Securitization Aug08

Securitization
North America Asset-Backed Securities
Global Securitization Research
Global Markets Research
Securitization Monthly August 2008

U.S. ABS public/144A issuance (US$bn)
2008 $139.2 YTD $635.7 YTD $634.9 YTD 0 300 600 900
Market Update
Research Team
Karen Weaver
Research Analyst (+1) 212 250-3125 karen.weaver@db.com
Table of Contents U.S. ABS sector breakdown (YTD)

Other 6% Equipment 2%
$863.6
Subprime RMBS 1%
In the Spotlight ..............................................3 Karen Weaver, CFA (1) 212 250 3125 Katie Reeves (1) 212 250 2507 Autos, Credit Cards, Student Loans ..............8 Katie Reeves (1) 212 250 2507 Lily Lau (1) 212 250 5360 Mortgage ABS...............................................15 Arthur Frank, CFA (1) 212 250 6156 Marc Silie (1) 212 250 6405 Subprime RMBS ...........................................17 Katie Reeves (1) 212 250 2507 Ying Shen (1) 212 250 1158 CMBS .............................................................20 Richard Parkus (1) 212 250 6724 Jing An, CFA (1) 212 250 5893
675
CDO 11% Auto 24%
2007
2006
$1,249.9
Credit Card 39%
1,200
1,500
Source: Thomson Financial Securities Data
Student Loan 17%
U.S. fixed-rate spreads to swaps (bp)

800 700 600 500 400 300 200 100 0 8/07 10/07 12/07
3-year Auto 5-year HEL
U.S. floating-rate spreads (bp)

900 600 300
750
---------------------------------------------------European ABS ...............................................39 Ganesh Rajendra, CFA (44) 207 545 2082 Conor OToole (44) 207 545 9652 Ivan Phlson-Mller (44) 207 547 2877 Japanese ABS ...............................................41 Yukio Egawa (81) 3 5156 6163 Michiko Sakai (81) 3 5156 6157 Appendix .......................................................44
140 115 2/08 4/08 6/08 8/08
0 8/07 10/07 12/07

2yr Card (LIBOR) 2yr Student Ln (LIBOR)
80 75 2/08 4/08 6/08 8/08

3yr ARM SEQ
5-year Credit Card 5-year CMBS
Source: Deutsche Bank Credit card and HEL ARMs spreads to 1-month LIBOR; student loan spreads to 3-month LIBOR Note: Includes public and 144A. Spread charts reflect generic secondary triple-A levels for each asset class
Source: Deutsche Bank Three-year auto spreads are for prime benchmark autos
Year-over-year U.S. ABS issuance change

Total Other Equipment Credit Card Student Loan Auto CDO Subprime RMBS $0 bil. $100 bil. (71.7%) (4.5%) (42.3%) (22.9%) (93.1%) (99.0%) $200 bil. $300 bil. 2007 YTD
Source: Thomson Financial Securities Data, Deutsche Bank
(78.1%)
$400 bil. 2008 YTD
$500 bil.
$600 bil.
$700 bil.
Securitization
Deutsche Bank Securities Inc. All prices are those current at the end of the previous trading session unless otherwise indicated. Prices are sourced from local exchanges via Reuters, Bloomberg and other vendors. Data is sourced from Deutsche Bank and subject companies. Deutsche Bank does and seeks to do business with companies covered in its research reports. Thus, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. Independent, third-party research (IR) on certain companies covered by DBSI's research is available to customers of DBSI in the United States at no cost. Customers can access IR at http://gm.db.com or by calling 1-877-208-6300. DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1.
6 August 2008
Asset-Backed Securities Securitization Monthly
Page 2
Deutsche Bank Securities Inc.
6 August 2008
In the Spotlight
Research Contacts
Karen Weaver (1) 212 250 3125 Katie Reeves (1) 212 250 2507
Housing demand is usually strongly correlated to the unemployment ratehousing demand falls when unemployment rises and vice versa. Fluctuations in employment lead to changes in income and when incomes change, the demand for housing changes. However, unlike other markets, where a drop in demand corresponds almost immediately to a drop in prices, housing prices are downwardly sticky, as sellers/homeowners often resist lowering their prices. Therefore even when unemployment has risen and housing demand has slowed, declines in housing prices tend to come gradually and more slowly. Todays housing correction is unique in that, so far, it has had little to do with unemployment. As the data in recent weeks had renewed concern about the employment outlook, its useful to look at past housing downturns and examine the lags between employment and home prices. If unemployment is indeed rising, it is highly unlikely that home prices will stabilize. In fact, in the past, housing markets continued to falter for some time even after employment had stabilized. Figure 1: Housing prices are notoriously sticky in one (nominal) measure, theyd never declined at all until very recently
Existing Single-Family Home Sales ($000) 7000 6000 5000 300 4000 200 3000 2000 1000 1975 1978 1982 1986 1989 1993 1997 2000 2004 100 500
400 OFHEO House Price Index

Page 3
0 2008
NAR Existing Home Sales (rs)

Source: OFHEO, National Association of Realtors
OFHEO House Price Index
6 August 2008
Figure 2: Logically, an inverse correlation exists between housing demand and unemployment
YOY Change in OFHEO Housing Price Index 12 10 Unemployment Rate 8 6 4 2 0 1/75 9/78 5/82 1/86 9/89 5/93 Negative Correlation = -0.65* 1/97 9/00 5/04 1/08 17% 14% 11% 8% 5% 2% -1%
Unemployement Rate
* No lag
Source: OFHEO, BLS
YOY Change in OFHEO Housing Price Index
One of the critical issues facing the economy is how much longer before we see home prices stabilize. Historically, home price corrections have taken many years. For example, we examined the most recent (prior) downturns of four major regional U.S. metropolitan statistical areas (MSAs). Even in nominal terms (i.e. before adjusting for inflation), it took about four to six years for housing prices to bottom (peak-to-trough) and another three to eight years for prices to climb back to their most recent peak. Figure 3: Historically, housing market corrections have taken many years
Home prices* MSA (Home price peak) Los Angeles (3Q 1990) San Francisco (1Q 1990) New York (1Q 1989) Boston (4Q 1988)
Source: OFHEO, Deutsche Bank
Price-Decline Peak to Trough 22% 12% 9% 12%
Peak to Trough Horizon 6yr 5yr 6yr 4yr
Peak to Peak Horizon 10yr 8yr 9yr 9yr
* Nominal. Home price peak is the quarter in which peak occurred.
Of course, unlike the current crisis, these past downturns coincided with regional economic weakness, manifested in employment trends. For these major regional MSAs, we compared historical unemployment trends to home price movements during the last downturn (slightly different time frames for each). Figures 47 plot unemployment at the MSA and statewide level against the home price peaks and trough of the previous housing cycle.1 For the four MSAs, we observe that in prior downturns, housing prices started to fall as unemployment rose and continued to fall well after employment recovered. Housing prices did not trough until eight to 12 quarters (or two to three years) after unemployment peaked (the peak of unemployment is represented by a triangle in Figures 47).
In the cases where unemployment data on the MSA level does not go back far enough, state unemployment is used.
Page 4
6 August 2008
The Los Angeles and San Francisco markets represent two major regional markets whose prior home price peaks were in the early 1990s; both markets followed two similar correction paths. In LA, housing prices reached a peak in 3Q 1990. Two years later regional unemployment peaked at 10%. By 1996, housing prices in Los Angeles had fallen by 22%. Housing prices in this market did not return to its peak level until 2000for a peak-to-peak horizon of 10 years, in nominal terms. (We use nominal terms in this analysis because mortgage debt is nominal.) In the San Francisco market, a housing price peak was reached in 1Q 1990. As unemployment rose, housing prices began to fall. By 1992, unemployment peaked at 7.1%. Three years later, in 1995, housing prices troughed, having declined by 12%. Housing prices did not retrace their old peak again in San Francisco until 1998, eight years later. In New York, housing prices peaked in 1Q 1989. As unemployment for the region started to rise, housing prices began falling. By 1992, unemployment peaked at 8.9%. It took another three years after unemployment peaked for housing prices to reach bottom. By 1998, New York housing prices climbed to its earlier peak, reflecting a nine year peak-to-peak horizon. Boston reflects a slightly different experience as unemployment in this region did not significantly lead declines in house prices. Boston reached a housing price peak in 3Q 1988; four years later, in 1992, housing prices had tumbled by 12%. Unemployment, which had started to rise in the late 1980s, reached its peak of 8.9% in 1992. Figure 4: Los Angeles MSA In the last housing downturn, housing prices didnt bottom until four years after the unemployment rate peaked
12% 10% 8% 6% 4% 2% 0% 1/81 1/85 1/89 1/93 1/97 1/01 12/01 California Statewide Unemployment Rate Los Angeles MSA
Source: OFHEO, BLS Source: OFHEO, BLS
Figure 5: In San Francisco, it took three years for housing prices to bottom after the unemployment rate peaked
12% 10% 8% 6% 4% 2% 0% 1/81 1/85 1/89 1/93 1/97 12/01 1/01 California Statewide Unemployment Rate San Francisco MSA Home price peak Home price trough
Home price peak
Home price trough
Page 5
6 August 2008
Figure 6: In the New York MSA, housing prices bottomed three years after the unemployment rate peaked
12% 10% 8% 6% 4% 2% 0% 1/81 1/85 1/89 1/93 1/97 12/01 1/01 New York Statewide Unemployment Rate New York MSA
Source: OFHEO, BLS
Figure 7: In Boston, the unemployment peak and housing price trough occurred in the same year
12% 10% 8% 6% 4% 2% 0% 1/81 1/85 1/89 1/93 1/97 1/01 12/01 Massachusetts Statewide Unemployment Rate Boston MSA
Source: OFHEO, BLS
Home price peak
Home price trough
Home price peak
Home price trough
Unlike past downturns, unemployment did not precipitate this current home price correction. Now, with unemployment expected to worsen, that could significantly extend the timing (letting alone the magnitude) of the correction. Weve never seen a market where housing prices bottomed while unemployment was still rising. Figures 812 show the unemployment and home price trends for these four MSAs today. Notably, current unemployment levels in all four are below DBs economists forecast (Figure 8). They are also well below their respective unemployment peak in the last downturn, with current unemployment levels ranging from 4.4% to 6.0% in these MSAs, versus unemployment peaks in the past cycles of 7.1% up to 10%. Further deterioration in the employment picture will likely usher in a phase two of the subprime-crisis-induced home price correction, in our view.
Page 6
6 August 2008
Figure 8: Unemployment rate expected to rise into 2009

1Q 08 U.S. Unemployment Rate
Source: Deutsche Bank
2Q 08 5.0%
3Q 08F 5.5%
4Q 08F 5.7%
1Q 09F 6.0%
2Q 09F 6.3%
4.9%
Figure 9: A correction in the Los Angeles market could be some ways off Housing prices peaked in 4Q 06; Unemployment is still 400 bp below its previous peak of 10%
7.5% Home price peak
Figure 10: San Francisco correction might be nearer Housing prices peaked in 3Q 06; Unemployment currently stands 190 bp below its previous peak
7.5%
Home price peak
6.5% 6.0 5.5%
6.5%
5.5% 5.2
4.5%
4.5%
3.5% 1/02 4/03 7/04 10/05 1/07 4/08 5/08
3.5% 1/02 4/03 7/04 10/05 1/07 4/08 5/08
Los Angeles
Source: OFHEO, BLS Source: OFHEO, BLS
San Francisco
Figure 11: A correction in the New York market is not yet in sight Housing prices saw its peak just a few months ago, in 4Q 07; Unemployment is currently 500 bp below its previous peak
7.5% Home price peak
Figure 12: Boston unemployment is 450 bp away from its previous peak in the last downturn
7.0%
Home price peak
6.5%
6.0%
5.5% 4.9 4.5%
5.0% 4.4 4.0%
3.5% 1/02 4/03 7/04 10/05 1/07 4/08 5/08
3.0% 1/02 4/03 7/04 10/05 Boston

Source: OFHEO, BLS
1/07
4/08 5/08
New York MSA

Source: OFHEO, BLS
Page 7
6 August 2008
Autos, Credit Cards, Student Loans

Research Contact
Katie Reeves (1) 212 250 2507 Lily Lau (1) 212 250 5360
Autos
During the month of July, $3 billion of new auto ABS paper came to market amidst a wave of negative headlines on the Big Three automakers, historically weak sales and continued weakness in used car valuations, as well as record level gas prices. Issuance was led by USAA, World Omni and Ford. By month-end Deutsche Banks triple-A prime retail auto ABS spread series had gapped out 15 to 25 bp depending on tenor. Not surprisingly the subprime triple-A spread series widened more significantlyspreads for 2-year fixed-rate paper were wider by 25 bp, while spreads for the 5-year tenor ended the month 100 bp wider. The decision by the Big Three U.S. automakers to significantly pull back from their once profitable leasing programs dominated headlines in July. The steep deterioration in the residual values of full-size SUVs has resulted in significant losses for those U.S. auto finance companies that have a heavy concentration in these vehicles. Chrysler Financial shuttered its lease program on August 1, GMAC announced it will eliminate subsidized leasing in Canada and cut its leasing business in the U.S. by half, and Ford will significantly tighten credit terms and raise prices of certain U.S. SUV and truck leases. For ABS, the reduction in lease programs will affect new issue volume going forward. With that said, the auto lease sector has always been a relatively small segment of the total auto ABS volume. Auto lease ABS deals comprised about 10% of the auto ABS volume in 2006 and 2007. The biggest players in this space (using new issue data from 2006 forwards) have
Prime auto fixed-rate spreads to swaps (bp)

200 170 140 110 80 50 20 -10 8/07 10/07 12/07 2/08 4/08 3yr Prime 6/08 8/08 140 105
Auto public/144A issuance (US$bn)

2008 $32.4 YTD
2007
$42.0 YTD
$72.1
2006
$43.9 YTD
$88.8
2yr Prime
20
40
60
80
100
Auto issuers (YTD)

Wachovia 5.4% Carmax 6.0% World Omni 6.6% Chrysler 8.7% GMAC 14.8% Honda 5.2% Volkswagen 3.6% Ford 32.0%
Subprime auto fixed-rate spreads to swaps (bp)

600 500 400 300 200 100 0 8/07 10/07 12/07 2/08 4/08 6/08 8/08 325 500
USAA 8.8%
Nissan 8.9%
2yr Subprime
3yr Subprime
Source: Thomson Financial Securities Data Source: Deutsche Bank Note: Issuance data includes U.S. dollar denominated deals only and exclude tranches with legal maturities under 1 year. Spread data reflects triple-A secondary offer side levels.
Page 8
6 August 2008
been Chrysler (through its Capital Auto Receivables Trust (Lease) platform) and Nissan. Other issuers include Pinnacle Capital, Susquehanna Auto Lease Trust and Volkswagen. The effects of declining residual values may impact the performance of these transactions going forward, particularly those with heavier concentrations of trucks and full-size SUVs. Figure 13 shows the deals that are currently on negative watch. In its press release, S&P cites the marked decline of used vehicle prices for SUVs and pickup trucks, and the significant exposure2 the affected transactions have to these vehicles. In order to determine the adequacy of the credit enhancement levels, the rating agencies noted that it will compare the base residual values at closing with the current values. Figure 13: High concentrations of SUVs and trucks prompt negative watchlisting for certain auto lease ABS by S&P
Trust CAL Securitization Trust CAL Securitization Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Capital Auto Receivables Asset Trust Nissan Auto Lease Trust Nissan Auto Lease Trust Nissan Auto Lease Trust Nissan Auto Lease Trust Nissan Auto Lease Trust Nissan Auto Lease Trust Nissan Auto Lease Trust Nissan Auto Lease Trust
Source: Standard and Poors
Series 2008-1 2007-1 2007-SN2 2008-SNB 2007-SNE 2007-SN2 2007-SN2 2007-SN2 2007-SN2 2007-SN2 2007-SNG 2008-SNA 2008-A 2007-A 2007-A 2007-A 2008-A 2008-A 2008-A 2008-A
Class A-3 A-3 D A A C B A-4 A-3 A-2 A A A-2a A-4 A-3 A-2 A-2b A-3a A-3b A-4
Orig $ Amt 500 400 77.4 509.11 746.99 83.9 70.9 518 590 470 504.36 543.18 98 365.08 200 375 75 155 70 22.72
Date of Action 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08 25-Jun-08
Current Rating AAA AAA BBB AAA AAA A AA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA
Rating Action On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg On Watch - Neg
Generally speaking, values of used vehicles have declined. The Manheim Used Vehicle Index for the month of June, which stands at 107.8, reflects a 6.2% decline year-over-year. The value of this index has been posting y-o-y declines steadily since September 2007. The Manheim index is based on all completed sales transactions at Manheim auctions and is meant to track the used vehicle market overall. Not all vehicles reflect similar deterioration. While full-size SUVs and trucks have shown large price declines, most fuel-efficient compact cars have retained or increased their residual values. The performance of a given securitized portfolio will vary based on its underlying collateral vehicle mix.
S&P, Ratings On Various Auto Lease ABS Deals With Used SUV And Truck Exposure Put On Watch Negative, June 25, 2008.
Page 9
6 August 2008
Figure 14: Manheim Used Vehicle Index Current level reflects YOY decline of 6.2%
10.00%
5.00%
0.00%
-5.00% -6.18% -10.00%
-15.00% Jan-96
Source: Manheim Consulting
Jul-97
Jan-99
Jul-00
Jan-02
Jul-03
Jan-05
Jul-06
Jan-08
Automotive Lease Guide (ALG), a recognized third-party provider of residual value forecasts, recently announced a significant repositioning of its residual value forecasts to take into account the effect of the recent spike in fuel prices and the diminishing demand for trucks and SUVs. On average, the residual values of compact cars in the following table have risen by 7.5%; the residual values for trucks and full-size SUVs have experienced a decline of 8.5%. (Compact cars are shown in the first set of columns; trucks/SUVs are shown in the second set of columns.) Figure 15: The residual value of compact cars have risen versus trucks and SUVs YOY % Change Model Average Honda Fit Chevrolet Aveo Hyundai Accent Kia Rio Toyota Yaris Suzuki SX4 Nissan Versa
Source: Automotive Lease Guide
2007 46.0 33.5 33.3 32.3 43.7 39.5 43.3
2008 56.0 42.8 41.8 39.2 50.3 45.3 48.4
YOY % Change 10.0 9.3 8.5 7.0 6.6 5.8 5.1
Model Average Ford Expedition Chevrolet Tahoe Chevrolet Suburban GMC Yukon Xl GMC Yukon Chevrolet Avalanche Toyota Sequoia Nissan Armada
2007 44.6 42.2 39.2 40.0 42.7 45.5 44.6 44.9
2008 32.4 31.8 29.7 31.3 34.2 38.5 38.3 39.1
YOY % Change -12.2 -10.4 -9.5 -8.6 -8.5 -7.0 -6.3 -5.8
As a result of declining residual values, S&P placed on negative watch the ratings of two DaimlerChrysler floorplan ABS transactions, DaimlerChrysler Master Owner Trust series 2005-C and 2006-A. The rating action was prompted specifically by the rating agencys revised residual value assumptions for the underlying pools of collateral and not on the current performance of the collateral. As with the lease transactions mentioned above, the precipitous and rapid decline in the residual value of certain vehicles have resulted in a revision of the rating agencys residual value assumptions. S&P has requested that Chrysler Financial increase the amount of credit enhancement in the affected trusts by $60 to $75 million (either through additional collateral or cash) within the next 30 days to maintain the current ratings. Away from residual values, other Chrysler transactions faced ratings pressure in July. In a separate rating action, S&P placed the ratings of 14 classes of DaimlerChrysler auto loan ABS
Page 10 Deutsche Bank Securities Inc.
6 August 2008
issued from two trusts (DaimlerChrysler Auto Trusts series 2007-A and 2008-A) on negative watch, citing higher-than-expected losses and delinquencies. The rating agency has noted that the recovery rates for the two affected trusts are lower than the recovery experience of earlier DaimlerChrysler transactions (see Figure 16). (In the graph, the solid lines represent 2007 and 2008 transactions, while the dotted lines represent pre-2007 transactions. Recovery rates are defined as a percentage of liquidations for the month.) For example, for month 5, the recovery rate for the 2007 transaction is 42.65% and for the 2008 transaction it is 39.61%. In the earlier deals the recovery rates are between 45.37% and 54.77%. Recoveries are the proceeds expected after liquidation of a vehicle, and, together with loss frequencies, dictate overall loss levels. Further pressuring these deals, neither of the impacted trusts has reached its target overcollateralization level. S&P has not yet determined whether additional credit enhancement is necessary for these transactions. Older trusts were also reviewed by the rating agency due to deteriorating pool performance. The ratings on these trusts, however, were affirmed as these trusts contained sufficient credit enhancement levels. Figure 16: Recovery rates of newer DaimlerChrysler transactions have been lower than previous deals
70 Recovery as a % of Liquidation 60 50 40 30 20 10 0 1
Source: Intex
DCAT 2005-A DCAT 2006-A DCAT 2008-A
DCAT 2005-B DCAT 2007-A
11
13
15
17
19
21
23
25
27
29
31
33
35
37
Age by Months
Page 11
6 August 2008
Credit Cards
After posting strong issuance volume in the first half of the year versus 1H 2007, card issuance moderated in July as $4.5 billion of new card ABS entered the market. This represents a 43% decline from the average monthly issuance volume in the first half of the 2008. Julys activity brings the year-to-date volume in sector to $52.1 billion, 6.4% below 2007s level. The new issue volume that was offered in July came from top-tier issuers, and priced at wider spreads than what was seen earlier in the summer. (This is consistent with where spreads have trended in the secondary market as well.) Additionally, slightly more than $2 billion of the volume was issued with a tenor of one- or one-and-a-half years, which is very short for a sector more accustomed to seeing three-to-seven year deals. In July, we saw the first Canadian receivables-backed credit card securitization to be offered in the U.S. market, in US dollars. Issued as a private 144A transaction, the Golden Credit Card Trust (Royal Bank of Canada) series 2008-3 consists of a $400 million triple-A rated senior class and a $23.5 million triple-B rated subordinate class. The emergence of Canadian securitization in the U.S. market was paved by recent changes to the existing Canada-U.S. Tax Treaty. These changes, which became effective 1Q 2008, call for the gradual elimination of withholding tax, over a three-year period, on interest payments to foreign investors. We expect the legislation to spur interest in more cross-border transactions; however, in the near term we expect issuance to be limited to Canadian issuers with some U.S. name recognition. Charge-offs continued to rise for Fitchs prime and subprime credit card indices, as reported in June. For the prime index, charge-offs rose only 3 bp to 6.43%, which represents a much slower pace of increase than the pace seen at beginning of the year. However, this metric now stands 183 bp or 40% above last years level at this time, and we expect charge-offs to continue to climb. Prime late-stage delinquencies, which have been rising since mid-2007 and are a leading indicator of future charge-offs, fell 7 bp for the month to 3.10%. This metric Credit card fixed-rate spreads to swaps (bp)
170 150 130 110 90 70 50 30 10 -10 8/07 10/07 2yr
Credit card public/144A issuance (US$bn)

150 130 115 85 80 2008 $52.1 YTD
2007
$55.7 YTD
$94.1
2006 12/07 3yr 2/08 5yr 4/08 6/08 7yr 8/08 10yr 0
$42.7 YTD
$66.9
25
50
75
100
Credit card issuers (YTD)

Bank of America 22.9% American Express 17.2% Chase 25.5%
Credit card floating-rate spreads to LIBOR (bp)

160 140 120 100 80 60 40 125 110 80
Cabela's 1.0% Discover 6.9% Capital One 9.7%
Citibank 16.9%
20 0 8/07 10/07 12/07 3yr

2/08 5yr
4/08 7yr
6/08
8/08
Source: Thomson Financial Securities Data Note: Spread data reflects triple-A secondary offer side levels.
Page 12
6 August 2008
stands 50 bp above last years level at this time. Subprime charge-offs continued to rise, increasing by 49 bp to 13.01%. This represents a 417 bp, or 47%, increase over last years level at this time. Subprime delinquencies rose 3 bp to 5.86%. Figure 17: Prime charge-offs and delinquency continue to rise, albeit slower for June
8% 7% 6% 5% 4% 3% 2% 1% 0% 5/92 5/94 5/96 5/98 5/00 5/02 Charge-off % Delinquency % 5/04 5/06 1.5% 3.0% Delinquency 4.0%
Figure 18: Subprime delinquencies have stabilized; losses continue to rise

20% 18% 12%
3.5%
Subprim e card charge-off
16% 14% 12% 10% 8% 6% 4% 2%
10%
Prime card charge-off
8% Delinquency
2.5%
6%
2.0%
4%
2%
1.0% 5/08
0% 6/96
0% 6/98 6/00 6/02 6/04 Charge-off % Delinquency % 6/06 6/08
In mid-July, the Financial Accounting Standards Board (FASB) decided to postpone for one year proposed changes to FASB Statement 1403 and Interpretation 46R4; the amendments had been scheduled to be implemented by year-end 2008 and are now slated for January 2010. The controversial amendments would jeopardize the true sale treatment of many securitizations and require financial institutions to consolidate those securitizations back on their balance sheet. Following this, and assuming no change to the current regulatory capital regime, banks would then be required to raise more capital to meet the subsequent increase in regulatory capital requirements. While general concerns that the effect of these changes could prolong market dislocation 5 abound, market participants are particularly concerned about the impact on the credit card businesses. Credit card securitizations, unlike other securitizations have been particularly vulnerable to accounting policy changes, given the latitude that credit card issuers have to replace and/or change terms on the securitized balances.
3 4 5
Accounting for Transfers and Servicing of Financial Assets and Extinguishments of Liabilities. Consolidation of Variable Interest Entities.
Letter to the SEC and the FASB from Ranking Member on the House Financial Services Committee, Spencer Bachus, dated July 22, 2008.
Page 13
6 August 2008
Student Loans
In July, $1.5 billion of student loan ABS came to market; year-to-date issuance stands at $23.8 billion, which is 42% below 2007s level for the same period. The bulk of the months issuance came from Sallie Mae with a $982 million offering at the end of the month. Access Group entered the term ABS market for the first time this year with its $463.6 million Access Student Loan Trust 2008-1. Both of these transactions were backed by FFELP collateral, which benefits from a substantial U.S. government guarantee. The most notable offering in this months market may have been the $118.8 million MRU Student Loan Trust 2008-A, the first private credit student loan ABS transaction so far this year. The 144A transaction was the second offering in the term ABS market from MRU (My Rich Uncle) Holdings. The bonds were offered in five tranches with two 7.9-year triple-A rated tranchesthe $25 million class A1A, which priced with a fixed-rate coupon of 7.4% and the $75.3 million class A1B, which priced with a floating-rate coupon of 3mL + 300 bp. (The comparable tranches in the 2007 MRU transaction were priced via the auction-rate market.) According to S&P, the only rating agency that rated this transaction, the higher cost of funding resulted in higher enhancement levels. The 2008 transaction required an initial subordination level of 14.25% with a target level of 21.50% versus 17.25% for the 2007 transaction.6 This transaction had a relatively long marketing period, and it remains to be seen whether the fact that it got done will serve to help re-open the very quiet market for alternative student loans.
S&P, MRU Student Loan Trust 2008-A, Pre-sale Report, June 10, 2008.
Student loan FFELP spreads to 3-month LIBOR (bp)

175 155 135 115 95 75 55 35 15 -5 8/07 10/07 1yr
Student loan public/144A issuance (US$bn)

150 115 90 80 65 2008 $23.8 YTD
2007
$41.3 YTD
$56.8
2006 12/07 2yr 2/08 3yr 4/08 6/08 5yr 8/08 7yr 0
$37.3 YTD
$66.6
10
20
30
40
50
60
70
SLABS issuance by issuer (YTD)

Nelnet 18.7% Student Loan Corp 15.7% South Carolina Student Loan 2.5% Access 1.9% MRU 0.5% SLM 60.6%
Alternative student loan spreads to 3-month LIBOR (bp)

400 350 300 250 200 150 100 50 0 8/07 10/07 12/07 2/08 4/08 AAA 7-5yr 6/08 8/08 350 300
AAA 2-5yr
Source: Deutsche Bank Source: Deutsche Bank Note: Issuance data includes all public transactions as well as auction rate notes offered as part of a public deal. Spread data reflects triple-A secondary offer side levels.
Page 14
6 August 2008
Mortgage ABS
Research Contacts
Arthur Frank, CFA (1) 212 250 6156 Marc Silie (1) 212 250 6405
MBS spreads widened in late July and early August as sponsorship declined
After steadily tightening from early March until late May, mortgages have been in a widening mode during most of the summer so far, as domestic banks and other real money investors slowed their MBS purchase activities, as did Asian investors. While Fannie and Freddie MBS supply has been declining due to higher guarantee fees for higher LTV and lower FICO loans, pushing borrowers with weak credit scores or small down payments toward FHA financing, GNMA issuance continues to set new records every month. In addition, some investors have become fearful that mortgage credit losses by Fannie Mae and Freddie Mac will eat into their capital sufficiently to make further growth of their retained portfolios impossible without raising additional capital, which is difficult for them with their stock prices depressed. In our view, GSE capital concerns could very well slow the pace of their MBS retained portfolio growth, which has been robust for the last three months. We will know more about their current capital positions after the upcoming releases of their Q2 earnings reports, but we cannot count on the GSEs to continue to grow their MBS portfolios during the second half of 2008 as rapidly as they grew them during the second quarter. Figure 19: Agency Mortgages Historical Performance (as of the Aug 4, 08 close) Trader HR implied Trader HR implied Price Price Change Perf vs Swap Perf vs Tsy 1W 1M 1W 1M 1W 1M 30Y CC -0-03 3 -0-09 1 -0-11 2 -0-15 3 4.5 91-00 4 -0-06 4 -6-05 4 -0-06 3 -0-11 5 -0-18 5 -0-20 3 5 94-15 4 -0-06 4 -5-06 4 -0-05 7 -0-12 4 -0-17 0 -0-18 3 5.5 97-15 0 -0-05 4 -3-29 4 -0-05 1 -0-10 2 -0-14 5 -0-16 0 6 100-04 0 -0-03 0 -2-16 0 -0-03 0 -0-04 3 -0-10 2 -0-09 1 6.5 102-13 0 -0-00 4 -1-12 0 -0-00 6 -0-03 6 -0-06 0 -0-06 6 7 104-14 0 0-00 4 -0-20 4 0-00 1 -0-03 7 -0-03 2 -0-06 3 15Y CC -0-04 0 -0-09 6 -0-10 6 -0-15 7 4 92-17 4 -0-11 4 -5-05 0 -0-11 3 -0-22 3 -0-21 3 -0-29 6 4.5 95-17 4 -0-04 4 -4-16 0 -0-04 3 -0-14 6 -0-13 3 -0-21 4 5 97-30 4 -0-06 0 -3-12 0 -0-05 7 -0-14 0 -0-13 5 -0-19 4 5.5 100-02 4 -0-03 4 -2-07 0 -0-03 6 -0-06 4 -0-10 1 -0-11 6 6 101-29 4 0-00 0 -1-08 4 -0-00 3 -0-06 3 -0-05 3 -0-11 1 6.5 103-14 0 -0-02 0 -1-06 0 -0-02 2 -0-03 4 -0-04 2 -0-05 5
Over the last month, the 30-year FNMA current coupon underperformed its swap hedge ratio by 9 ticks, while underperforming its Treasury hedge ratio by 15.5 ticks, as MBS sponsorship was mostly limited to hedge funds and money managers, who themselves were not consistent buyers but rather opportunistic traders of the MBS market. The 30-year current coupon mortgage now looks inexpensive to both Treasuries and swaps on our historical regression model, taking into account the level of swaption implied volatility, yield curve shape, swap spreads and rate levels. However, we advocate only a neutral weighting at this time, given the high level of market volatility and the doubtful near-term MBS sponsorship by domestic banks and the GSEs. On the other hand, while the 30-year FNMA current coupon spread is currently well inside its March wides (now 242 bp over the 5-10 Treasury blend, versus 293 on March 6, and 155 bp over the 7-year swap rate, versus 190 on March 6), the carry advantage versus similar duration Treasuries is a robust 20 bp per month, too generous to justify an MBS underweight. This recent MBS widening has not been limited to agencies,
Deutsche Bank Securities Inc. Page 15
6 August 2008
as non-agency AAAs, both ARMs and fixed-rate, were in a major widening mode during the first half of July before stabilizing over the past two weeks, and for the most part are now trading below their March lows. For example, super-senior option ARM prices have fallen by about 8 points over the past month, from prices in the low 70s in early July to the low-to-mid 60s now, while prime fixed-rate pass-throughs have fallen about 4.5 points over the same period, from 8 points in early July to 12.5 points now behind comparable FNMA TBAs. While the near-term technicals for the non-agency MBS market remain poor, we would advise against selling at these depressed market levels. Staying close to a market weight in both agency and non-agency MBS is the most prudent policy in the current very volatile and at times illiquid MBS market. While some super-senior Alt-A hybrid ARMs appear to have fallen below fair value, it will be difficult for them to rally much from here until the housing market stabilizes, which we do not think is likely to happen in 2008 (see Update: The Outlook for U.S. Home Prices, a May 30, 2008 publication of Deutsche Banks subprime research team). Figure 20: Non-agency Mortgages Historical Performance (as of the Aug, 4, 08 close)
Description Super-Senior Hybrid ARM Prime Fixed-Rate Pass-Through* Alt-A Fixed-Rate Pass-Through Super-Senior Option ARM Current Coupon FN30 Current Coupon Trust IO * in ticks behind FN30 5.5%
Bond RALI 06-QA1 A21 RAST 05-A6CB A7 RALI 07-QO4 A1 FN30 6.0 FNS 370 IO
Price 67-16 76-16 63-00 100-04 28-30
1-W Chg 1-M Chg 0-00 0 0-00 0 0-00 0 -0-03 0 0-30 1 -4-16 0 -6-16 0 -8-00 0 -0-10 6 1-11 7
JPMMT 07-S3 1A1 <12-16> <0-00 0> <4-16 0>
The latest Mellon survey indicates that large money managers have on average slightly increased their MBS overweight in June, with the median MBS allocation rising slightly to 49.92% from 48.26% in May, while the mean allocation rose to 45.15% in June from 43.34% the prior month, while a neutral allocation in accordance with the Lehman Aggregate Index was only 35.38%. Junes MBS mean and median allocations are the highest since December 2007. With the MBS carry advantage to Treasuries now at about 20 bp per month, we think that most money managers will maintain their MBS overweight versus Treasuries. Conventional fixed-rate agency MBS supply fell by 17.2% in June to $75.15 billion, due largely to higher GSE guarantee fees for high LTV and low FICO loans, which went into effect on June 1. On the other hand, GNMA issuance shot up in June, with 30-year fixed-rate GNMA I and II issuance rising 23.6% in June to $24.48 billion. The VA and especially the FHA are clearly growing their market share in 2008, as other avenues of mortgage finance shut down for borrowers with small down payments and/or low FICO scores. We expect GNMA issuance to continue to grow for the remainder of 2008, and for this growth to accelerate in 2009 as the Hope for Homeowners section of the recently enacted Housing and Economic Recovery Act of 2008 leads to delinquent Alt-A and subprime borrowers refinancing into FHA loans, which will mostly be securitized as non-TBA deliverable GNMAs. We think the prudent position today is a market weight of FNMA and Gold pass-throughs, but a modest underweight of GNMAs, whose supply we expect to continue its recent rapid growth. In pass-throughs, we recommend that investors maintain an up-in-coupon bias in 30-year conventionals, with our favorite coupons being 6.5s and 7s, as falling housing prices and tighter mortgage underwriting make it harder for many mortgagors to refinance. Fannie and Freddie premium prepayments came in below expectations in both May and June, and we expect further slowing in July and August. In premiums, we like high LTV pools, as LTV has become the key variable in determining mortgage refinanciability in an environment where appraisals are coming in lower and primary mortgage insurance premiums are increasing for borrowers with LTVs above 80%. However, we do not have the same enthusiasm for higher coupon GNMAs, as servicer buyouts of seriously delinquent mortgages elevate the prepayment speeds of moderately seasoned GNMA premiums, while easier FHA streamline refinancing keeps the speeds of less seasoned GNMA premiums robust compared to conventionals.
Page 16 Deutsche Bank Securities Inc.
6 August 2008
Subprime RMBS
Research Contacts
Katie Reeves (1) 212 250 2507 Ying Shen (1) 212 250 1158
Secondary market spreads for some categories of subprime cash RMBS headed still wider in July, largely in sympathy with the broader market for financials. Technicals continue to hurt the sector, with significant CDO and SIV liquidation lists bringing no shortage of supply. Cash 5-year triple-A fixed-rate spreads widened dramatically during the month, from 600 to 750 bp. Short triple-A floaters moved the opposite direction, however. For example, 3-year triple-A ARM sequential spreads started the month at 700 bp, hovered at 725 bp for much of the month, before declining to 675 bp by the end of the month. Cash subordinates were unchanged on the month, at discounted price levels. Looking at single-As, for example, depending on vintage, name, and structure, we are seeing cash prices from between 3-9px. While this does not mark a change for the month, it has come down significantly from the levels seen at the end of 2007, when such bonds were trading more in the 15-30px range. The ABX was either down to unchanged over the month. (See chart at bottom, right, for the price levels over the past year for the triple-A ABX classes.) IndyMac Bank, the 9th largest mortgage originator (at the end of 2007, according to Inside Mortgage Finance), was closed by the Office of Thrift Supervision (OTS) on July 11, and is now being operated by the FDIC in conservatorship. While it is not the first mortgage lender to fail in the current cycle, it is one of the largest. In 2007, IndyMac accounted for 3.2% of the mortgage market (also according to Inside Mortgage Finance). However, it was the second-largest Alt-A lender in 2007, after Countrywide, and made up 11% of Alt-A originations. The banks total servicing portfolio at the end of 2007 was $198 billion, making it the 9th largest mortgage servicer. While IndyMac is predominately an Alt-A securitizer, it also issued subprime RMBS from the Indymac Residential Asset Backed Trust (BBG: NIABS) shelf. By the terms of the documents, the conservatorship of IndyMac is a servicer Event of Default. As a result, the trustee has the option of terminating IndyMacs rights and obligations as servicer. Additionally, a 2/3 investor vote can also force the trustee to replace the servicer. The trustee would then appoint a successor servicer, which would need to be a Fannie Mae/Freddie Mac-approved seller/servicer, with a net worth of at least $15 million, who is willing to assume the role. At least in the near term, we expect the FDIC to keep what IndyMac resources it can to service the portfolio, while it seeks to ultimately sell the servicing. A question will be whether the 50 bp servicing fee in the subprime RMBS documents is adequate to attract a successor servicer. In order to change the servicing fee in the documents, the documents would need to be amended, which would also require a 2/3rd vote, a logistically onerous process.
Subprime RMBS fixed and floating-rate spreads (bp)

900 750 600 675
Subprime RMBS public/144A issuance (US$bn)
AAA ABX price levels

110
2008
$2.1 YTD 90 70 $236.8 50 30 8/07 11/07 2/08 5/08 8/08 88.8 67.0 47.5 44.0
2007 300
$197.0 YTD
0 8/07 10/07 12/07 5yr fixed-rate HEL 2/08 4/08 6/08 8/08 3yr floating seq HEL
2006
$301.7 YTD
$566.1
ABX.HE.06-1 0 100 200 300 400 500 600 700 ABX.HE.07-1

ABX.HE.06-2 ABX.HE.07-2
Source: Deutsche Bank Source: Deutsche Bank Note: All charts show data for U.S. dollar denominated deals only and exclude tranches with legal maturities under 1 year.
Page 17
6 August 2008
One of the few details that the FDIC has provided on IndyMac is that it has temporarily halted foreclosures on that part of the portfolio that IndyMac holds (approximately $15 billion of its $200 billion servicing portfolio of loans). Sheila Bair, the Chairwoman of the FDIC, has been one of the strongest proponents of modifying loans where at all possible, as an alternative to foreclosure. In theory, at least, the FDICs control of IndyMac now offers a loss mit lab of sorts for the FDIC, where it can get a sense for the true quantity of loans for which modification may make sense. By looking just at the bank-owned portfolio, the FDIC does not need to contend with securitization-related constraints on loan modifications. Last week President Bush signed the Housing and Economic Recovery Act, the longanticipated legislation that includes the program which would allow certain borrowers to refinance into more affordable FHA-insured loans. (That specific program is referred to as the Hope for Homeowners Act within the broad piece of legislation.) On July 28 we published our most recent review of that programs details, and implications.7 We maintain our view that the program will likely fall short of helping the 400,000 homeowners estimated by the Congressional Budget Office (CBO), given uncertainty about how motivated borrowers and servicers will be to participate. The focus has now shifted to how quickly the FHA will be able to get the new program up and running. The new law puts the effective date for the program as October 1, 2008. However, the program has yet to be created. Officials from the Department of Housing and Urban Development (HUD), which includes the FHA, have publicly doubted being ready by that date. 8 However, chief architects of the program on Capitol Hill, including Representative Barney Frank, are putting pressure on the regulators to move quickly. And Representative Frank has also called on lenders to immediately start voluntarily suspending foreclosures associated with loans that the lender believes could be candidates for the new program. Concern over the programs timing may have been one of the factors behind Freddie Macs move last week to announce changes to some of their servicing guidelines.9 The combined result of the changes amounts to increasing the economic attractiveness of certain loss mitigation alternatives versus foreclosure. Included in the changes: Increasing foreclosure timelines in 21 states, to 300 daysthis will likely have the practical effect of extending actual foreclosures times in those states as well. And as a result, servicers will also find themselves needing to advance for a longer period of time for delinquent loans impacted by this change, Doubling the amount paid for loan modifications, from $400 to $800, Doubling the amount paid for short sales, from $1,100 to $2,200, Doubling the amount paid for repayment plans, from $250 to $550, Prohibiting servicers from charging a fee to borrowers related to a loan modification, Easing requirements for delinquent loans to qualify for a loan modification. For example, Freddie Mac did not previously allow loans that have already been modified, to be modified again; this requirement is now gone. And, Ending compensation for servicers for adhering to their foreclosure time lines.
Servicers for securitizations have always had the duty to choose the loss mitigation alternative that will maximize recoveries for the given loan. These steps will move that
7 Please see Deutsche Banks Research Update: Congress Passes Housing Bill Update to May 22 Article, July 28, 2008. 8 9
HUD: Foreclosure Program Will Not Be Ready by Oct., American Banker, July 28, 2008. See, full bulletin from Freddie Mac on these changes at http://freddiemac.com/sell/guide/bulletins/pdf/bll073108.pdf.
Page 18
6 August 2008
balance closer to favoring work-outs (versus foreclosures), possibly resulting in fewer foreclosures. This is clearly in line with the overall policy climate in Washington to curb foreclosure activity where possible. However, critics are likely to say that many of these changes may only serve to delay loss recognition where losses are inevitable. Data on re-default rates for modified loans is difficult to come by, as most servicers only started modifying loans in size over the past 12-24 months. However, in mid-July, Moodys released its third report in a series on loan modification activity,10 which included some detail on how modified loans are performing. Data for the report was gathered by Moodys by surveying ten servicers, comprising approximately half of the subprime mortgage market. Among the reports highlights was the statistic that 42% of loan modifications performed during H1 2007 were 90+ days delinquent as measured at the end of March 2008. The report notes that the loans captured by that 42% re-default rate were more likely to have been modified by deferring principal or capitalizing arrearages, and were more likely to be seriously delinquent at the time of modification. In late 2007 and early 2008, Moodys posits that servicers began to perform more loan mods that took the form of interest reductions, and that less seriously delinquent loans were considered for modifications. As a result, Moodys believes that the 42% re-default rate could come down, as the performance of more loans from those later cohorts is captured by their surveying efforts.
10
See Moodys Subprime ARM Loan Modification Update, Moodys Investors Service, July 14, 2008.
Page 19
6 August 2008
CMBS
Research Contacts
Richard Parkus (1) 212 250 6724 Jing An, CFA (1) 212 250 5893
Market activity
Heightened market volatility over the past month related to fears about the solvency of Fannie Mae and Freddie Mac, as well as regional banks (i.e. National City, Indy Mac, WAMU), have taken a toll on the CMBS market. Most cash CMBS spreads and CMBX indices widened through July. In CMBX, indices from more recent series widened substantially more than those of CMBX.1. During July, 30% subordination cash AAAs were 40 bp wider, while CMBX AAA indices were 12-28 bp wider. In AJs cash was 125 bp wider, while CMBX indices were 22-101 bp wider. In both AAs and As, cash widened more than the CMBX indices. In the cash market, bid list activity was generally below average; except for a couple of days in each week in July. CMBS was well bid and cash spreads outperformed post July Fourth, but concerns regarding the GSEs and other financial companies drove the spreads wider and activity down. Money managers have been very active in trading cash CMBS. Insurance companies and banks caught up later with a little more focus on yield. Seasoned AAAs have been the most active and activity down the credit curve has been very quiet. For the AAA, AJ, AA and A classes, where most of the trading has been taking place, cash underperformed synthetics in July. In particular, the AAA synthetics (series 4)/cash basis set the record of negative 72 bp on July 31. The synthetics/cash basis among the other three classes, although far from their respective historic lows, changed drastically in July. For instance, the A synthetics/cash basis was positive 164 bp at the end of June, but it narrowed to negative 5 bp due to the vast underperformance of the cash bonds. See Figure 21. The credit curves for CMBX.1 and CMBX.4 are presented in Figures 22 and 23. Interestingly, for CMBX.1, the AAA-AJ and AJ-AA sections of the curve have been flattening since late June, while the AA-A, A-BBB and BBB-BBB- sections have steepened. In the CMBX.4, the AJAA and AA-A sections of the curve have been flattening since late June, while the AAA-AJ, ABBB and BBB-BBB- sections have steepened. Its also interesting that in both series the ABBB section are the steepest. One might have expected that in CMBX.1 the BBB-BBBsection of the credit curve would be the steepest as it is generally viewed as the cuspy-est in terms of loss risk. Also interesting is that in both series the AJ-AA section of the credit curve is the flattest section. In particular, the extra risk premium from moving down the credit curve from AJ to
C M BS S pre ads
C M BX R ate s
Floating-Rate Sector
Fixed Rate Conduit Sector

Spread to Swap Curve (bp) 1-wk 1-mo 3-m o 8 / 1 Change Change Change AAA 3yr AAA 5yr AAA 7yr AAA SS A-M A-J AA A BBB BBBSwaps (5yr) Swaps (10yr) 200 240 255 235 335 525 725 1050 1700 2200 94 72 20 25 30 25 25 25 50 100 200 300 5 3 80 45 45 45 75 115 175 300 400 500 0 1 30 60 60 90 90 150 175 325 400 600 19 11 AAA AA A BBB
CMB X 5
1-mo 8 / 1 Change
CMBX 4
1-mo 8 / 1 Change
CMBX 3
1-mo 8 / 1 Change
CMB X 2
1-mo 8 / 1 Change
CMBX 1
1-mo 8 / 1 Change
Spread to 1-Month LIBOR (bp) 1-wk 1-mo 3-mo 8 / 1 Change Change Change 275 475 700 975 0 25 100 125 95 50 150 125 100 25 100 75
AAA A-J AA A BBB BBBBB
163 496 712 1048 1722 2224 2720
25 85 88 119 239 351 259
163 495 708 1047 1728 2240 2742
26 93 87 126 240 357 291
163 492 710 1004 1728 2243 2869
26 95 91 183 261 380 423
129 386 591 750 1322 1728 2513
11 42 70 145 243 256 138
111 275 389 549 888 1181 --
13 24 42 93 130 271 --
LIBOR (1 mo)
Page 20
6 August 2008
AA is much less than the extra premium from moving from AJ to AA. Since the AJ is a much thicker class than the AA, one might have expected the opposite result. Figure 21: Historical CMBX.4/cash bases for AAA, AJ, AA and A
300 250 200 150 Basis Points 100 50 0 -50 -100 -150 -200 -250 1/29/08
Source: Deutsche Bank and Markit
AAA
AJ
AA
2/29/08
3/29/08
4/29/08
5/29/08
6/29/08
7/29/08
Figure 22: CMBX.1 credit curve

AAA-AJ 500 450 400 350 300 250 200 150 100 50 0 Nov-07 Dec-07 Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08 Jul-08 AJ-AA AA-A A-BBB BBB-BBB-
Page 21
6 August 2008
Figure 23: CMBX.4 credit curve

AAA-AJ 900 800 700 600 500 400 300 200 100 0 Nov-07 AJ-AA AA-A A-BBB BBB-BBB-
Dec-07
Jan-08
Feb-08
Mar-08
Apr-08
May-08
Jun-08
Jul-08
With CDX IG 5Y 10 bp tighter and all five CMBX AAA indices wider in July, CMBX has underperformed CDX by 24-37 bp over the period. CDX IG 5Y had returned briefly to a more normal relationship of being 10-15 bp wider than CMBX.4 AAA in the first half of July, but is now trading 32 bp through. The CDX IG 5Y / CMBX.4 basis has entered negative territory again. See Figures 24 and 25. Figure 24: CMBX AAA indices versus CDX IG 5Y
CMBX.1 AAA 300 275 250 225 200 Basis Points 175 150 125 100 75 50 25 0 12/14/2007
CMBX.2 AAA
CMBX.3 AAA
CMBX.4 AAA
CDX IG 5Y
1/14/2008
2/14/2008
3/14/2008
4/14/2008
5/14/2008
6/14/2008
7/14/2008
Page 22
6 August 2008
Figure 25: The CMBX.4/CDX IG 5Y basis over time

40
CMBX.4 AAA-CDX IG 5Y Basis (bp)
20 0 -20 -40 -60 -80 -100 12/14/2007 1/14/2008 2/14/2008
CDX IG Y5 - CMBX.4 AAA
3/14/2008
4/14/2008
5/14/2008
6/14/2008
7/14/2008
There is one specific concern related to the GSEs that has surfaced in the market recently: Fannie Mae and Freddie Mac are both major buyers of CMBS securities. They typically purchase 30% subordination AAA classesthe A1A classesthat are created specifically for them. A1A classes are structured so as to absorb the principle risk from the multifamily loans in the collateral pool. From 2004 to 2006, approximately $44 billion in A1A multifamily classes were purchased by Fannie Mae and Freddie Mac. There is now some concern that the GSEs (Freddie Mac in particular) could choose to liquidate some of their securities holdings as an alternative to raising additional capital. While there is some risk of this, we think that it is quite unlikely. The GSEs each own hundreds of billions of dollars in mortgage pass-through securities. These markets are among the deepest and most liquid of all capital markets, and the GSEs could sell many billions in these securities quickly and cheaply. The market for CMBS multifamily carve-out classes (A1A classes), on the other hand, is less liquid in comparison to the pass-throughs. In fact, only the GSEs have ever purchased these securities. It would be extremely inefficient for the GSEs to sell CMBS multifamily securities when they could sell mortgage pass-throughs instead. In the property market, two more retail chains are facing financial challenges. Steve & Barrys LLC has just become the latest retailer to file for Chapter 11 bankruptcy protection. The bankruptcy has raised immense media interest, including a lengthy article in the Wall Street Journal discussing the clothing chains operational and accounting practice. We analyzed the retailers impact on CMBS deals and concluded that the impact has been very minimal. We calculated the store impact by multiplying the percentage of a propertys gross leasable area (GLA) occupied by the store by the loans weight in the CMBS deal in which it resides. The largest store impact is 5.5% on the WBCMT 2005-WL5A deal: the Steve & Barrys store occupies 10% GLA in the Brazos Malls that collateralizes 55% of the outstanding deal balance. Thirty-seven other CMBS deals also have exposure to the retail chain, but all of them have less than 1.3% of the deal exposure. Another large retailer, Mervyns, has been fighting to keep its stores open but eventually filed for bankruptcy in July. According to our data, three floating-rate deals have over 5% exposure to Mervyns: GSMC06-FL8, COMM05F11 and GMAC06-C1.
Page 23
6 August 2008
CMBS Delinquency Update: June 2008

Below, we provide separate analyses of collateral performance for the fixed-rate conduit and large loan/floating-rate sectors. The fixed-rate conduit sector currently comprises of 524 conduit deals containing 67,207 loans with an aggregate balance of $753.2 billion.11 The large loan/floating-rate sector consists of all deal modeled in Intex whose deal type is classified as Floating Rate, Single Borrower Single Property or Single Borrower Multiple Properties. There are 124 such deals containing 616 loans with an aggregate balance of $98.2 billion.
Conduit Sector Delinquency Analysis

The aggregate delinquency rate for the fixed-rate conduit sector increased from 36 bp in December 2007 to 46 bp in May 200810 bp over the course of five months, or an average of approximately 2 bp per month. From May to June, the aggregate delinquency rate increased by 3 bp to 49 bp, a similar rate of increase. See Figure 26. As of June, there were 523 delinquent loans out of a population of 67,207 conduit loans. The aggregate delinquency rate is now approximately one-fourth of its historical peak of 1.95%, which occurred in October 2003, and not far off its historical low of 27 bp, which occurred in March 2007. Figure 26: Aggregate delinquency rates for conduit loans (99-08)
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Jan-99
Source: Deutsche Bank and Intex
30-Day
60-Day
90+ day
Total (Rt. Axis)
2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Jan-00
Jan-01
Jan-02
Jan-03
Jan-04
Jan-05
Jan-06
Jan-07
Jan-08
30-day delinquency rates are clearly trending upwards now. This can be seen more clearly in Figure 27. On the positive side, the upward trend in the 60-day delinquency rate, which is a better credit indicator than the 30-day delinquency rate, is clearly quite modest, having risen only 10-15 bp over the past 12 months. It actually decreased in June.
11 The population of conduit CMBS deals we use consists of all deals modeled in Intex whose collateral type is defined as Conduit or Fusion.
Page 24
6 August 2008
Figure 27: Aggregate delinquency rates for conduit loans (06-08)

0.25 30-Day 0.20 60-Day 90+ day Total (Rt. Axis) 0.8 0.7 0.6 0.5 0.4 0.10 0.3 0.2 0.1 0.00 0.0
0.15
0.05
M ay -0 7 Ju l-0 7 Se p07
Ju l-0 6 Se p06
M ay
Figure 28 provides basic statistics for the aggregate 30-day, 60-day, 90+ day delinquency rates, as well as the foreclosure and REO rates. Figure 28: Aggregate delinquency statistics for conduit loans
` Date Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08 Aggregate # Loans 69,644 69,169 68,609 68,611 67,796 67,207 Balance ($ Bil) 762.1 758.9 755.0 759.9 755.5 753.2 30-Day # Loans 139 155 140 123 121 127 % (Bal) 0.090 0.102 0.122 0.108 0.095 0.129 60-Day # Loans 38 52 54 55 50 55 % (Bal) 0.033 0.040 0.040 0.042 0.056 0.043 90+ Day # Loans 138 137 156 165 163 163 % (Bal) 0.126 0.134 0.155 0.162 0.170 0.167 Foreclosure # Loans 40 50 51 55 58 62 % (Bal) 0.027 0.036 0.039 0.052 0.053 0.053 REO # Loans 104 107 112 104 110 116 % (Bal) 0.086 0.100 0.091 0.087 0.094 0.106 Total # Loans 459 501 513 502 502 523 % (Bal) 0.363 0.412 0.446 0.450 0.468 0.498
The deterioration in delinquency rates for seasoned conduit loans (loans of age greater than 24 months) is even less than that for the conduit population as a whole, which is quite unusual, as credit deterioration is typically manifested mainly in seasoned loans. See Figure 29. The 60-day delinquency rate, in particular, shows only modest deterioration. The delinquency rate for seasoned loans, at 60 bp, is extremely close to aggregate delinquency rates by historical standards. Moreover, it is up only 16 bp from its historic low of 44 bp in July 2007. This suggests that younger loans, the 2006 and 2007 vintage loans in particular are probably underperforming in a relative sense.
M ay -0 8
-0 6
ar -0 7
ar -0 6
ov -0 6
ov -0 7
Ja
Ja
Ja
ar -0 8
n0
n0
n0
Page 25
6 August 2008
Figure 29: Aggregate delinquency rates for seasoned loans (age > 24 months)
0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00
7 M ar -0 7 M ay -0 7 Ju l-0 7 Se p07 6 M ar -0 6 M ay -0 6 Ju l-0 6 Se p06 8 M ar -0 8 M ay -0 8 ov -0 6 Ja n0 Ja n0 Ja n0 ov -0 7
1.4 30-Day 60-Day 90+ day Total (Rt. Axis) 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Moving on to individual property sectors, hotel loans continue to perform extraordinarily well, at least for the moment. The number of 30-day delinquent loans increased from five in May to seven in June. One of these newly delinquent loans, however, was relatively large, causing a jump in the 30-day delinquency rate. The 60-day delinquency rate remained stable. The economic downturn is likely to take a significant toll on hotel performance over the next year or two. Moreover, the stress will be exacerbated by rising fuel prices, affecting both drive-in locations via the price of gasoline and fly-in locations via the impact on airlines (much higher prices and reduced capacity). The one positive for hotels is the extremely cheap U.S. dollar which has helped to significantly increase foreign travel to the U.S. The main beneficiary, however, will likely be large fly-in locations on the coasts.
Page 26
6 August 2008
Figure 30: Historical delinquency rates for conduit hotel loans

30-Day 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00
7 M ar -0 7 M ay -0 7 Ju l-0 7 Se p07 6 M ar -0 6 M ay -0 6 Ju l-0 6 Se p06 8 M ar -0 8 M ay -0 8 ov -0 6 Ja n0 Ja n0 Ja n0 ov -0 7
60-Day
90+ Day (Rt. Axis)
Total (Rt. Axis) 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Delinquency rates in the industrial/warehouse sector remain fairly stable over the past six months. The 30-day and 60-day delinquency rates show little sign of deterioration. The same applies for seasoned industrial loans. Figure 31: Historical delinquency rates for conduit industrial/warehouse loans
30-Day 0.6 0.5 0.4 0.3 0.2 0.1 0.0
M ar -0 7 M ay -0 7 Ju l-0 7 Se p07 M ar -0 6 M ay -0 6 Ju l-0 6 Se p06 M ar -0 8 M ay -0 8 ov -0 6 ov -0 7 Ja n06 Ja n07 Ja n08
60-Day
90+ Day (Rt. Axis)
Total (Rt. Axis) 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
While deterioration in the multifamily sector has received much attention in the press, it is currently less significant than commonly believed, at least in our view. The 60-day delinquency rate has been stable since mid 2007, with the exception of the large portfolio of MBS Realty multifamily loans (located in Texas) that rolled through in December. Moreover, the number of 30-day delinquent multifamily loans is little changed over the past 12-18 months.
Page 27
6 August 2008
Multifamily likely has the best prospects for any of the major property sectors, being the beneficiary of the fall-out of the single-family housing sector in the near-term and very positive demographics in the medium to longer term. We expect multifamily to perform relatively well, on the whole, through the economic downturn. Figure 32 lists the states with the five highest multifamily delinquency rates. Not surprisingly, Michigan leads the way with a 4.94% 60+ delinquency rate (balance-weighted). Texas comes in fourth at 3.47%. Figure 32: Multifamily 60+ day delinquency rates by state
60+ Delinquency Rate ($) 4.94 4.80 4.50 3.47 3.20
State Michigan Tennesee Oklahoma Texas Indiana

# of MF Loans # of Delinquent MF Loans 426 247 196 2191 261 18 10 12 73 9
60+ Delinquency Rate (#) 4.23 4.05 6.12 3.33 3.45
Figure 33: Historical delinquency rates for conduit multifamily loans

30-Day 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Ja n0 M 5 ar -0 M 5 ay -0 Ju 5 l-0 Se 5 p0 N 5 ov -0 Ja 5 n0 M 6 ar -0 M 6 ay -0 6 Ju l-0 Se 6 p0 N 6 ov -0 Ja 6 n0 M 7 ar -0 M 7 ay -0 7 Ju l-0 Se 7 p0 N 7 ov -0 Ja 7 n0 M 8 ar -0 M 8 ay -0 8
60-Day
90+ Day (Rt. Axis)
Total (Rt. Axis) 2.5 2.0 1.5 1.0 0.5 0.0
For the office sector, both 30-day and 60-day delinquency rates were up significantly in June over May, albeit still at extraordinarily low absolute levels. However, the actual number of loans that were 30 days and 60 days delinquent were almost exactly the same in both months. The difference is that the delinquent loans in June were significantly larger, on average, than those in May.
Page 28
6 August 2008
Figure 34: Historical delinquency rates for conduit office loans

30-Day 0.25 0.20 0.15 0.6 0.10 0.4 0.05 0.00
Ja n0 M 5 ar -0 M 5 ay -0 Ju 5 l-0 Se 5 p0 N 5 ov -0 Ja 5 n0 M 6 ar -0 M 6 ay -0 6 Ju l-0 Se 6 p0 N 6 ov -0 Ja 6 n0 M 7 ar -0 M 7 ay -0 7 Ju l-0 Se 7 p0 N 7 ov -0 Ja 7 n0 M 8 ar -0 M 8 ay -0 8
60-Day
90+ Day (Rt. Axis)
Total (Rt. Axis) 1.2 1.0 0.8
0.2 0.0
Credit performance deterioration is most clear in the retail sector at this point, although with an aggregate delinquency rate of 31 bp (up only 3 bp from May), it remains excellent. Figure 35: Historical delinquency rates for conduit retail loans
30-Day 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 60-Day 90+ Day (Rt. Axis) Total (Rt. Axis) 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00
M ay -0 6 Ju l-0 6 Se p06
M ay -0 7 Ju l-0 7 Se p07
Figure 36 presents the number of loans that are 60 days or more delinquent for each of the five major property types. The data suggests that only office and retail are beginning to see a significant upward trend in the number of delinquent loans. On the other hand, office and retail also have the lowest delinquency rates at 27 bp and 31 bp, respectively.
M ay -0 8
ar -0 6
ar -0 7
ov -0 6
ov -0 7
Ja n0
Ja n0
Ja n0
ar -0 8
Page 29
6 August 2008
Figure 36: The number of 60-day delinquent loans for each of the five major property types
Hotel 18 16 14 12 10 8 6 4 2 0
-0 5 ay -0 5 Ju l-0 Se 5 p0 N 5 ov -0 Ja 5 n0 M 6 ar -0 M 6 ay -0 6 Ju l-0 Se 6 p0 N 6 ov -0 Ja 6 n0 M 7 ar -0 M 7 ay -0 7 Ju l-0 Se 7 p0 N 7 ov -0 Ja 7 n0 M 8 ar -0 M 8 ay -0 8 n0 ar 5
Indus trial
Office
Retail
Multifam ily (Rt. Axis ) 60 50 40 30 20 10 0
Ja
Figure 37 presents basic delinquency statistics on the core property types. Figure 37: Delinquency data on the five major property types
Aggregate Property Type Hotel Date Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08 Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08 # Loans 3,494 3,463 3,430 3,461 3,416 3,385 4,829 4,795 4,757 4,728 4,656 4,588 16,369 16,238 16,032 15,965 15,684 15,474 11,508 11,425 11,345 11,358 11,236 11,157 22,704 22,604 22,480 22,537 22,359 22,242 Balance ($ Bil) 43.1 42.8 42.6 43.1 42.8 42.7 27.6 27.5 27.3 27.2 26.9 26.8 109.9 109.3 108.3 108.2 107.2 106.5 208.6 207.8 206.9 208.9 208.1 207.7 212.1 211.5 210.8 212.2 211.1 210.6 30-Day # Loans % (Bal) 8 0.108 8 0.086 8 0.081 5 0.068 5 0.046 7 0.140 7 9 10 3 4 9 51 58 58 52 60 49 15 23 15 18 16 16 39 34 28 26 22 25 0.178 0.147 0.178 0.104 0.054 0.157 0.222 0.253 0.383 0.328 0.339 0.338 0.041 0.063 0.039 0.066 0.046 0.076 0.086 0.071 0.079 0.091 0.068 0.106 60-Day # Loans % (Bal) 1 0.014 2 0.033 2 0.015 3 0.031 2 0.031 1 0.005 3 2 1 2 0 0 21 23 18 24 20 26 4 8 10 7 9 10 5 9 17 15 11 12 0.034 0.031 0.004 0.026 0.000 0.000 0.092 0.160 0.109 0.133 0.138 0.126 0.013 0.019 0.026 0.017 0.023 0.042 0.044 0.012 0.038 0.038 0.035 0.029 90+ Day # Loans % (Bal) 3 0.025 4 0.040 5 0.077 3 0.035 3 0.029 5 0.060 4 4 4 3 4 3 81 80 90 92 89 81 18 14 16 19 20 26 15 17 23 29 28 26 0.117 0.085 0.100 0.057 0.078 0.075 0.626 0.617 0.736 0.775 0.811 0.661 0.050 0.042 0.040 0.040 0.049 0.058 0.018 0.060 0.069 0.092 0.090 0.085 Foreclosure # Loans 3 4 4 3 4 4 1 2 3 2 2 21 27 25 26 29 29 3 2 4 6 7 10 5 7 8 9 9 11 % (Bal) 0.018 0.022 0.037 0.033 0.043 0.043 0.000 0.060 0.063 0.107 0.104 0.105 0.121 0.154 0.139 0.191 0.197 0.179 0.004 0.004 0.014 0.025 0.027 0.033 0.015 0.019 0.022 0.020 0.024 0.030 REO # Loans 6 7 7 6 6 6 5 5 5 4 4 3 38 41 45 43 48 55 15 14 15 15 16 17 29 28 28 26 25 25 % (Bal) 0.115 0.118 0.117 0.111 0.111 0.111 0.063 0.062 0.062 0.060 0.061 0.053 0.215 0.219 0.243 0.234 0.260 0.337 0.052 0.047 0.056 0.053 0.057 0.062 0.069 0.070 0.068 0.065 0.065 0.065 Total 0.280 0.299 0.326 0.277 0.260 0.359 0.392 0.384 0.408 0.354 0.297 0.391 1.276 1.403 1.611 1.662 1.745 1.641 0.160 0.175 0.175 0.200 0.201 0.271 0.232 0.232 0.276 0.307 0.282 0.314
Industrial
Multifamily Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08 Office Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08 Jan-08 Feb-08 Mar-08 Apr-08 May-08 Jun-08
Retail
Page 30
6 August 2008
Apart from delinquency rates, loss severity rates on defaulted loans are extremely low by historical standards, and appear to be headed even lower. First, we describe our methodology, which is a little different from other reported loss severity rates. The loss severity rate we report includes the loss to the trust on the following types of events: 1. 2. 3. Foreclosure/liquidation Discounted payoff Note sale
In particular, we capture and include those events in which the loss to the trust is zero. There are a significant number of such zero loss events and neglecting them misrepresents loss severity. In fact, of the 29 loss events in May, seven gave rise to zero losses. Loss severity rates calculated in this way have been stable since mid 2005 and trending down since late 2007. The average loss severity rates for May and June were 12% and 11%, respectively. Figure 38: Average loss severity rates (all property types)
# Liquidated Loans (Rt. Axis) 70 Avg Loss Severity Rate (%) 60 50 40 30 20 10 0 Jan-02
Average Severity (Balance-Weighted) 70 60 # Liquidated Loaans

Page 31
50 40 30 20 10 0 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
In addition to trends in delinquency and loss severity rates, we also look indicators of whether conduit borrowers are experiencing difficulties refinancing their maturing loans. Specifically, we look at the percentage of loans that have not paid off one month after their maturity date, and three months after their maturity date. Were these percentages to begin to rise, we would suspect that borrowers were facing increasing difficulties. See Figure 39. While the data shows that the percentage of loans still outstanding one month after maturity has increased from approximately 5% to 10% over the past six months, there has not yet been any significant increase in the percentage of loans still outstanding three months after maturity. The evidence suggests that the amount of time required to refinance has increased, but is not yet a major problem for older vintage conduit loans that are reaching their maturity dates.
6 August 2008
Figure 39: Percentage of loans that had not refinanced one and three months after their maturity dates
50 45 40 35 % of Loans 30 25 20 15 10 5 0 Jan-01
900 % Loans Outs tanding 1 Month After Maturity Date % Loans Outs tanding 3 Months After Maturity Date # Loans Maturing During the Month (Rt. Axis ) 800 # Loans Maturing During Month 700 600 500 400 300 200 100 0 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08
Large loan sector delinquency analysis

In this section we look briefly at delinquency trends in the loans from the major non-conduit subsectors of CMBS: multi-borrower floating-rate deals and single borrower deals (both fixed and floating-rate). This sector comprises a much smaller number of loans, but they are, on average, much larger than conduit loans. In total, there are 124 deals with 616 loans having an aggregate balance of $98.2 billion. Importantly, they also tend to be secured by transitional properties (e.g. in the process of being renovated or repositioned) as opposed to stabilized, as in the case of fixed-rate conduit loans. Thus, property performance (and credit risk) tends to be more sensitive to broader economic conditions. It should also be noted that, at the moment, the risk in this sector is related more to loans that must be refinanced in the near term, and less to term defaults due to deteriorating property performance (the exception being condo conversion loans). The delinquency rates we provide will not reflect maturity defaults unless the loans become delinquent after passing their maturity dates. However, we examine the recent maturity default record for large floating-rate loans in a later section. Delinquency rates for the non-conduit sector are presented in Figure 40. The total delinquency rate stood at 34 bp in June, 15 bp below that of the conduit sector. There are currently no 30-day or 60-day delinquent loans, but there are three 90+ day delinquent loans, two loans in the process of foreclosure and one loan in REO.
Page 32
6 August 2008
Figure 40: Delinquency rates for the non-conduit sector

30-Day 1.2 1.0 0.8 0.6 0.4 0.2 0.0
l-0 0 l-0 1 l-0 2 l-0 3 l-0 4 l-0 5 n00 n01 n02 n03 n04 n05 n06 l-0 6 n07 l-0 7 Ju Ju Ju Ju Ju Ju Ju Ju n08
60-Day
90+ day
Total
# of Loans (Rt. Axis) 1000 900 800 700 600 500 400 300 200 100 0
Ja
Ja
Ja
Ja
Ja
Ja
Ja
The total delinquency rate for non-conduit loans is up from a low of 10 bp as of July 2007. While delinquency activity has clearly picked up in some sense, it remains extremely low. In fact, it is apparent from Figure 40 that this sector performed extremely well through the recession of 2002-2003, with the total delinquency rate reaching approximately half of that of the conduit sector. Looking a little deeper into the statistics, of the eight loans that have been 60 days delinquent or worse in 2008, five are condo conversion loans and three are hotel loans. One of the condo conversion loans Greenwich Residential, in GCC06FL4has since become current. However, the borrower is negotiating a modification with the special servicer, the results of which have not yet been disclosed. The recent appraised value of the property is $38.9 million, down from an appraised value of $55.2 million in 9/06. The property has total debt of $43.0 million, but only $27.4 million in the trust, so expectations, at this point are for losses on some of the subordinate debt, but not the debt held by the trust.
Ja
Ja
Page 33
6 August 2008
Figure 41: Delinquency data for the non-conduit sector

# of Loans 853 830 853 835 840 857 855 836 827 861 875 861 837 820 800 798 767 734 744 772 720 709 689 716 675 645 654 651 637 627 591 600 577 578 591 605 574 565 570 570 549 525 504 518 515 531 581 591 571 568 566 567 571 558 542 663 651 652 630 647 642 637 636 628 619 616 Total Balance ($MM) 54,123 53,592 53,104 52,840 54,113 56,956 58,742 58,318 57,207 56,935 59,439 60,802 58,655 57,826 57,644 58,824 58,041 58,351 58,548 60,010 57,740 59,300 59,155 62,611 60,690 58,991 60,382 61,333 62,422 61,875 57,941 60,538 58,193 58,658 60,098 61,945 59,054 61,641 64,489 65,726 64,821 62,450 59,754 63,620 67,224 73,638 85,931 86,907 84,402 83,351 81,743 82,372 83,465 90,057 88,390 100,100 98,282 100,434 98,370 100,232 99,421 98,804 100,985 100,704 98,753 98,218 30-Day Delinq (#) 9 4 5 3 5 4 3 2 3 3 2 1 6 3 2 2 2 3 3 1 6 12 6 5 6 9 6 5 4 5 9 10 6 7 7 1 1 1 1 30-Day Delinq ($MM) 83.9 87.2 51.6 196.8 38.3 81.2 29.5 21.3 0.0 104.0 0.0 27.6 12.3 32.5 18.4 31.0 8.1 8.1 0.0 24.6 137.8 113.7 8.1 98.1 222.7 107.5 46.8 87.2 300.8 38.4 136.9 25.5 22.6 151.3 88.6 38.0 88.7 76.4 7.5 0.0 6.2 21.7 0.0 4.9 0.0 0.0 0.0 4.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 62.8 74.0 0.0 0.0 45.3 144.1 20.3 0.0 0.0 0.0 60-Day Delinq (#) 2 2 3 2 3 3 4 2 1 2 2 1 2 2 1 1 3 1 60-Day Delinq ($MM) 77.6 77.6 97.6 55.5 204.3 58.1 54.8 17.4 12.1 18.5 13.5 12.1 27.4 16.7 15.3 35.0 56.8 0.0 6.8 0.0 0.0 38.8 0.0 32.5 0.0 22.7 0.0 0.0 0.0 0.0 11.0 0.0 11.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.2 12.5 0.0 4.9 0.0 0.0 12.5 4.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 61.3 74.0 0.0 0.0 45.3 144.1 0.0 0.0 0.0 90+ Day Delinq (#) 5 4 4 4 2 4 3 5 6 2 2 1 2 1 1 2 90+ Day Delinq ($MM) 106.4 111.9 120.3 87.6 43.2 139.3 137.5 132.5 147.2 114.2 0.0 46.9 39.0 23.0 0.0 11.7 35.0 46.4 0.0 0.0 11.0 0.0 38.5 38.5 35.0 35.0 19.4 19.4 0.0 165.3 0.0 0.0 0.0 11.0 0.0 0.0 0.0 0.0 22.5 18.9 18.9 18.9 7.9 7.9 7.9 4.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 61.3 135.3 135.3 74.0 74.0 218.1 90.1 218.1 FC ($MM) 16.2 15.2 15.1 30.3 15.1 15.0 15.0 15.0 14.9 14.9 14.9 14.8 5.8 30.6 5.8 21.1 19.4 19.4 15.5 0.0 0.0 0.0 16.3 16.3 0.0 0.0 35.0 35.0 35.0 35.0 35.0 35.0 0.0 0.0 0.0 0.0 0.0 0.0 32.2 32.2 32.2 32.2 37.1 11.2 11.2 11.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 61.3 61.3 61.3 85.8 85.8 REO (#) 12 8 10 13 13 13 14 14 15 17 16 17 16 16 16 17 15 16 15 15 13 14 15 14 14 14 15 13 12 13 13 11 12 12 12 8 7 6 6 6 6 5 6 7 6 5 6 5 5 4 4 4 4 4 4 4 3 2 2 2 2 2 2 2 1 1 REO ($MM) 200.6 136.8 162.9 220.2 194.1 217.1 259.0 257.8 259.5 277.6 275.0 283.6 277.6 277.2 276.8 282.9 244.8 332.0 309.1 308.3 292.9 303.0 389.7 353.6 367.6 366.4 367.2 347.7 326.7 327.7 324.9 318.6 353.3 352.4 352.0 239.5 231.3 215.8 215.6 215.3 213.8 210.3 216.3 236.1 154.5 142.7 153.7 142.3 142.1 121.0 120.8 120.6 120.4 120.2 106.2 106.0 99.6 80.0 79.8 79.7 79.5 79.4 79.2 79.1 35.0 35.0 Total Total (#) ($MM) 31 484.7 21 428.7 25 447.4 26 590.4 26 495.0 27 510.7 27 495.9 26 444.1 25 433.8 27 529.2 21 303.3 26 385.0 23 362.1 24 380.0 25 316.3 25 381.7 24 364.2 23 406.0 18 331.4 17 332.9 17 441.7 19 455.5 19 452.6 24 539.0 27 625.3 23 531.6 22 468.4 21 489.4 22 662.6 22 566.4 20 507.9 16 379.0 18 386.9 22 514.7 22 440.6 14 277.5 14 319.9 13 292.2 12 277.7 10 266.4 11 271.1 11 289.3 11 273.7 10 260.0 9 178.5 7 158.7 6 153.7 7 159.2 6 146.5 4 121.0 4 120.8 4 120.6 4 120.4 4 120.2 4 106.2 4 106.0 4 162.4 4 215.3 4 215.1 4 215.0 5 260.1 7 404.1 7 378.9 6 358.4 5 210.9 6 338.9
Date 1/1/03 2/1/03 3/1/03 4/1/03 5/1/03 6/1/03 7/1/03 8/1/03 9/1/03 10/1/03 11/1/03 12/1/03 1/1/04 2/1/04 3/1/04 4/1/04 5/1/04 6/1/04 7/1/04 8/1/04 9/1/04 10/1/04 11/1/04 12/1/04 1/1/05 2/1/05 3/1/05 4/1/05 5/1/05 6/1/05 7/1/05 8/1/05 9/1/05 10/1/05 11/1/05 12/1/05 1/1/06 2/1/06 3/1/06 4/1/06 5/1/06 6/1/06 7/1/06 8/1/06 9/1/06 10/1/06 11/1/06 12/1/06 1/1/07 2/1/07 3/1/07 4/1/07 5/1/07 6/1/07 7/1/07 8/1/07 9/1/07 10/1/07 11/1/07 12/1/07 1/1/08 2/1/08 3/1/08 4/1/08 5/1/08 6/1/08
FC (#) 3 3 3 4 3 3 3 3 3 3 3 3 2 3 2 3 3 3 2
1 2 2 1 1 1 1 2
2 1 2
1 1
1 1
1 1 1 1 1 1
1 1 1
3 2 2 2 1 1 1 1
2 2 2 2 3 1 1 1
1 1
1 1
1 1
1 2 1
1 2
1 2 2 1 1 3 2 3
1 1 1 2 2
Page 34
6 August 2008
Maturity default experience of 2008 maturing floating-rate loans

We have repeatedly expressed the view that the main near-term risk facing CMBS is not a spike in term defaults as a result of declining CRE prices or deteriorating collateral performance, but rather a spike in maturity defaults, mainly in (often highly levered) large floating-rate loans.12 We have already seen that fixed-rate collateral performance continues to be outstanding and shows little sign of deterioration. However, serious cracks are already appearing in the ability of large floating-rate loans to secure refinancing. Fortunately, only a very small number of large floating-rate loans mature in 2008 that do not have extension options. However, this does not diminish the potential scope of the problem. We also note that there are two important exceptions to our view of solid near-term collateral performanceconstruction loans and condo development loans. Both are currently experiencing rapid deterioration. Fortunately, construction loans rarely, if ever, make their way into fixed-rate conduit deals, although they can be found (in small numbers) in some CRE CDOs. Condo conversion loans are also uncommon in fixed-rate conduit deals, as they are not generally stabilized properties. Once again, however, they do appear in floating-rate deals and CRE CDOs. In our February 12, 2008 publication, we identified nineteen floating-rate CMBS loans (aggregate balance of $1.68 billion) that have final maturity dates (i.e. no additional extension options) in 2008. Two of the loans were noted as having defaulted at that time. As five months have now passed, we return to examine how these loans have performed to date. The results are presented in Figure 42. A summary of our findings are as follows: Eight loans have been fully paid off. Among these loans, two were paid off post final maturity dates (two months late and three months late, respectively), two were paid off just in time and the remaining four loans were paid off prior to the final maturity date. Two additional loans have defaulted since our February note, bringing the total number of defaults loans to four. Two of these defaults were maturity defaults and two were term defaults. Three of the four defaulted loans were condo conversion loans. Both term defaults were caused by insufficient condo sales due to the challenging housing market. In fact, of the four condo conversion loans listed in the table below, three have already defaulted. The fourth default was on a multifamily property located in Las Vegas. Among the four defaulted loans, none is expected to experience a loss on the first mortgage pooled in the CMBS deals. The remaining seven loans are current, with one loan having been extended by the special servicer. Two of the seven loans have high risk of default according to our analysis. The Jamestown loan was granted a three-month extension by the special servicer. It had reported a reasonable DSCR, but apparently failed to refinance before the final maturity date. At least four loans have incurred late principal payments, loan extension or loan forbearance, which reflects the flexibility and willingness for the special servicers to work out the loans to maximize the bond holders best interest.
12 A maturity default refers to a default that occurs at maturity, and is usually the result of the borrower being unable to refinance a loan (or willing at the required terms). A term default, by contrast, refers to a default that occurs prior to maturity. Most term defaults are the result of deterioration in the propertys performance or some other event that affects the propertys value. Typical loss severity rates are markedly lower for maturity defaults than for term defaults.
Page 35
6 August 2008
Given the current challenges in financing commercial real estate transactions amid softening but still robust commercial real estate fundamentals, we believe that the majority of the floating-rate loans will fully exercise all extension options and some might require additional time to secure sales or refinance. Therefore, when pricing investmentgrade floating-rate CMBS bonds, we recommend extending the maturity dates further (three to six months more) for the loans with high maturity default risks to take potential loan modifications, extensions and forbearance into account.
COMM 2005-FL11 Large Loan Surveillance

We periodically collect loan level information and review loan performance for the large loans pooled in Deutsche Bank issued floating-rate deals. The latest surveillance report on the COMM 2005-FL11 deal will be published later this week. The following are the preliminary findings: Since issuance, ten loans have been paid off and one loan has had partial releases, reducing the outstanding deal balance from $2,770 million to $1,155 million, 42% of its original level. The remaining loans pay interest only with floating interest rates indexed to one-month LIBOR with spreads between 0.69% and 2.15%. All pooled whole loans, A notes, or senior participations of whole loans in the trust are rated investment grade by Moodys, Fitch and S&P. The pools current DSCR is 2.25x and LTV is 49.3% compared to the DSCR of 3.23x and the LTV of 45.1% as of cut-off date. All loans have current DSCRs between 1.55x and 2.92x and current LTVs between 11.0% and 65.0% compared to the DSCR ranged between 1.99x and 5.86x and LTVs ranged between 23.0% and 65.0% as of cut-off date. Fitch has affirmed the ratings for this deal on 24th January 2008. As of July 2008, the Whitehall/Starwood Golf Portfolio loan and the Universal City Hilton loan are on the servicers watchlist due to Loan Maturity. (Both loans have extension options.) The Crossgates Commons loan is on the servicers watchlist due to declining and below 1.0x DSCR. All three loans are current. Please see Loan Details in the Surveillance Report for further information. The DDR/Macquarie Mervyns Portfolio had a DSCR of 2.06x as of June 2007. However, with the widely known financial challenges faced by the Mervyns company, we will pay close attention to the retailers strategy to secure short-term financing and keep the door open for business.
Page 36
Deutsche Bank Securities Inc. Page 37
6 August 2008
Figure 42: Floating-rate CMBS Loans Maturing in 2008 (July Update)

Deal BALL03B2 COM07F14 GCC06FL4 GMAC00F1 LBFR03C2 LBFR06C2 LBFR06C5 MSC06XLF Loan Name Westland Shopping Center Macklowe EOP Manhattan Portfolio The Tides The Key Bank Building One IBM Plaza Mandalay on the Hudson 5670 Wilshire Blvd Waikoloa Land Maturity 4/8/2008 2/8/2008 2/8/2008 2/8/2008 3/8/2008 12/8/2008 5/8/2008 7/8/2008 City Westland New York Miami Beach Anchorage Chicago Jersey City Los Angeles Honolulu State MI NY FL AK IL NJ CA HI Current Trust Balance 50,000,000 1,130,000,000 13,047,002 2,901,676 130,211,771 8,096,211 50,538,690 7,030,000 Current Total Balance Property Type NRA/ Unit 228,711 n/a 45 62,281 1,359,036 269 409,325 13,797 Trust Debt Per Unit $219 n/a $289,933 $47 $96 $30,097 $123 $510 Total Debt Per Unit $278 n/a $482,503 $70 $96 $30,097 $143 n/a Status Paid off Paid Paid Paid Paid Paid Paid Paid off off off off off off off Paid-off date/Comments Feb-08 Apr-08 May-08 Sep-07 Mar-08 May-08 May-08 Mar-08 Being special serviced since March. The property is only 31% occupied with 0.51 DSCR so the DSCR is too low to grant a threemonth extension. The current three-month forebearance expires in July 08 and the borrower is trying to sell the property. Loan loss is not expected as the proceeds should be able to cover outstanding balance.(Current NCF capped at 8% will value the property at $9.3MM.) Transferred to special servicer in April. Failed condo conversion. Appraised $127MM as of April 08. Loss on first mortgage is unlikely. Property will likely be sold before a foreclosure is completed because the borrower has filed for bankruptcy.
63,500,000 RT 1,600,000,000 OF 21,712,650 Condo N/A OF 130,211,771 OF 8,096,211 MF 58,479,854 OF 7,030,000 Land
LBFR05C4
321-329 Riverside Avenue Office Senior Companion Note
4/8/2008
Westport
CT
8,400,000
14,080,000 OF Condo Conversion 128,100,000 (Hotel) Condo Conversion 16,066,498 (MF)
49,690
$169
$283
Maturity default
CSF05CN1
Hotel 71
4/7/2008
Chicago
IL
61,281,847
n/a
n/a
n/a
Maturity default
LBFR06C2
Avalon at Seven Hills
12/8/2008
Las Vegas
NV
13888724]
320
$50,208
Defaulted in Nov.07 and being special serviced. 40% condos were sold and the rental portion is 60% leased. Currently in foreclosure.Jan.08 value estimiate ($20.9MM) is above the remaining $50,208 Term default debt balance so no loss is projected. Special servicer is in the process of filing foreclosure documents. 19 units are sold, representing 9% potential gross sales proceeds. Property is currently 90% occupied. The special servicer is selling the collateral as a whole. Value is estimated between $18-25MM. A loss n/a Term default is unlikely. $57 Current Low possibility of default. 1Q08 NOI 3.18. Current as of June 08. No major rollover. Low possibility of default. Sole tenant has vacated in March. 100% vacant.P&S was executed and scheduled closing is July 17, 2008. Low possibility of default. Loan Current. DSCR as of 1Q08 is 1.85x and occupancy is 88%. High possibility of default. The borrower has been funding the shortfalls. The loan is current as of June 08. High possibility of default. 41% occpupied as of Dec 07. DSCR=1.01 as of '07. Low possibility of default. Paid through June 08 and current. DSCR as of Dec. 07 is 1.33x. The loan has low risk of default given the desirable location and leverage. Low possibility of default due to the DSCR of 1.48 as of Sep.07. The loan is being special serviced and extended to Sep.07.
LBFR06C2 BSC04BB3
Village Oaks Riverside Center
1/1/2009 11/8/2008
Tampa Utica
FL NY
17,232,764 28,238,000
Condo Conversion 17,232,764 (MF) 35,827,413 RT
234 633,503
$73,644 $45
CTG04FL1 JPC04FL1
Hensley Distribution Center Oasis Apartments
1/1/2009 8/8/2008
Tempe Las Vegas
AZ NV
3,132,849 2,286,250
4,125,000 IN 3,835,000 MF Condo Conversion 17,247,626 (MF) 31,000,000 OF
124,585 128
$25 $17,861
$33 $29,961
Current Current
LBFR06C2 WBC06W07
The Crossings at Otay Ranch 10/8/2008 Leestown Square 12/8/2008
San Diego Louisville
CA KY
17,247,626 19,500,000
n/a 441,649
n/a $44
n/a $70
Current Current
WBC07W08 CTG04FL1
717 Fifth Avenue Jamestown Mall
9/8/2008 9/8/2008
New York Florissant
NY MO
130,000,000 3,567,648
130,000,000 Mixed Use 4,709,221 RT
117,251 308,441
$1,109 $17
$1,109 $23
Current Current
Source: Deutsche Bank, Intex, Trepp, RealPoint
6 August 2008
Moodys/REAL CPPI showed further price decline in May

Moodys has just released the May 2008 CPPI data. The May Price Index is measured at 174.97, a decline of 3.5% from April 2008. The latest price decline in May is the largest onemonth price decline since 2001. Cumulatively, the CPPI has dropped 8.8% from its peak level in October 2007 and 5.7% from its level a year ago. In addition, the holding period between repeat sales has extended over the last two years. Slightly longer holding periods mean more accumulated price appreciation, which may make owners of those assets more motivated to sell. In particular, the average holding period has extended to just over seven years from just over five years in July 2006. Average transaction price is falling steadily as transaction activity continues to shift to lower priced assets. According to Moodys, almost 75% of all transactions taking place in May priced less than $7.5 million, as compared to 50% in May 2007, and slightly over 50% in 2007 as a whole. In addition, slightly more than 1% of transactions taking place in May 2008 were in assets priced over $50 million, as compared to over 5% of all transactions taking place in May 2007 and 7% in 2007 as a whole. Figure 43: Moodys/REAL CPPI May Update
Monthly Returns (National All Properties) 5.00 4.00 Monthly Price Index Change (%) 3.00 2.00 1.00 0.00 -1.00 -2.00 -3.00 -4.00 2001
Source: Deutsche Bank and Moodys
Index Value (Right Axis) 200 190 180 170 160 150 140 130 120 110 100 90 Index Value
2002
2003
2004
2005
2006
2007
2008
Page 38
6 August 2008
European ABS
Research Contacts
Ganesh Rajendra (44) 207 545 2082 Conor OToole (44) 207 545 9652 Ivan Phlson-Mller (44) 207 547 2877
The past month was one of the most volatile for financial credit and (especially) stocks in recent memory, being witness to the biggest ever one-day decline as well as gain in two separate trading sessions for global bank stocks. Performance earlier in the month was dragged weaker by the GSE turmoil (which ultimately required government intervention), but a better-than-expected bank reporting season aided a perceptible recovery later in the month. Financial credit also traded tighter by the end of the month, but the outperformance was more muted compared to equities, which we see as warranted given the negative rating actions in the major banking sectors. The extreme volatility seen in the financial markets over the past month (coupled with the holiday season) kept trading flows very light in the European asset-backed market. Spread performance continued to generally mirror trends in financial credit. Senior AAA vanilla cash spreads ended the month at around 120 bp, wider by some 5 to 10 bp since the beginning of July. (Single-name senior CDS is generally trading on top of cash bonds in the current market). Conduit CMBS sold-off most noticeably (underperformance of which is justified by the sectors negative rating frequency recently), widening around 20 bp across capital structures over the month. Thus far in the credit crisis, technical pressures in the European structured finance market has remained far less powerful than in the US asset-backed market, particularly of course in the credit impaired sectors such as subprime mortgages (and related CDOs) which continue to witness frequent portfolio liquidations. (In July, the key ABX indices fell to record lows). Forced selling remains less prevalent in the European market, as weve discussed previously, given firstly that banks have more readily underwritten ABS held in conduits, SIVs or money market funds and, secondly, that disenfranchised leveraged investors such as CDO managers and credit hedge funds were much smaller participants in European securitized product in the preceding market. Still, the technical balance in the European asset-backed market remains one of net sellers, as evidenced by the frequency of bid-lists circulating in recent weeks (though we would note, many of such lists do not trade). We continue to see risks that banks will become better sellers in the coming months as balance sheet pressures potentially increase, which in turn could act as a renewed technical drag on spread performance later this year.
Excerpt from our European Securitisation Monthly August 2008.
Monthly ABS issuance (253.5 bn)

100 90 80 70 60 50 40 30 20 10 Jan Mar May Jul Sep Nov EUR bn YTD 2008 2007 2006 2005 2004
Collateral breakdown (2008ytd)

Credit Card Other ABS Auto 0.5% 7.4% 3.1% CMBS 1.6% CDO 12.2%
Rating trends
Up /downgrades 280 240 200 160 120 80 40 25% 50% 75% CDO & Other Non-Consumer ABS Consumer ABS/RMBS CDO & Other Non-Cons.ABS Drift (rhs) Consumer ABS/RMBS Drift (rhs) 100%
-40
0% Aug-07 Oct-07 Dec-07 Feb-08 Apr-08 Jun-08 Aug-08
RMBS 75.2%
(2008YTD)
-80 -120 -160 -200 -240 -280 -75% -50% -25%
Source: Deutsche Bank, rating agencies
Page 39
6 August 2008
Recommendations
Current secondary market prices are heavily dispersed across comparable sectors and bonds, with this tiering reflecting perceived credit and liquidity qualities and also the buyer/seller logic unique to each trade. We generally continue recommending defensive asset types in the more off-the-run sectors, where the pricing dislocation tends to be the most pronounced but where there is little, if any, credit deficiencies to show for it. We also see value in slow-pay senior bonds of seasoned collateral in credit disaffected sectors such as Spanish mortgages or SME CLOs and UK non-conforming RMBS, where we believe the potential credit convexity (that is, senior cash flow acceleration owing to continue credit deterioration) is currently undervalued. By the same token, we remain cautious on the fastpay senior tranches given the scope for extension. We also take a more wary view of European auto ABS for nowwhich trades at the tightest spread among any senior sector (AAA levels of around 90 bp)in light of potential contagion risks from the US auto ABS sector, which has sold-off noticeably in recent weeks following the retrenchments by the major auto makers. While there is unlikely to be any fundamental justification for European auto bonds to trade in sympathy at this stage, we note nonetheless that flows and technical factors tend to outweigh fundamental considerations in the current market.
Triple-A floating-rate spreads (bp)

300 bp Prime RMBS Leverage loan CLO Bank unsecured SME CLO Conduit CMBS Covered Bonds 250
Triple-B floating-rate spreads (bp)

900 800 700 600 500 bp Prime RMBS Leverage loan CLO Corporates SME CLO Conduit CMBS
Credit spread differentials

700 600 500 400 300 AAA-BBB differential (bp) Prime RMBS Credit Card ABS Non Conf. RMBS Leveraged loan CLO Conduit CMBS iBoxx Financials Index
200
150
400 300 200
100
200 100
Feb-08 Mar-08 Aug-07 Oct-07 Nov-07 Dec-07 Sep-07 Apr-08 Jul-07 May-08 Jun-07 Jun-08 Jan-08 Jul-08
50
100
Nov-07
Mar-07
Mar-06
Mar-08
Nov-06
Sep-06
Sep-07
Jan-06
Jan-07
Jan-08
Jul-07
May-06
May-07
May-08
Jul-06
Jul-08
-50
May-06
Nov-06
May-07
Nov-07
May-08
Source: Deutsche Bank, DBIQ
Page 40
6 August 2008
Japanese ABS
Research Contacts
Yukio Egawa (81) 3 5156 6163 Michiko Sakai (81) 3 5156 6157
While the corporate bond market is active despite the earnings season, the Japanese securitization market remains quiet. Issuance of ABS during the month of July amounted to JPY 560 billion, making year-to-date issuance of JPY 3.2 trillion, or 33% below the same period last year. We have lowered our full-year issuance projection to JPY 6 trillion, which would translate into a year-on-year decline of roughly 30%. Issuance in the RMBS and CMBS sectors is being constrained by the individual circumstances of a few large originators, and the growing reluctance by securities firms to arrange or underwrite new securitizations. The second series of triple-A13 rated FILP (Fiscal Investment and Loan Program) CLOs were launched on July 30. The CLOs are backed by loan receivables originated by the Japanese government (the Ministry of Finance) that are owned by 12 zaito agencies, all of which are currently wholly-owned by the government. The fixed-rate coupon on the 10-year, JPY 100 billion deal was fixed at 1.98%, 42 bp over the corresponding JGB yield, an attractive spread versus comparable bonds. For example, this is a 17 bp pick-up over recent debt issued by a zaito agency, the 10-year JPY 50 billion bonds by Development Bank of Japan (Aaa/AA-/AA/AAA Moodys/S&P/R&I/JCR), which recently priced at 25 bp over JGB, and a 7 to 12 bp pick-up over a generic single-A rated14 major bank senior debt, at 30 to 35 bp over JGBs. The deal enjoys initial subordination of approximately 20%. The top two obligors with the highest concentration in the pool are the Japan Housing Finance Agency (Aaa/AA/AAA Moodys/S&P/R&I) with 17.7% concentration and Urban Renaissance Corporation (not rated by S&P; Aaa/AA/AA Moodys/R&I/JCR) with 15.8% concentration. Both of these obligors have close relationship with the Ministry of Land, Infrastructure, Transport and Tourism (MILITT). S&P states, in its various reports on the deal and its rating methodology, that it used its CDO rating model, CDO Evaluator, in assigning its AAA rating to the deal. The rating agency assumed, on average, a correlation factor of 0.42 and recovery of 55% (loss severity of 45%). S&Ps ratings on the Japanese government bonds, Japan Bank for International Cooperation and the Japan Housing Finance Agency are all AA (double-A) and those on other governmentowned entities are AA- (double-A minus) or lower. It appears, judging from these examples,
13 14
Rated AAA by S&P and R&I). S&Ps rating.
Monthly/cumulative ABS issuance (JPY bn/trn) (2008 YTD)

Monthly (JPY bn)
2,200 2,000 10 1,800 1,600 1,400 1,200 1,000 800 600 400 200 0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Upgrades/downgrades of Japanese ABS (no. of tranches) (2008 YTD)

90
12
Breakdown by asset type (2008 YTD)

Others 1.9% CDO 7.5% Equipment Leases 9.6% Credit Receivables 5.9%
Cumulative (JPY trn)

90 80 70 60 50 40
8
81 73 65 54 39 28 17 50 42 43 59 64 56
79
75 74 62
30 20 10
17
CMBS 17.6%
0 10 20 1 1 3 8 14 8 3
30 40 50
10
Consumer Loans 5.2%
60 70
Upgrades
Downgrades
80 90
04 04 04 04 05 05 05 05 06 06 06 06 07 07 07 07 08 08 08 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q
2006
2007
2008
RMBS 52.4%
Note: Excluding those assigned short-term ratings Source: Deutsche Securities Inc.
Page 41
6 August 2008
S&Ps ratings on government-owned entities in Japan are capped at AA. The agency asserts that the correlation factor of 0.42 is relatively high, as it typically uses 0.3 (or, 30%) correlation factor among corporate bond issuers belonging to the same industry group. In our view, given that the collateral behind this transaction is rated at least two notches lower than the bonds offered, and that all of the obligors are wholly-owned by the government, it is quite possible that S&Ps AAA-rating of the CLOs was met with skepticism from market participants. In an attempt to understand this process, we examined S&Ps CDO Evaluator. In this model different default rates are assigned to each of corporates, ABS and CDO Quantiles. For example, 10-year cumulative default rate for corporates are 0.3621% at AAA, 0.5365% at AA+, 0.8717% at AA, 1.1302% at AA-, 1.4584% at AA- and 1.7816% at A+. For CDOs these rates are much higher; 0.728% at AAA, 1.013% at AA+, 1.4925% at AA and 1.8764% at AA-. The agency considers recovery for the underlying pool but assigns ratings based on default probability of the CDO tranches. Thus, assuming a 10-year A+ rated corporate bond with default probability of 1.4584% and a 55% recovery assumption is repackaged as a CDO with a default probability of 0.6563%, or 1.4584% times (1-0.55), the bond will get a rating of AAA because the target default probability of AAA-rated CDO tranche is 0.728%. Thus, S&Ps AAA ratings on CDO tranches do not necessarily reflect an equivalent credit quality of corporate bonds rated AAA. In this case, perhaps, the AAA rating should be interpreted as A+ or higher on the corporate bond rating scale. A Moodys report entitled Global Structured Finance Recap: A Summary of 2007 Review and 2008 Outlooks Across Asset Classes with Methodological Updates (June 4, 2008) makes numerous negative comments in relation to various sectors of the U.S. securitization market (citing higher-than-anticipated defaults, numerous downgrades, negative outlooks for many sectors and industries, and substantial revisions to its rating methodology), but is largely positive (or at worst neutral) with regard to various Japanese asset classes. This is consistent with our own belief that Japanese securitized products remain sound. Despite the fact that the vast majority of Japanese securitized products do not suffer from the problems that have afflicted some U.S. securitized products, primary market activity is clearly in a slump. So what can be done to breathe life back into the Japanese securitization market? On July 7 the Securities Industry and Financial Markets Association (SIFMA), the American Securitization Forum (ASF), and the European Securitsation Forum (ESF)three important industry groups comprised of entities with an interest in the U.S. and European bond and securitized products marketsestablished a working group aimed at restoring confidence in the securitization and structured credit markets. Similar initiatives could benefit the Japanese securitization market and may be undertaken in Japan as well. Breakdown by ratings (2008 YTD)
BBB+~BBBA+~A- 2.3%
Historical ABS Issuance (JPY trn) (2008 YTD)

12
Breakdown by seller industry (2008 YTD)

Consumer Credit (Shinpan) 2.8%
3.5% AA+~AA4.0%
B+~B0.2%
11 10 9 8 7 6 5 4 3
Others
CDO
Others, N/A 28.8% Non-Life Insurance 0.4% Captive Finance 5.0%
Leasing 5.1% Banks 15.5% Life Insurance 1, 1%
CMBS
RMBS
Consumer Loans Credit Receivables Equipment Leases

00 01 02 03 04 05 06 07 08
AAA 89.4%
2 1 0
Government affiliated 42.4%
Note: Excluding those assigned short-term ratings Source: Deutsche Securities Inc
Page 42
6 August 2008
Recommendation
In our view, for buy and hold investors that are not particularly concerned about secondary market liquidity, Japanese ABS continue to be an attractive alternative to lower-yielding Japanese corporate bonds. We believe that with the exception of a few specific securitizations in limited sub-sectors, the credit fundamentals of broader ABS market remain firm. Investors should perhaps seek opportunities in the secondary market, or inventory being held by intermediaries as primary market activities continue to be constrained. Japanese and non-Japanese securities firms have grown increasingly reluctant to arrange or underwrite new securitizations given the turmoil in the U.S. and European securitization market. Additionally growing inventories of securitized products and increasingly strict internal restrictions on traders positions have done little to assuage participants reluctance. Indeed, some securities firms may continue to avoid involvement in new issues until or unless they can slim down their balance sheets by selling their inventories on the secondary market.
Page 43
6 August 2008
Appendix
U.S. fixed-rate ABS spreads to Treasuries (bp)
900
841 HEL (High 8/08)
800 Spreads to Treasuries (bp) 700

640 MH (High 4/08)
841
600 500 400 300 200 100 0 11/94 5-year Credit Card 4/98 2-year Auto 9/01 5-year Home Equity Loan 2/05 8/08 7/08
29 CC (Low 3/97) 47 MH (Low 3/97) 23 Auto (Low 5/97) 63 HEL (Low 10/97)
616
241 CC (High 4/08) 225 Auto (High 4/08)
205 190
5-year Manufactured Housing
U.S. collateral type breakdown (floating) (YTD) 71.7% of public ABS issuance
Home Equity Loan 2.9% Auto 14.0% Credit Card 51.5% Student Loan 31% Equipment 0.7%
U.S. collateral type breakdown (fixed) (YTD) 28.3% of public ABS issuance
Other 2.2% Equipment 7.9% Student Loan Home Equity 0.2% Loan 0.2%
Credit Card 34.4%
Auto 55.1%
Note: Collateral type breakdown is for public issuance only Source: Deutsche Bank
Page 44
6 August 2008
Appendix 1
Important Disclosures Additional information available upon request
For disclosures pertaining to recommendations or estimates made on a security mentioned in this report, please see the most recently published company report or visit our global disclosure look-up page on our website at http://gm.db.com.
Analyst Certification
This report covers more than one security and was prepared by more than one analyst. The views expressed in this report accurately reflect the personal views of each undersigned lead analyst about the subject issuers covered by each, and the securities of those issuers. In addition, the undersigned lead analysts have not and will not receive any compensation for providing a specific recommendation or view in this report. Karen Weaver
Deutsche Bank debt rating key Buy: These bonds are expected to outperform other issues in the sector/industry group over the next three to six-month period. Hold: These bonds are fairly valued currently. If owned, no need to sell, but we await events/ releases/ conditions that would make the bond attractive enough for us to upgrade. In the interim, the bond will likely perform as well as the average issue in the sector/industry group. Sell: There exists a significant likelihood that these bonds will underperform relative to other issues in their sector/industry group, at least over the next three months.
Page 45
6 August 2008
Regulatory Disclosures 1. Important Additional Conflict Disclosures

Aside from within this report, important conflict disclosures can also be found at https://gm.db.com/equities under the "Disclosures Lookup" and "Legal" tabs. Investors are strongly encouraged to review this information before investing.
2. Short-Term Trade Ideas

Deutsche Bank equity research analysts sometimes have shorter-term trade ideas (known as SOLAR ideas) that are consistent or inconsistent with Deutsche Bank's existing longer term ratings. These trade ideas can be found at the SOLAR link at http://gm.db.com.
3. Country-Specific Disclosures
Australia: This research, and any access to it, is intended only for "wholesale clients" within the meaning of the Australian Corporations Act. EU countries: Disclosures relating to our obligations under MiFiD can be found at http://globalmarkets.db.com/riskdisclosures. Japan: Disclosures under the Financial Instruments and Exchange Law: Company name - Deutsche Securities Inc. Registration number - Registered as a financial instruments dealer by the Head of the Kanto Local Finance Bureau (Kinsho) No. 117. Member of associations: JSDA, The Financial Futures Association of Japan. Commissions and risks involved in stock transactions - for stock transactions, we charge stock commissions and consumption tax by multiplying the transaction amount by the commission rate agreed with each customer. Stock transactions can lead to losses as a result of share price fluctuations and other factors. Transactions in foreign stocks can lead to additional losses stemming from foreign exchange fluctuations. New Zealand: This research is not intended for, and should not be given to, "members of the public" within the meaning of the New Zealand Securities Market Act 1988. Russia: This information, interpretation and opinions submitted herein are not in the context of, and do not constitute, any appraisal or evaluation activity requiring a license in the Russian Federation.
Page 46
David Folkerts-Landau
Managing Director Global Head of Research Global Company Research Stuart Parkinson Chief Operating Officer Europe Pascal Costantini Regional Head Principal Locations
Deutsche Bank AG London 1 Great Winchester Street London EC2N 2EQ Tel: (44) 20 7545 8000 Deutsche Bank AG New York 60 Wall Street New York, NY 10005 United States of America Tel: (1) 212 250-2500 Deutsche Bank AG Boston 225 Franklin Street Boston, MA 02110 United States of America Tel: (1) 617 988 8600 Deutsche Bank AG Hong Kong Cheung Kong Center, 2 Queen's Road Central Hong Kong Tel: (52) 2203 8888 Deutsche Bank AG Singapore One Raffles Quay South Tower Singapore 048583 Tel: (65) 6423 8001 Deutsche Bank AG Japan 2-11-1 Nagatacho Sanno Park Tower Chiyoda-ku, Tokyo 100-6171 Tel: (81) 3 5156 6701 Deutsche Bank AG Australia Deutsche Bank Place, Level 16 Corner of Hunter & Phillip Streets Sydney NSW 2000 Tel: (61) 2 8258 1234
Global Fixed Income Strategies & Economics Marcel Cassard Global Head
Guy Ashton Global Head Asia-Pacific
Germany Andreas Neubauer Regional Head
Americas Steve Pollard Regional Head
Michael Spencer Regional Head
Deutsche Bank AG Frankfurt Groe Gallusstrae 10-14 60272 Frankfurt am Main Germany Tel: (49) 69 910 0
Subscribers to research via email receive their electronic publication on average 1-2 working days earlier than the printed version. If you would like to receive this or any other product via email please contact your usual Deutsche Bank representative. Publication Address: Deutsche Bank Securities Inc. 60 Wall Street New York, NY 10005 United States of America (1) 212 250 2500 Internet: http://gmr.db.com Ask your usual contact for a username and password.
Global Disclaimer
The information and opinions in this report were prepared by Deutsche Bank AG or one of its affiliates (collectively "Deutsche Bank"). The information herein is believed to be reliable and has been obtained from public sources believed to be reliable. Deutsche Bank makes no representation as to the accuracy or completeness of such information. Deutsche Bank may (1) engage in securities transactions in a manner inconsistent with this research report, (2) with respect to securities covered by this report, sell to or buy from customers on a principal basis, and (3) consider this report in deciding to trade on a proprietary basis. Opinions, estimates and projections in this report constitute the current judgment of the author as of the date of this report. They do not necessarily reflect the opinions of Deutsche Bank and are subject to change without notice. Deutsche Bank has no obligation to update, modify or amend this report or to otherwise notify a recipient thereof in the event that any opinion, forecast or estimate set forth herein, changes or subsequently becomes inaccurate. Prices and availability of financial instruments are subject to change without notice. This report is provided for informational purposes only. It is not an offer or a solicitation of an offer to buy or sell any financial instruments or to participate in any particular trading strategy. The financial instruments discussed in this report may not be suitable for all investors and investors must make their own informed investment decisions. Stock transactions can lead to losses as a result of price fluctuations and other factors. If a financial instrument is denominated in a currency other than an investor's currency, a change in exchange rates may adversely affect the investment. Past performance is not necessarily indicative of future results. Unless governing law provides otherwise, all transactions should be executed through the Deutsche Bank entity in the investor's home jurisdiction. In the U.S. this report is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this report is approved and/or communicated by Deutsche Bank AG Frankfurt authorized by the BaFin. In the United Kingdom this report is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange and regulated by the Financial Services Authority for the conduct of investment business in the UK and authorized by the BaFin. This report is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. and in Singapore by Deutsche Bank AG, Singapore Branch. In Japan this report is approved and/or distributed by Deutsche Securities Inc. The information contained in this report does not constitute the provision of investment advice. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Deutsche Bank AG Johannesburg is incorporated in the Federal Republic of Germany (Branch Register Number in South Africa: 1998/003298/10). Additional information relative to securities, other financial products or issuers discussed in this report is available upon request. This report may not be reproduced, distributed or published by any person for any purpose without Deutsche Bank's prior written consent. Please cite source when quoting. Copyright 2008 Deutsche Bank AG
GRCM2008PROD013927

Securitization Aug08

Загружено:

Сведения о документе

Исходное описание:

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Securitization Aug08

Загружено:

Авторское право:

Доступные форматы

Securitization

North America Asset-Backed Securities

Global Securitization Research

Global Markets Research

Securitization Monthly August 2008

Table of Contents U.S. ABS sector breakdown (YTD)

CDO 11% Auto 24%

Credit Card 39%

Student Loan 17%

Source: Thomson Financial Securities Data

U.S. fixed-rate spreads to swaps (bp)

U.S. floating-rate spreads (bp)

140 115 2/08 4/08 6/08 8/08

0 8/07 10/07 12/07

80 75 2/08 4/08 6/08 8/08

5-year Credit Card 5-year CMBS

Year-over-year U.S. ABS issuance change

$400 bil. 2008 YTD

Asset-Backed Securities Securitization Monthly

Deutsche Bank Securities Inc.

Asset-Backed Securities Securitization Monthly

400 OFHEO House Price Index

NAR Existing Home Sales (rs)

OFHEO House Price Index

Deutsche Bank Securities Inc.

Asset-Backed Securities Securitization Monthly

YOY Change in OFHEO Housing Price Index

Price-Decline Peak to Trough 22% 12% 9% 12%

Peak to Trough Horizon 6yr 5yr 6yr 4yr

Peak to Peak Horizon 10yr 8yr 9yr 9yr

* Nominal. Home price peak is the quarter in which peak occurred.

Deutsche Bank Securities Inc.

Asset-Backed Securities Securitization Monthly

Home price peak

Home price trough

Deutsche Bank Securities Inc.

Asset-Backed Securities Securitization Monthly

Home price peak

Home price trough

Home price peak

Home price trough

Deutsche Bank Securities Inc.

Asset-Backed Securities Securitization Monthly

Figure 8: Unemployment rate expected to rise into 2009

Home price peak

6.5% 6.0 5.5%

3.5% 1/02 4/03 7/04 10/05 1/07 4/08 5/08

3.5% 1/02 4/03 7/04 10/05 1/07 4/08 5/08

Home price peak

5.5% 4.9 4.5%

5.0% 4.4 4.0%

3.5% 1/02 4/03 7/04 10/05 1/07 4/08 5/08

3.0% 1/02 4/03 7/04 10/05 Boston

New York MSA

Deutsche Bank Securities Inc.

Asset-Backed Securities Securitization Monthly

Autos, Credit Cards, Student Loans

Prime auto fixed-rate spreads to swaps (bp)

Auto public/144A issuance (US$bn)

Source: Thomson Financial Securities Data

Auto issuers (YTD)

Subprime auto fixed-rate spreads to swaps (bp)