buyto-let properties has so far been matched by an equally strong demand for rental accommodation, driven by immigration into the metropolitan areas and by the growth of single households, often created by students and young professionals leaving their parental home. The argument continues that these are exactly the population segments that are most likely to flow out of the respective regions as soon as economic prospects deteriorate. If capital appreciation were threatened then there exists the possibility that some buy-to-let investors might start looking to liquidate properties and ‘lock-in’ paper profits in cash. If this were to occur then any house price declines could be exacerbated by any buy-to-let concentration in the region. Critics of the growth of buy-to-let lending also point towards the ‘blurring’ between buy-to-let and owneroccupied mortgages. Owner-occupied properties are
UK Non-Conforming RMBS: Catching a Cold?: October 2007 10
Source: Fitch, Transaction pool cuts
This regional data clearly suggests a strong buy-tolet concentration in London and the South-East, with London having the largest exposure, while other regions show proportions in a rather narrow band of between 8% and 11%. This picture is confirmed when looking into postcode area levels with the top 11BTL concentrations of the 105 postcode areas all situated within Greater London. However, a closer look at a sub-regional level reveals that there are more hotspots of buy-to-let lending across the country. Indeed, on a postcode district level figures show that some areas in Manchester, Liverpool and Nottingham have higher concentrations than any of the London districts. Still,
Structured Finance
the numbers suggest that, were the performance of the buy-to-let market to deteriorate, then London would be likely to suffer more than any other region. Table 4: Estimated Median Interest Coverage Ratios Among Buy-to-Let by Region
(%) East Anglia East Midlands London North West Other South East Outer Metro South West Wales West Midlands Yorks and Humber OLTV among BtL - median 83.33 84.29 81.85 85.00 82.19 82.61 80.60 84.69 83.70 84.85 ICR among BtL - median 132.70 135.80 140.20 133.50 138.60 137.70 134.50 134.80 136.00 132.10
Table 5: Estimated Non-Prime Concentration Among OwnerOccupied by Region
Obs 1 2 3 4 5 6 7 8 9 10 11 Obs for Region subprime Wales 129,789 North West 297,158 West Midlands 231,621 North 159,692 Yorks and Humber 240,028 East Midlands 185,330 London 295,744 East Anglia 96,705 South West 236,038 Other South East 294,911 Outer Metro 293,671 Non-prime Concentration (%) 9.6 8.8 8.6 8.6 8.0 7.7 6.2 5.8 5.6 5.4 5.3
Source: Fitch, Transaction pool cuts
Source: Fitch, Transaction Pool Cuts
However, this risk is partly mitigated by the fact that LTVs are lower and rental coverage greater in London and the South East than in other regions. As the following table demonstrates, London, Outer Metro and the South East show relatively low median loan-to-value ratios at origination (OLTV) and high median interest coverage ratios (ICR). For details regarding estimated BTL mortgage geographic concentration within the UK, please refer to Appendix 1.
Non-Prime
The recent market turmoil in the US has shown that a minority segment of the entire mortgage market can have a significant effect on the whole. The credit crunch experienced on a global level across financial markets was triggered by the deterioration of the performance of so-called sub-prime loans, ie loans to borrowers known to have previously had difficulties in servicing financial liabilities. While this market was estimated to account for approximately 16% of entire mortgage originations, the deterioration of credit quality caused house prices to reduce in many US states and to produce significant losses even on higher credit quality segments (Alt-A). This happened despite the performance of prime loans, which form the majority of the market, remaining remarkably stable.
In the US, on a regional level, reviews of mortgage data have found that California has had the largest proportion of non-prime loans and, as a consequence, suffered more than most from the downturn in that segment. The following table attempts to identify which regions are particularly vulnerable to a deterioration of non-prime performance in the UK. Fitch assumes that on a national level, non-prime originations are at 7.1% of all lending to owner occupiers, as suggested by figures published by Datamonitor. This number is also fairly close to the share of approximately 6.6% found within the sampled pool. (Note, this aggregate number is used only for scaling the absolute proportion of non-prime among all owner-occupied lending and does not affect the order among regions.) The data on the level of regions show higher concentrations of non-prime originations in other areas of the country - such as Wales and the North West - than for London and the south. The south eastern regions outside London appear to have the lowest concentration of all. A further point to note is that the dispersion of concentrations seems less pronounced than for buy-to-let. A closer inspection at the postcode area level (see Appendix 1) shows that concentrations in Wales are largely attributable to Newport and Cardiff alone . It also confirms that even the regions with the highest concentrations show non-prime proportions that do not exceeding 11% of all owner-occupied originations. However, at the individual postcode district level (PCD), we find that non-prime hotspots exist within Manchester, Birmingham and Liverpool, some of which approach 20%. Some East London districts also exhibit non-prime proportions exceeding 15% of owner-occupied originations. At the other end of the scale, Central London areas
UK Non-Conforming RMBS: Catching a Cold?: October 2007 11
Structured Finance
Table 6: Loan Characteristics Among Non-Prime Owner Occupied Loans by Region
OLTV median (%) 79.53 80.92 83.72 79.45 79.85 80.45 82.56 79.65 79.28 80.05 80.00 DTI - median (%) 37.55 36.62 41.52 34.83 35.65 38.95 40.12 38.85 35.63 36.87 35.34 IM - median 3.0 2.9 3.1 2.8 2.9 3.0 3.1 3.0 2.9 2.9 2.9 Self Cert (%) 63.4 59.1 71.5 55.6 58.7 69.4 68.5 67.9 57.9 60.4 56.6 CCJ (%) 6.5 8.1 5.1 8.3 7.4 6.2 5.9 6.9 7.5 8.2 8.6
East Anglia East Midlands London North North West Other South East Outer Metro South West Wales West Midlands Yorks and Humber
Source: Fitch, Transaction pool cuts
show the lowest originations.
proportion
of
non-prime
Prime
A closer look into the characteristics of the loans reveals some further insights. Based on the sampled pool of more than 200,000 non-prime owneroccupied loan originations - and in contrast to buyto-let findings - we can clearly see that non-prime borrowers in London and its surroundings are much more exposed to risks from stretched affordability, with OLTV, DTI)and income multiples (IM) being the highest risks. Particularly highly stretched borrowers have also been noted in US regions which have subsequently been hit by deteriorating performance and house price decline. California serves as an example. As noted elsewhere in this report, the table also confirms that the proportion of self-certified loans among non-prime originations is clearly the highest in London and its surroundings, with approximately 70% having been originated in this way, (in the US this figure stood at 42% in 2006 for the nation as a whole, and 54% for California). However, in contrast, indicators of adverse credit information for the borrower at origination stage, such as the degree of county court judgments (CCJ), - as measured by the default probability adjustments applied by Fitch’s default model - indicate a higher risk in Northern Regions and Wales, rather than in London and South East. Adjustments for prior mortgage or rental arrears also do not appear exceptionally high in London. (For more details on the definition of loan characteristics and the criteria applied in rating RMBS transactions see the criteria report “UK Residential Mortgage Default Model Criteria”, dated 5 February 2007 and available at www.fitchratings.com). For details regarding estimated non-prime mortgage geographic concentration within the UK, please refer to Appendix 1.
UK Non-Conforming RMBS: Catching a Cold?: October 2007 12
As experienced in the historic downturn of the UK mortgage market during the late 1980s and early 1990s, stress situations may arise even in the absence of non-conforming lending. Risky prime lending, which is most evident in high loan-to-value ratios and stretched affordability measures, can easily be impacted by a deteriorating economic environment and lead to large losses realised on mortgage portfolios. Closer inspection of the sampled data reveals that OLTV does not appear to pose a significant concern among prime lending in general, as median ratios are typically below 70%. However, these figures combine both first-time buyers and re-mortgages and thus fail to highlight the lack of equity put down by first-time buyers with high LTV mortgages in regions with historically high house-price appreciation. The DTI and the income multiple (IM) however, clearly display a strong North-South divide as southern regions exhibit much more affordability issues than Northern regions. Table 7: Loan Characteristics Among Prime Owner Occupied Loans by Region
OLTV among DTI among prime prime median (%) median (%) 65.8 36.0 69.4 33.9 69.4 37.1 69.6 31.5 70.2 33.3 65.3 36.3 66.7 36.6 63.3 36.0 69.0 33.3 69.1 34.6 68.2 32.7 IM among prime median 2.9 2.7 3.0 2.5 2.6 3.0 3.0 2.9 2.6 2.7 2.6
East Anglia East Midlands London North North West Other South East Outer Metro South West Wales West Midlands Yorks and Humber
Source: Fitch, Transaction pool cuts
Structured Finance
In summary, the UK housing and mortgage market shows some similarities with developments in the US before the sub-prime crisis, such as the surge in the non-conforming sector, loosened underwriting criteria and highly appreciating house prices. In more detail however, concentrations in nonconforming activity in London seem to be counterbalanced with somewhat tighter underwriting. The most worrying element appears to be the to-date untested performance of buy-to-let investors. London would be hard hit if this segment of the market turned out to deteriorate. The strongest mitigants to risks are the continued inward migration into the UK and the stable construction business, which make large oversupply and rapidly falling house prices rather unlikely in the short term.
Origination and Underwriting
adequate security for the loan in the event that an applicant fails to pay. Furthermore, for the most part, lenders on both sides of the Atlantic price for risk when making loans, whereby interest rates charged to borrowers are higher for borrowers perceived as being high risk. The biggest development with respect to UK mortgage underwriting practices has been the introduction of statutory regulation by the Financial Services Authority (FSA) in 2004 on the back of which, transparency afforded to consumers applying for a mortgage has increased significantly. This has closed the mortgage regulation gap to the US, which has been monitored in various degrees at a federal and local government level for a number of years. However, whereas in the US only lenders and servicers are regulated, the FSA has introduced new rules to regulate lenders, intermediaries and other sources of mortgage origination. This was a major extension of the FSA’s role, involving direct supervision of thousands of firms. Rules to promote increased transparency have also been introduced. In the US, only recently has regulation in some states been introduced to license and register intermediaries. The FSA’s Mortgage Code of Business (MCOB) applies to four types of firms: mortgage lenders, mortgage administrators (third party servicers), mortgage arrangers and mortgage advisers. In effect, this means that any consumer application for a mortgage on an owner-occupied property will be covered by FSA mortgage regulation principals. Among the advantages of the new regime is clearer information for consumers about both mortgages and mortgage services through newly introduced 'key facts' documents. Consumers are able to compare mortgages more easily, understand the nature and cost of service they are receiving and have access to the Financial Ombudsman if they have a complaint. Lenders have had to adjust their thinking with respect to lending policies, as the general trend historically has been to lend principally based on suitable security. However, the introduction of regulation and ‘responsible lending’ means that the lender has to focus much more on the borrower’s ability to repay the loan, both now and in the future, and have this backed with guidance - a ‘know your customer’ exercise. Whereas in the past there have been instances of loan approvals purely due to low loan-to-value, irrespective of the borrower’s affordability, this is no longer deemed acceptable under the auspices of MCOB. Regulation has taken affordability one step further requiring the lender’s affordability test to include the interest rate and
While the influence of mortgage product criteria could be a key driver as to future loan performance, so is the way in which loans are originated, underwritten and serviced. The comparison of credit characteristics of underwriting practices between markets in different jurisdictions is not a straightforward process. As distinct markets evolve, they do so in their own unique way and market practices and norms can diverge significantly. In the US, a borrower’s credit quality is estimated by using a credit score. There are three main providers of this credit score, however the one used in the majority of US sub-prime deals is the FICO Score. This score is based on the borrower’s previous credit history, including defaults and late payments on previous secured and unsecured debt. Understandably, for reasons of commercial sensitivity, the exact weightings and methods used in deriving this score are not disclosed. In the UK, when originating a sub-prime loan, a borrower’s credit quality is not generally measured through a score but is instead measured by the nature and number of adverse credit features that the borrower has had. However, the UK market is now seeing a shift towards the credit scoring of sub-prime applicants, with lenders such as GMAC-RFC and Edeus leading the way. . However, the majority of the UK non-conforming market continues to adopt a criteria-based underwriting approach where each product is allowed certain adverse credit features for example a certain amount of CCJs or prior mortgage arrears - before it is ineligible for that product. Whilst methods of assessing credit do differ between the UK and US, there are a number of headline similarities between both markets. The underwriting process in both jurisdictions generally looks at two broad features of a potential loan as a credit proposition: firstly, to assess an applicant’s ability to repay the loan and whether the property is
UK Non-Conforming RMBS: Catching a Cold?: October 2007 13
Structured Finance
payment following expiration of the ‘teaser’ or discount rate. It can be argued that the implementation of MCOB has resulted in a higher percentage of loans being declined. However, it has also been suggested that, in a potentially worrying trend, lenders have loosened their underwriting criteria (for example, allowing higher LTV or DTI) to enable continued levels of gross lending that meet their aggressive targets. This ‘criteria creep’ has yet to bring about significant increases in arrears and repossessions, although even in a benign economy, these were the first signs of worsening performance in 2006. A rising interest rate climate in 2007, as mentioned above, will test these loans further. Limited documentation applications have a higher propensity to default, and these usually have income details omitted or do not have the benefit of a full valuation. Self-certification is becoming increasingly popular in the UK. Borrowers are more vulnerable to default if their income is overstated, irrespective of the jurisdiction, but the implementation of MCOB, and the onus put on lenders and intermediaries with regard to assessing affordability, has perhaps helped to reduce this risk within the UK. In the US, various products, like Adjustable Rate Mortgages (ARMs), affordability products, hybrid ARMs (where the initial rate is fixed and after a set period reverts to an adjustable rate) and interest only loans are on offer, but defaults have occurred as a result of the payment increase potential faced by borrowers at the reset date - a trend that is now becoming apparent in the UK market. Interest only ARMs were traditionally offered to high net worth customers, however the product was expanded to the Alt-A market (a market where applicants typically do not submit all the documentation that would be required to qualify for a prime loan - usually chosen by people who have unsteady sources of income or simply have too little documented income to qualify for a prime loan for the house they want to buy) and sub-prime markets. The expansion of the product down the credit scale resulted in borrowers extending their purchase and/or borrowing capacity by being provisionally accepted at an initially lower interest rate. The resultant increase in rates would see borrowers having difficulty in finding a refinancing option to mitigate the payment shock. In the UK, the majority of loans are placed on two to five year fixed rate products - essentially similar to US hybrid ARM products. Although ‘payment shock’ at the end of a fixed rate period is a concern in the UK, the Key Facts Illustration (KFI) given to
UK Non-Conforming RMBS: Catching a Cold?: October 2007 14
consumers at origination should prepare them for a higher monthly instalment. This is not always the case however, as it would be understandable for a borrower to forget about the reset date which may be five years after the completion of the mortgage. As such, the provision of the KFI is not necessarily a mitigating factor and other tools (such as telephone and letter contact nearer to the time of reset) should be used to remind the borrower of the new, higher payment. Often, affordability tools also use the full standard variable rate with a one or two per cent loading, even if a fixed rate product is taken out. Similarly, interest only loans may be underwritten assuming full capital and interest repayment. This further mitigates the payment shock risk. Affordability assessment in the UK has seen a move towards Debt-to-Income (DTI) measures and away from traditional income multiples, concentrating more on the amount of disposable income required to support the loan. The commitments taken into account are normally taken from the office of national statistics (ONS) - achieving a reasonably consistent representation of other commitments. Within the US, the DTI, supplemented by the FICO score provides a representation of the borrower's creditworthiness. However, each lender has its own strategy, including the level of risk it finds acceptable for a given credit product, and there is no single ‘cut off’ score used by all lenders in the US.
Servicing
The quality and stability of a servicer’s operation has a direct impact on its default management capabilities and, ultimately, on security losses. This is the same for servicers both in the US and the UK. Nevertheless, the sub-prime servicing market in the US differs greatly in terms of business strategy compared to the same market within the UK. A notable proportion of sub-prime originators within the US tend to service their own mortgages, whereas in the UK, many sub-prime lenders outsource primary servicing and administrative activities to third party administrators with the lender retaining responsibility for special servicing. This is an important difference considering the impact of the current global credit crunch. In the US, many originators of sub-prime mortgages have ceased trading in the past six months, and this has inevitably brought about the same fate for their servicing arm. However, the existing portfolios must continue to be serviced and with servicers willing to pay a premium for servicing rights, either the servicing portfolio or the servicing division of a defunct lender may be sold to a third party. The UK market has seen two non-conforming originators cease trading but their loans continue to be serviced by a third-party provider. -
Structured Finance
Mortgage servicers within the US must conduct their operations in accordance with a number of regulations. These regulations include ‘The Federal Trade Commission Act’, ‘The Fair Debt Act’, and ‘The Real Estate Settlement Procedures Act’. The most recent federal regulation, Reg AB, was introduced by the Securities and Exchange Commission (SEC) to formalise and expand rules and regulations for asset-backed securities that are registered with the SEC and sold to the public. Recent years have seen an increase in predatory lending statutes. Predatory lending and servicing is a significant concern within the US residential mortgage markets, particularly sub-prime, and as well as affecting originators, many laws allow for attorneys to include the servicer in any law suit. Many predatory issues are directly within the servicer’s control, and therefore their responsibility including application of payments, assessment of late charges, penalties and fees, debt collection practices and handling of disputes and complaints. Servicers within the UK market are regulated by the Financial Services Authority (FSA), adhering to specific regulatory guidelines and internal audit and compliance reviews. This type of scrutiny ensures the servicer is also able to concentrate on minimising internal risks and collect payments appropriately. Nevertheless, the level of regulation is not as great as seen in the US, where the requirements of a servicer in this regard are stricter and require more focus from senior management. While the regulatory guidance around ‘Treating Customers Fairly’ in the UK may be less transparent, it does ensure that servicers are consistent in their approaches regarding collection and arrears management activities. This consistency requires servicers to document procedures around customer services, and such adherence to these internal guidelines limits the potential for predatory activities so long as the servicer’s guidelines are perceived as ‘fair’ by regulators. As mentioned, in the US some sub-prime originators service their loans as “seller servicers”, so if the originator has financial difficulties or loses its funding lines then the servicer also faces a funding issue. This is primarily due to the fact that revenue for the smaller lenders is derived almost exclusively from origination and servicing fees along with interest earnings on escrow and reserve accounts held by the seller-servicer since few if any mortgage loans are retained on the lender’s balance sheet. Although the originator may exit the market, it is likely the servicing entity would be sold off in its entirety, the servicing portfolio sold or the servicing division spun off into a stand-alone platform. In the UK, primary servicing of sub-prime mortgages is
UK Non-Conforming RMBS: Catching a Cold?: October 2007 15
typically carried out by sub-servicer or third-party administrator albeit some non-conforming lenders have developed their own in-house servicing platforms in recent years. While several of the major UK third party servicers are owned by stable and/or investment grade rated entities such as building societies or investment banks, others are supported by relatively smaller institutions (compared with the US) and could have limited financial support to see them through the potential loss of one or two origination clients. Accordingly, the UK market has recently seen Victoria Mortgage Funding go into administration and London Mortgage Company, a subsidiary of Lehman Brothers being closed down. The third-party administrator for these originators, Vertex Mortgage Services (Vertex), rated ‘RPS3+(prime)UK’ and ‘RPS3+(sub-prime) UK’ has therefore lost two key clients. The potential loss of revenue, along with other factors such as not being part of a rated entity and small client base, resulted in Fitch placing Vertex on Rating Watch Negative. The situation will continue to be monitored by the agency. Even though Vertex will continue to receive servicing fees for the remaining Victoria and LMC portfolios they continue to administer, the loss of these clients given Vertex’s fairly small portfolio hinders the potential for continued growth and revenue long term. Vertex also lacks the support of a financial strong parent company. The financial situation along with the current turmoil in global credit markets impacting UK new origination volumes and the overall affect on Vertex’s operations resulted in Fitch placing Vertex on Rating Watch Negative in September 2007. The situation will continue to be monitored by the agency.
Structured Finance
Collateral Characteristics
Given that at a strategic level there are a number of similarities between the UK and US with respect to underwriting practices in the non-conforming market, it is important to closely examine the types of products originated and the risks associated with them.
Stated Income Loans/Self Certification
The US generally has four types of income verification: Full Documentation, Lite Documentation, Stated Income Documentation, and No documentation. Table 8: US Mortgage Documentation Type
Typical level of verification Must submit one written form of verification to the lender showing 12 months’ stable income. If selfemployed Federal Tax return is acceptable. Lite Documentation 3-6 months bank statements or one pay stub. Stated income Decision to lend is based on the Documentation monthly income stated on the application form. No Documentation No proof and decision to lend is based on credit score and trade lines.
Source: Fitch
certification a valid means of obtaining a loan. In the US, stated income products were also used by a large number of employed applicants who would have had no problem proving their income and who gained a cheaper mortgage if they did so. This feature was abused – sometimes by borrowers but also, by brokers who directed applicants into taking out this product without the applicants’ full understanding. For these borrowers, their loans may have been affordable at the outset, but following a number of interest rate rises both in the US and UK, such loans end up becoming very unaffordable if borrowers are on variable rates. The majority of sub-prime lenders in the UK also lend to employed self-certified applicants and to this extent, the UK mortgage market is also susceptible to this type of mortgage fraud.
Buy-to-Let/Investment Properties
Type Full Documentation
In the UK there is not the same degree of formalised differentiated policies. There is a concept of selfcertification and a concept of full documentation. The degree of underwriting and checking the lenders actually do within these two broad categories varies significantly. Typically for full documentation (for all lender types, prime and non-conforming) three months’ worth of pay-slips or a P60 will be required. However, some lenders, whilst asking for these documents from the borrower, will not actually check all borrowers and will instead check only a risk-based sample. For selfcertification the degree of validation also varies in the UK, but typically as a minimum, a call or letter will be sent to the employer confirming whether the applicant is employed and their position. Furthermore, the proportions of such loans within RMBS transactions are higher in the UK than the US. Self-certification was introduced as a product designed to meet an evolving economy, where being self-employed, doing contract work and/or working with multiple employers was becoming increasingly common. For borrowers with this profile it makes sense that proving income may be difficult or administratively burdensome, making selfUK Non-Conforming RMBS: Catching a Cold?: October 2007 16
One of the main developments in the UK mortgage market over the last decade has been the evolution of buy-to-let lending. From a standing start 10 years ago, but-to-let (BTL) lending now accounts for approximately 10% of new lending, according to CML figures. Given the recent strong performance of the UK BTL segment this has probably been to the benefit of the performance of UK transactions. However, the current rising interest rate environment may have a dual negative impact on the BTL sector. Rising interest rates may stretch highly leveraged tenants’ affordability, meaning existing tenants could find it difficult to manage rental commitments. Secondly, it means mortgage finance becomes more expensive and, to the extent that this increased cost cannot be passed on to the tenant, it means that formerly profitable lets become loss making. For sophisticated BTL investors such circumstances will generally not come as a surprise and will have been factored into the investment decision at the outset. It will not therefore, mean the difference between success and failure. However, the popularity and success of BTL as a product in recent years means that there are many UK lenders who lend BTL business on a sub-prime basis. The media image and impact of TV property makeover programmes has been such that BTL is often portrayed as a one way bet on the property market. Given the levels of house price inflation and benign macro economic situation, it has been relatively easy to make money in the UK BTL market. Therefore losses and defaults have been low. However, in a more strained economic situation it is to be expected that less sophisticated BTL investors and/or BTL investors who have had a history of poor credit performance in the past will be affected first. The equivalent products in the US are residential investment loans. There is evidence to suggest that
Structured Finance
the bulk of US sub-prime transactions do not have significant exposure to BTL though it is likely that BTL are sometimes mis-represented at application to be owner-occupied. Evidence suggests that a sizable portion of early payment defaults in recent vintages may be these types of loans. The more relevant comparison is the US alternative-A (Alt-A) market, where delinquencies among investment properties are much higher relative to all delinquencies. While the exposure of UK transactions to BTL loans does not have a direct parallel in the US, the performance of BTL loans that have been made to small scale unsophisticated borrowers may strain certain UK transactions if the housing market takes a turn for the worse.
Heavy Adverse
Heavy/unlimited adverse products are offered by most sub-prime lenders in the UK (or were before the summer as noted in the Affordability Spotlight above). They are unlimited in that the lender will advance money regardless of the amount of adverse credit features that the borrower has. The lender attempts to price for this risk by limiting LTVs and charging more for the loans. Because loans are scored using FICO in the US, mortgage products are not necessarily marketed as being unlimited or heavy adverse; however, in reality products existed in the US that catered for borrowers with blemished credit records. A product can be heavy or unlimited due to the amount of CCJs, BOs/IVAs or mortgage arrears or how recent they were. Heavy/unlimited adverse borrowers do generally perform worse than borrowers with less impaired credit profiles, with more borrowers exhibiting early arrears and higher default rates. The high level of house price appreciation over the last five years in the UK has meant that even borrowers with significant mortgage arrears were able to mortgage away from those arrears to a new lender. Given the recent re-pricing and re assessment of product ranges, many lenders have opted to withdraw products for heavy/unlimited products. Those lenders that have retained heavy/unlimited products have reduced the LTVs at which they are prepared to lend to. The effect of this is likely to be that heavy/unlimited borrowers will be increasingly retained in mortgage pools, leading to a worsening of performance for UK RMBS.
Second Charges and ‘Piggy-back’ Lending
difference. There are multiple variations of this borrowing strategy and the most established versions are the 80/10/10 and 80/15/5 arrangements. In these situations, a first charge mortgage is applied for which would be 80% of the property value, then there would be a second charge loan for either 10% or 15%, depending on the borrower’s circumstances. The balances would be a cash deposit of either 5% or 10%. In the UK, second charges are almost exclusively made in order to release equity from an existing mortgaged property - usually using this to consolidate unsecured debts - rather than as a means of arbitraging lending criteria. Whilst it is possible that in the UK a borrower’s deposits may be generated from borrowed money, the borrowing would tend to be from family or in the form of an unsecured loan and therefore would not form part of an RMBS transaction. The UK mortgage market does have exposure to second charges and some deals do have exposure to levels of second charges on a par with those seen in the US. These deals are likely to be most exposed from a deterioration in performance. However, the overall exposure to second charges, and also the generally lower LTV that UK lenders underwrite to, should offer some protection to UK RMBS. That said, transactions with a large percentage of second charges are susceptible to higher than expected loss severity that may place strain on the structure, irrespective of the large amount of excess spread generated by such deals. In this circumstance, the role of the servicer will become pivotal in mitigating certain transactions’ exposure to loss. For more information as to how different collateral types – including those listed above – are treated in Fitch’s RMBS criteria, please see either “UK Residential Mortgage Default Model Criteria”, dated 7 February 2007 or “ResiLogic: U.S. Residential Mortgage Loss Model”, dated 14 August 2007. Both reports are available at www.fitchratings.com.
The strategic purposes of taking out a second charge from a borrower’s perspective are different in the US when compared to the UK. In the US, if a borrower has less than the 20% deposit needed to avoid paying a mortgage insurance premium it is possible he or she might, at the same time as applying for the first charge loan, apply for a second charge mortgage (a piggy-back loan) to make up the
UK Non-Conforming RMBS: Catching a Cold?: October 2007 17
Structured Finance
Chart 16: Fitch UK Non-Conforming Index by Programme
3 months plus, including current possessions
Index Clavis Securities Leek Money Partners Residential Mortgage Securities ALBA Eurosail Marble Arch Residential Securities Newgate RMAC Bluestone Farringdon Mortgages plc Preferred Southern Pacific
(%) 30 25 20 15 10 5 0 1 3 5
7
9
11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 Months since issuance
Source: Fitch
RMBS Performance
Despite increasing arrears in the UK, losses generally remain within expectations. This is in contrast to the US where losses have lead to a raft of well publicised downgrades. According to Fitch’s Q2 2007 UK Non-conforming Index, arrears in this sector were averaging at just under 13%, up from 12% a quarter earlier but still lower than the peak of almost 15% seen in Q1 2002. Currently, the 2006 vintage is demonstrating lower arrears than previous years. However, care must be taken when interpreting the different years as the collateral backing each can vary. For example, the transactions comprising the 2006 index to date have a larger share of so called “near prime” loans. As future updates are prepared, it is expected that the
2006 line will rise as more deals containing more risky collateral are included. Going forward, arrears levels are expected to rise. This is particularly true for those deals issued in late 2005 and 2006. Since the loans in these pools were originated there have been five increases in the Bank of England base rate. While borrowers on fixed rates, and to a lesser extent, discounted rates have been protected against these increases to date, they will feel the full force of the revised rates as their teaser rates expire. As most teaser periods run for around two years, the first borrowers to experience all interest rate rises on reset will be doing so around now. Therefore, arrears are now expected to start climbing further.
Chart 17: Fitch UK Non-Conforming Index by Year of Issue
3 months plus, including current possessions
Index (%) 16 14 12 10 8 6 4 2 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 Months since issuance Source: Fitch 2003 2004 2005 2006
UK Non-Conforming RMBS: Catching a Cold?: October 2007 18
Structured Finance
Chart 18: Fitch Index Compared to Bank of England Base Rate and 3M LIBOR
(%)
Fitch index (LHS)
BoE base rate (RHS)
3M LIBOR (RHS)
(%)
15 14 13
7
6
5 12 11 10 Q401 Q302 Q203 Q104 Q404 Months Q305 Q206 Q107 4
3 Q307ª
Source: Fitch, Bank of England
Chart 22 shows the index against both bank of England base rate and LIBOR. As can be seen, this index has historically tracked movements in interest rates with a lag of around two years. As noted in the Affordability Spotlight above, LIBOR poses additional stresses for those borrowers whose mortgages are LIBOR linked rather than BBR linked. Whereas the gap between them has historically been fairly constant, recent events have widened the gap to record levels. As a result, those borrowers who can least afford it, ie those in the subprime sector, are faced with the biggest increases in mortgage payments. Those borrowers with resets in the first half of September 2007 will have felt the full effect of this rate dislocation. In the US, factors that have become prevalent among loans that defaulted within a year of issue, were high combined LTV ratios and the lack of documentation to support stated income. FICO scores have become relatively less significant, although they are still considered a good predictor of risk. Over half of the 2006 vintage loans in the US that experienced early Chart 19: OLTV Distribution Comparison
Vintage OLTVs – US
Fixed 40 FX (%) 85 Hybrid IO 40 ARM Hybrid non IO 228 ARM Piggyback
payment defaults included piggyback or second loans, leading to higher combined LTVs. There was also a high proportion of borrowers who had not provided full documentation to support the income levels stated on their mortgage applications. Both piggyback and low documentation loans are significantly more prevalent among early defaults than among loans that have not defaulted in the first 12 months. In the UK, OLTVs have generally been lower than those seen in the US, with weighted average (WA) OLTVs as shown in chart 19. The data includes a number of deals that can be classified as near prime in that they are made to borrowers with very limited adverse credit history. Consistent with the underwriting criteria for these loans, they tend to have higher LTVs than loans in standard nonconforming transactions. The effect on the vintage LTVs of excluding near prime loans from the analysis can be seen in the chart.
Vintage OLTVs – UK
All (%) 80 78 Non-near prime
80
76 74
75 72 70 2000 Source: Fitch 2001 2002 2003 2004 2005 2006 70 2000 2001 2002 2003 2004 2005 2006
UK Non-Conforming RMBS: Catching a Cold?: October 2007 19
Structured Finance
Chart 20: Self Cert Distribution Comparison by Issuance Vintage
Vintage Self Cert – US
Fixed Piggyback 228 ARM Hybrid IO 40 FX Hybrid Non IO 40 ARM (%) 80 70 60 50 40 30 20 10 0 2000 Source: Fitch 2001 2002 2003 2004 2005 2006 2000 2001 2002 2003 2004 2005 2006
Vintage Self Cert – UK
All Non-Near Prime
(%) 70 60 50 40 30 20 10 0
Chart 19 also shows the LTVs for US loans over the same vintages and it is clear that, while the relatively small fixed rate loan sub-sector has LTVs comparable with UK sub-prime vintages, all other types of US sub-prime loans have higher LTVs than those in the UK. Therefore, any stress on any US vintage from 2000 to 2006 will likely result in a higher loss than in the UK. Self-certification has become increasingly common in UK non-conforming transactions. However, it is interesting to note that, since peaking in 2004, the proportion of pools accounted for by such loans has decreased. Despite this trend, lending to borrowers who have provided no proof of income is more prevalent in UK RMBS pools than in the US (data for the latter being shown in chart 20). Principal payment rates (PPRs), or the rate at which pools amortise, are broadly similar in the UK and the US, although those in the US have tended to build up quicker than has been seen in the UK. Thus, portfolios backing US sub-prime RMBS Chart 21: Prepayment Distribution Comparison
US
2000 (%) 70 60 50 40 30 20 10 0 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 Source: Fitch 2004 2001 2005 2002 2006 2003
transactions have been amortising much quicker than those behind UK deals. However, recent data show this rate to be slowing and Fitch has revised its assumptions downwards accordingly. As a result, credit enhancement has built up more quickly in US deals than in UK deals. The credit enhancement levels for those tranches originally rated ‘BBB’ for US and UK deals are shown in chart 23. The build up in credit enhancement for UK deals is much slower, partly as a result of the lower payment rates discussed above, and also as a result of the structure of many transactions that allow pro-rata pay down of notes to begin once credit enhancement levels have reached a certain level. In the US, such pay down of subordinate tranches occurs at the stepup date – which is comparatively later in time than in the UK where one of the pro-rata conditions is that credit enhancement for senior notes must double in size.
Fitch UK Non-Conforming Index by Year of Issue
(%) 70 60 50 40 30 20 10 0 1 5 9 13 17 21 25 29 Months since issue 33 37 41 Index 2005 2003 2006 2004
UK Non-Conforming RMBS: Catching a Cold?: October 2007 20
Structured Finance
Chart 22: Loss Comparison
US
2000 2003 2006 (%) 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 Months since issue Source: Fitch Source: Fitch 2001 2004 2002 2005
Fitch UK Non-Conforming Loss Index by Issuer
(%) 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 Months since issue Index 2002 2005 2000 2003 2001 2004
Losses in the US have been notably higher than in the UK, as can be seen from chart 22. Fitch anticipates higher losses to materialise in the UK over coming quarters, as borrowers find adjusting to higher interest rates problematic and are unable to find alternative lenders willing to lend to the same criteria as that which applied when they originally took their loans out. Despite the higher build up in subordination seen in the US, the higher level of arrears and losses there has lead to more negative rating actions for US RMBS tranches. In the US, the number of sub-prime tranches upgraded has increased year-on-year. However, these positive rating actions have, on an annual basis, been outnumbered by downgrades. The stalling mortgage market in the UK during 2005 meant that deals de-leveraged at a slower rate than
during years when the housing market grew. Borrowers were less able to refinance to alternative lenders as LTVs had not reduced sufficiently. The maturity and structure of deals also influences the number of upgrades, with many UK sub-prime RMBS deals changing from sequential pay down of notes to pro-rata pay down. Typically this occurs once credit enhancement has reached a certain level usually twice that at closing - and the potential for further upgrades is consequently limited. The Negative Rating Watches in the UK during 2006 and 2007 all related to the two Farringdon Mortgages deals from Rooftop Mortgages Ltd. These deals included substantially higher proportions of borrowers with adverse credit than other transactions in the sector and experienced higher than expected arrears and expenses. Elsewhere in the UK, Fitch is closely monitoring
Chart 23: Credit Enhancement Build-up Comparison
Sub-Prime BBB CE Seasoning Trends (With OC)
(%) 16 12 8 10 4 0 1 7 13 19 25 31 37 43 49 Months since issue 55 61 67 73 5 0 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 Months since issue 2000 2003 2006 2001 2004 2002 2005
UK Sub-Prime BBB CE Seasoning Trends
2000 2003 (%) 25 20 15 2006 2001 2004 2002 2005
Source: Fitch
UK Non-Conforming RMBS: Catching a Cold?: October 2007 21
Structured Finance
deals with heavy or unlimited borrower concentrations or deals where notes are due to be repaid using excess spread. For the latter, Fitch has placed several tranches from Southern Pacific and Preferred Residential transactions on Rating Outlook Negative. On the other hand, 82 tranches are currently assigned Rating Outlook Positive. These are predominately tranches from more seasoned deals, where sequential pay down of notes has allowed a build up in credit enhancement and arrears are in line with expectations.
Surveillance Initiatives
ii. those with above-average exposure to heavy and unlimited adverse products and higher LTV loans; iii. those with exposure to significant un-hedged positions, whether between BBR and Libor or fixedrate loans and Libor. These transactions will be subjected to extra stress tests to assess their ability to withstand sharply deteriorating performance from current levels and greater levels of interest rate dislocation. Results will be contrasted against existing credit enhancement levels and rating outlooks may be revised accordingly as a result. Some non-conforming deals where performance is relatively weak or where there are servicing concerns have already been placed on Negative Outlook as indicated above. Rating downgrades may follow subsequently when a discernible trend can be forecast from performance data.
Fitch is in the process of ongoing review of the rating outlooks for the following subsets of UK nonconforming RMBS transactions: i. those where significant resets are expected over the next 12 months - mostly late 2005 and 2006 vintage deals;
UK Non-Conforming RMBS: Catching a Cold?: October 2007 22
Structured Finance
Appendix 1
Concentration Tables
BTL concentration by Region Postcode Area (PCA) BTL Concentration
Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 PCA E EC WC NW W N SE SW UB HA IG BN M CR TW CT LS EN BS SL OX SO LU RM AL L BH PO FY CV CF MK BB PL NN WD CO NG TQ ME RG SM YO KT TN HU SS IP CB NR LE B GU % BtL Among All 28.35 27.36 26.16 25.22 24.72 23.49 20.96 20.85 19.91 18.28 18.11 16.37 16.20 15.67 15.52 15.22 13.55 13.51 13.45 13.18 13.11 12.77 12.52 12.50 12.43 12.41 12.22 12.19 12.16 12.06 12.02 11.77 11.57 11.47 11.29 11.25 10.95 10.89 10.87 10.84 10.77 10.69 10.54 10.37 10.33 10.21 10.18 10.11 10.08 10.01 9.88 9.80 9.74 Town London London London London London London London London Uxbridge Harrow Ilford Brighton Manchester Croydon Twickenham Canterbury Leeds Enfield Bristol Slough Oxford Southampton Luton Romford St Albans Liverpool Bournemouth Portsmouth Blackpool Coventry Cardiff Milton Keynes Blackburn Plymouth Northampton Watford Colchester Nottingham Torquay Medway Reading Sutton York Kingston Upon Thames Tonbridge Hull Southend-On-Sea Ipswich Cambridge Norwich Leicester Birmingham Guildford
BTL Concentration (Cont.)
Rank 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 PCA CH DA PR RH SN SA WV PE BD EX LN GL NE HP TR SR TS ST BA TF HG BL HX BR OL HD S DE DL CW DN LA WN SK WA LL DH CM TA SG DY WS WF NP WR SP DT CA SY HR TD LD % BtL Among All 9.60 9.58 9.25 9.18 9.16 9.15 9.03 8.99 8.97 8.79 8.69 8.53 8.53 8.32 8.29 8.27 8.26 8.24 8.24 8.23 8.22 8.20 8.16 8.07 7.96 7.85 7.84 7.76 7.69 7.67 7.67 7.63 7.49 7.46 7.45 7.40 7.35 7.05 7.04 7.01 6.91 6.89 6.88 6.46 5.85 5.66 5.55 4.94 4.69 4.57 4.16 2.96 Town Chester Dartford Preston Redhill Swindon Swansea Wolverhampton Peterborough Bradford Exeter Lincoln Gloucester Newcastle Upon Tyne Hemel Hempstead Truro Sunderland Cleveland Stoke-On-Trent Bath Telford Harrogate Bolton Halifax Bromley Oldham Huddersfield Sheffield Derby Darlington Crewe Doncaster Lancaster Wigan Stockport Warrington Llandudno Durham Chelmsford Taunton Stevenage Dudley Walsall Wakefield Newport Worcester Salisbury Dorchester Carlisle Shrewsbury Hereford Galashiels Llandrindod Wells
Source: Fitch, Transaction pool cuts
UK Non-Conforming RMBS: Catching a Cold?: October 2007 22
Structured Finance
Non-Prime concentration by Postcode Area (PCA) Non-Prime Concentration
Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 PCA NP TS WV CF M TD SR OL WS WN BD SA BB B DN L ST FY DH BL DL IG DY S E RM WF LU CH SE PE NN DE HD CA CV LL HX NG PR SK CR LE CW HU WA NE TF PL MK CT SS LS % Non-Prime Among OwnerOccupied 10.92 10.60 10.47 10.45 10.35 10.32 9.99 9.99 9.90 9.86 9.84 9.62 9.61 9.42 9.38 9.37 9.21 8.93 8.89 8.84 8.84 8.72 8.69 8.61 8.61 8.50 8.36 8.33 8.33 8.32 8.24 8.14 7.86 7.85 7.85 7.84 7.84 7.78 7.68 7.66 7.61 7.55 7.53 7.51 7.49 7.39 7.28 7.16 7.13 6.83 6.75 6.65 6.60 Town Newport Cleveland Wolverhampton Cardiff Manchester Galashiels Sunderland Oldham Walsall Wigan Bradford Swansea Blackburn Birmingham Doncaster Liverpool Stoke-On-Trent Blackpool Durham Bolton Darlington Ilford Dudley Sheffield London Romford Wakefield Luton Chester London Peterborough Northampton Derby Huddersfield Carlisle Coventry Llandudno Halifax Nottingham Preston Stockport Croydon Leicester Crewe Hull Warrington Newcastle Upon Tyne Telford Plymouth Milton Keynes Canterbury Southend-On-Sea Leeds
Non-Prime Concentration (Cont.)
Rank 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 PCA UB CO TQ ME LA TA EN WD TR LN BS N PO EX WR DA SY YO BN HR SN HA NR BH NW SL GL TN SM HG IP SO CM LD TW BR SG HP BA OX SW DT CB RH RG AL SP GU KT W EC WC % Non-Prime Among OwnerOccupied 6.60 6.43 6.32 6.23 6.22 6.19 6.17 6.16 6.14 6.11 6.06 6.05 5.97 5.88 5.85 5.80 5.77 5.68 5.51 5.46 5.42 5.38 5.23 5.20 5.19 5.09 5.08 4.95 4.86 4.78 4.75 4.70 4.69 4.61 4.53 4.49 4.47 4.46 4.43 4.39 4.28 4.18 4.15 4.06 4.03 3.99 3.85 3.78 3.65 3.22 3.01 2.19 Town Uxbridge Colchester Torquay Medway Lancaster Taunton Enfield Watford Truro Lincoln Bristol London Portsmouth Exeter Worcester Dartford Shrewsbury York Brighton Hereford Swindon Harrow Norwich Bournemouth London Slough Gloucester Tonbridge Sutton Harrogate Ipswich Southampton Chelmsford Llandrindod Wells Twickenham Bromley Stevenage Hemel Hempstead Bath Oxford London Dorchester Cambridge Redhill Reading St Albans Salisbury Guildford Kingston Upon Thames London London London
Source: Fitch, Transaction pool cuts
UK Non-Conforming RMBS: Catching a Cold?: October 2007 23
Structured Finance
Appendix 2
Detailed Methodological Explanations Dynamic Repossession Order Rate
The dynamic repossession order rate has been calculated by comparing reported repossessions for a period with the number of mortgaged properties during the same period. This rate pertains only to England and excludes Wales, Scotland and Northern Ireland. The figures for the repossession orders made on mortgaged properties are from the Ministry of Justice, to which responsibility was transferred from the Department of Constitutional Affairs in May 2007. Data extend up to H1 2007. The number of mortgaged properties for a particular period has been arrived at as follows: for the years 1985 to 2002, the number of owner occupied properties purchased using a mortgage in England for a particular year, was divided amongst the nine Government Office Regions, in proportion of the estimated number of households per region to the total (for England) for that year. This data is provided by the Department of Communities and Local Government. The same source provides actuals/estimates for the number of owner occupied properties purchased using a mortgage per region for the years 2003 to 2006. Estimates for the first half of 2007 have been arrived at by linear extrapolation.
Regional Risk Factors Analysis
Council of Mortgage Lenders (CML). This figure is in relation to the total lending on secured dwelling as reported by the Bank of England (BoE). For the purpose of this study the average of this ratio between H1 2006 and H1 2007 has been taken (11.37%). The aggregate proportion of non-prime among all owner occupied lending has been taken from Datamonitor, which reports 7.1% for 2006. The regional concentration makes use of the aggregate information in the following way: if the Fs denote the overall proportion of non-prime lending on a national level, and Frd,s is the proportion of region r among all non-prime and Frd,p id the proportion of region r among all prime originations, then the proportion of non-prime among all originations in region r is calculated as:
Cr = Frd , s
Frd , s s d , p ⎛ 100% - F ⎜ + Fr ⎜ Fs ⎝
⎞ ⎟ ⎟ ⎠
It is easy to show that the order of Cr across regions r is independent of Fs.
The aggregate proportion of buy-to-let among all mortgage originations has been calculated using the buy-to-let gross advances as published by the
Copyright © 2007 by Fitch, Inc., Fitch Ratings Ltd. and its subsidiaries. One State Street Plaza, NY, NY 10004. Telephone: 1-800-753-4824, (212) 908-0500. Fax: (212) 480-4435. Reproduction or retransmission in whole or in part is prohibited except by permission. All rights reserved. All of the information contained herein is based on information obtained from issuers, other obligors, underwriters, and other sources which Fitch believes to be reliable. Fitch does not audit or verify the truth or accuracy of any such information. As a result, the information in this report is provided “as is” without any representation or warranty of any kind. A Fitch rating is an opinion as to the creditworthiness of a security. The rating does not address the risk of loss due to risks other than credit risk, unless such risk is specifically mentioned. Fitch is not engaged in the offer or sale of any security. A report providing a Fitch rating is neither a prospectus nor a substitute for the information assembled, verified and presented to investors by the issuer and its agents in connection with the sale of the securities. Ratings may be changed, suspended, or withdrawn at anytime for any reason in the sole discretion of Fitch. Fitch does not provide investment advice of any sort. Ratings are not a recommendation to buy, sell, or hold any security. Ratings do not comment on the adequacy of market price, the suitability of any security for a particular investor, or the tax-exempt nature or taxability of payments made in respect to any security. Fitch receives fees from issuers, insurers, guarantors, other obligors, and underwriters for rating securities. Such fees generally vary from US$1,000 to US$750,000 (or the applicable currency equivalent) per issue. In certain cases, Fitch will rate all or a number of issues issued by a particular issuer, or insured or guaranteed by a particular insurer or guarantor, for a single annual fee. Such fees are expected to vary from US$10,000 to US$1,500,000 (or the applicable currency equivalent). The assignment, publication, or dissemination of a rating by Fitch shall not constitute a consent by Fitch to use its name as an expert in connection with any registration statement filed under the United States securities laws, the Financial Services and Markets Act of 2000 of Great Britain, or the securities laws of any particular jurisdiction. Due to the relative efficiency of electronic publishing and distribution, Fitch research may be available to electronic subscribers up to three days earlier than to print subscribers.
UK Non-Conforming RMBS: Catching a Cold?: October 2007 24