Vous êtes sur la page 1sur 43

SECURITIZATION RESEARCH

9 September 2011

SECURITIZED PRODUCTS WEEKLY


Trends and issues 3
Views on a Page Ajay Rajadhyaksha +1 212 412 7669 ajay.rajadhyaksha@barcap.com www.barcap.com 2

President Barack Obamas long-awaited jobs speech on Thursday outlined a $447bn package of tax cuts, infrastructure spending and transfer payments but lacked specifics on a mortgage refinancing plan. While some changes to HARP have been proposed, the agency MBS market will likely remain in a state of limbo absent further details. On the economic front, mostly weak data have further focused attention on a potential monetary policy easing at the Fed meeting in a couple of weeks. Fed Chairman Ben Bernankes reiteration of the presence of a range of tools to stimulate what he portrayed as a gloomy economy reinforced expectations of an easing. Agency MBS continued to be weighed down by government refi fears, even as non-agencies remained stable.

Prepayment commentary: Re-assessing refinancing risks

13

Total paydowns rose 13% m/m in the September report, mostly because of a three day increase in day count. Bank of Americas exiting the correspondent business makes it less likely that it will greenlight the HARP program. The proposed American Jobs Act signals that a change to the HARP program is imminent. However, we expect the speed impact from any immediate legislation to be moderate. Relative to conventionals, we think GNs are less prone to policy risk.

CMBS: Trust implications of super-high loss severities

23

With liquidation volumes remaining elevated, increasing numbers of loans are reporting severities north of 100%. We discuss some recent examples and examine the impact on deal cash flows.

Consumer ABS: Relative value among prime retail auto loan ABS

27

We assess the collateral performance and credit enhancement of the largest prime retail auto loan ABS issuers and identify relative value within the prime retail auto ABS sector.

Convexity/credit portfolio

33

The convexity portfolio lost 68bp over the week, bringing total ROE to 105.63%. Yearto-date ROE stands at -0.18%. The credit portfolio gained 22bp this week. The overall return since inception stands at 49.5%, and the year-to-date return is -0.78%.

PLEASE SEE ANALYST CERTIFICATIONS AND IMPORTANT DISCLOSURES STARTING AFTER PAGE 40

Barclays Capital | Securitized Products Weekly

VIEWS ON A PAGE
Category Agency basis Comments The mortgage basis could continue to face near-term headwinds from policy risk. The draft of President Obama's American Jobs Act indicates that some measures to spur refinancing are in the works, with some modifications to the existing HARP program appearing more likely. While changing the eligibility date, removing LLPAs, etc. should cause only a mild prepay pickup, the market seems jittery about more drastic measures, such as waiving reps and warranties. While we believe this is very unlikely, we recommend maintaining a down-in-coupon bias until there is more clarity. We retain our longs in 30y 4s and 15y 3.5s. Despite a new all-time low in mortgage rates, prepayment risk remains contained. Rates will need to rally below 4.1% to reach the same level of aggregate pay-downs of last year's peak. Bank of America's exiting the correspondent business makes it less likely that it will greenlight the HARP program. The proposed American Jobs Act signals that a change to the HARP program is imminent. However, we expect the speed effect from any immediate legislation to be moderate. For GNMA, we believe Bank of America will execute a substantial delinquency clean-up in the near future. Relative to conventionals, we think GNs are less less prone to policy risk. The mortgage delinquency pipeline made further improvement in July. REO stock fell to 476K from June's 492K. Mortgages 90+ days delinquent fell to 1.57mn, from 1.58mn in June, while those in the foreclosure process remained flat at 1.91mn. Since peaking in February 2010 at 4.22mn, the shadow inventory of homes 90+ days delinquent or in foreclosure has now fallen 17%, to 3.48mn. We remain constructive on non-agencies, with a bias toward jumbo and alt-A fixed rates. We believe they offer an attractive pickup over comparable assets and think they could return double-digit yields with leverage. Our favorite trade remains being long Alt-A and jumbo FRM with leverage. We also recommend being long PrimeX FRM and ARM1 as a leveraged and cheap way to own the jumbo IOs. Despite the threat of some near-term technical selling pressure from legacy holders, macroeconomic concerns and negative trends in stop advances/housing, we believe subprime LCFs look attractive after the recent sell-off. These bonds should benefit from elevated severities over the next 18-24 months, followed by a sharp decline in the back end. This trade can also be expressed on the synthetic side by going long ABX 07-1 AAA. In cash, we prefer sticking to servicers with reliable performance on mods and advances, such as Wells Fargo and Chase. We also find subprime seasoned mezz attractive and would recommend adding on dips. On the Countrywide settlement, our base case is still that it will get approved for most of the deals, but with some delays, with cash flowing to the deals in mid-2012. Recent spread widening, driven primarily by global macro concerns, provides opportunities for selective longs higher up in the capital stack. We initiated a long 2006 AM and long 2007 duper trade, which should remain protected from losses, even in a stress scenario. We maintain a long 2007 AJ position and remain positive on 2007 AMs and back-end re-REMICs. We stay neutral lower in the capital structure in cash; however, we maintain a long CMBX.AA.2 recommendation in the index space. Long fixed coupon jumbo/alt-A FRM SSNRs Prefer subprime LCFs over front pays, ABX 07-1 AAA PrimeX.FRM1 and PrimeX.ARM1 are a cheap and leveraged way to own jumbo FRM and ARM IO Prefer amortizing IOs loans from the jumbo FRM/ hybrid space Favor seasoned perma-fail subprime mezz despite very strong outperformance Buy back-end re-REMICs off 2007 GG10 A4s Buy 2007 AMs Buy 2007 Pac IOs Buy 2007 AJs Buy CMBX.AA.2 Long 2006 AM Long 2007 dupers Buy FFELP ABS Buy retail auto ABS Buy next-pay private credit student loan ABS Buy recent vintage rental fleet ABS Trade recommendations Long FNCL 4s vs. Tsy curve Long DW 3.5s vs. Tsy curve Long DW 3.5/FNCL 4.5

Prepayments

Housing

Residential credit

CMBS

Consumer ABS

At current spread levels, we believe FFELP ABS present compelling relative value among consumer ABS and reiterate our overweight on the sector. We also maintain our overweight on retail auto ABS and other auto-related sectors, including dealer floorplan and rental fleet securitizations. However, we are neutral on bank credit card ABS, given tight spread levels and limited upside potential. Lastly, we are overweight equipment and current-/next-pay private credit student loan ABS. The market is already priced for a weak economic outlook and additional long end purchases by the Fed. We are neutral on duration, owing to uncertainty regarding fiscal policy and the risk from the Feds disappointing the market. We recommend underweighting 3s versus the wings, as the risk of asset sales in the front end is non-trivial, and OTR10s and 30s versus older issues in respective sectors.

Rates, curve, and volatility

Source: Barclays Capital

19 August 2011

Barclays Capital | Securitized Products Weekly

TRENDS AND ISSUES


Overview
President Obama delivered his much-awaited jobs speech on Thursday night, laying out a $447bn package of tax cuts, infrastructure spending, and transfer payments. Although there was a mention of easier refinancing options for responsible homeowners, mortgage investors are likely to be disappointed by the lack of any specifics. The language in the proposed bill states that the administration will work with the GSEs, FHFA, and industry leaders to remove barriers that exist in the current refinancing program (HARP). To us, that language indicates a lack of agreement among the various stakeholders on the course of action. As we mentioned in Refi realities (August 26, 2011), we do not foresee any changes to HARP that can boost refinancing for burdened borrowers without affecting market participants in a significant way. The FHFA lawsuit against banks and a CBO report discounting any economic benefits from large-scale refinancing programs add to the headwinds facing any such proposal. As a result, agency MBS, especially in higher coupons, is likely to remain in limbo until the administration comes out with a final proposal. On the economic front, all eyes are now focused on the September 20-21 Fed meeting, as most economic data continued to disappoint. Claims came in above expectations, on the heels of last weeks flat payroll growth number, and there are also worrying signs of a slowdown in US manufacturing exports. The ongoing crisis in the euro area, with seemingly new adverse developments almost every day, continues to weigh on financial markets. In a speech in Minnesota, Chairman Bernanke continued to paint a gloomy picture of the US economy. He reiterated that a range of tools can be used. Markets are now dominated by speculation about the nature of easing, with terms such as twist and torque being thrown around. Our economics team expects further policy easing at the meeting. Within the securitized sector, agency MBS continued to be hurt by the uncertainties about refinancing programs, with higher coupons bearing most of the pain. However, the low refi index print remains a positive for the sector, along with the significant yield advantage over Treasuries. Ginnies also outperformed conventionals, as they are seen to be less susceptible to government-induced refi risk. Non-agencies mostly remained stable on light volumes. The FHFA lawsuit against banks, as well as an increase in investor-driven rep and warranty lawsuits, is likely to continue weighing on the sector. CMBS also remained flat over the week on low volumes.

Agency MBS: Policy risk weighs on MBS


Risk aversion picked up sharply this week, as sovereign credit issues in Europe and weak economic data weighed heavily on investor sentiment. The most recent shot across the bow was the unchanged non-farm payroll number last Friday, which led to a sharp flattening of the yield curve, sending 10y Treasury yields plunging to sub-2% levels. While low yields have put pressure on MBS performance, the mortgage market continues to be plagued by policy concerns. On the refinancing front, it had been widely speculated that President Obama might unveil a new refinancing initiative. While he did mention that to help responsible homeowners, were going to work with Federal housing agencies to help more people refinance, he stopped short of presenting any concrete initiatives. However, his mention of it in his

9 September 2011

Barclays Capital | Securitized Products Weekly

speech is significant in that it now appears that there will be some imminent changes to stoke refinancing activity. Based on the FAQs of the Administrations American Jobs act, 1 the administrations focus looks to be on working with the FHFA and the GSEs to remove barriers in the existing HARP program. As we discussed in Refi realities, August 26, 2011, we think that the most likely alternatives would be to remove loan level pricing adjustments, remove the 125% LTV cap, and change the HARP eligibility date. These changes by themselves should only have a moderate effect on refinancing activity. However, larger steps such as addressing rep and warranty relief could also be considered, although we think this is less likely. In the absence of concrete details, higher coupons are likely to remain under pressure. We recommend a down-in-coupon bias until there is more clarity. On the monetary front, MBS could face other headwinds. Our colleagues in the Economics team note that the Fed may try to push long term yields lower, which could cause further volatility for the mortgage basis. Given the policy-related uncertainty, most investors remained on the sidelines this week. Flows remained light overall, and despite a let-up in origination, FN 4s were down 12 ticks versus 10y Treasury hedges. The up-in-coupon move over the prior week reversed sharply, as higher coupons bore the brunt of the uncertainty over new refinancing initiatives. FNCL 5s through 6s were down 17-22 ticks over the week. On a Treasury curve- and vol-hedged basis, the coupon stack showed a clear down-in-coupon trend (Figure 1), with 3.5s and 4s down 3 ticks, while 5.5s and 6s were down 21-22 ticks. Bank demand remains strong in lower coupon 15y mortgages. The DW 3 roll continues to trade extremely special (about 6 ticks rich, assuming zero CPR), as bank demand is particularly heavy in this coupon, but has been met with insufficient supply. GNs continued to outperform relative to FNs over the week, reaching extremely high levels, with the GN/FN 4.5 swap closing at 2-26 on Thursday. As we have written before, this sector has benefited from the recent talk about government-induced refis, as GNs considered relatively immune to the risk.

Figure 1: Sharp widening in the coupon stack (ticks)


0 -5 -10 -15 -20 -25 FNMA 3.5s FNMA 4s FNMA 4.5s FNMA 5s FNMA 5.5s FNMA 6s

Figure 2: GN/FN swaps march higher (ticks)


90 85 80 75 70 65 60 55 50 1Jun 15Jun 29Jun 13Jul 27Jul 10Aug 24Aug GN/FN 5s 7Sep

vs 10y swap

vs 10y Tsy

vs Tsy curve and vol

GN/FN 4s
Source: Barclays Capital

GN/FN 4.5s

Note: Performance from Sep 1 to Sep 8 close. Source: Barclays Capital


1

The press release notes: The President has instructed his economic team to work with Fannie Mae and Freddie Mac, their regulator the FHFA, major lenders and industry leaders to remove the barriers that exist in the current refinancing program (HARP) to help more borrowers benefit from todays historically low interest rates. This has the potential to not only help these borrowers, but their communities and the American taxpayer, by keeping borrowers in their homes and reducing risk to Fannie Mae and Freddie Mac. http://www.whitehouse.gov/the-pressoffice/2011/09/08/fact-sheet-american-jobs-act

9 September 2011

Barclays Capital | Securitized Products Weekly

Regular refinancing activity remained tepid, with applications heading lower this week. The MBA refi index came in at 3,169, down 6.3% w/w. However, as an indication of last weeks applications, this number does not reflect current mortgage rates (the Freddie Mac PMMS no-point rate dropped 10bp to 4.3% this week). As we have written earlier, the falling refi index was observed in last years refi episode as well, as newly incentivized borrowers refi away quickly. In addition to the slower-than-expected September prepayment print, this remains a positive for agency MBS. While mortgages do face near-term headwinds, at 60+bp L-OAS across the stack, valuations appear attractive. Recent actions such as the FHFA lawsuit against banks lead us to believe any drastic measures, such as rep and warranty relief, which may be perceived as yet another bank bail-out, are extremely unlikely. Also, the working paper released by the Congressional Budget Office casts doubt on the effective stimulus provided by any largescale refi program, likely creating headwinds for the implementations of any such effort. That being said, we do think that near-term policy concerns argue for a rather conservative stance and a strong down-in-coupon focus.

CBO pours cold water over benefits of large-scale agency refi programs
The CBO released a working paper on large-scale mortgage refinancing programs this week, in which it examined the macroeconomic effect of a streamlined refinancing program for agency mortgages. Overall, it found that there is no significant economic benefit from implementing such a program. The effect on housing is also unlikely to be significant.

CBO's stylized refinancing program


Using a combination of some key features of various recent streamlined refinancing proposals, the CBO constructed a stylized refinancing program. It assumes no restrictions on LTV and income, a fixed interest rate at the prevailing rate, a 30y maturity and guarantee fee capped at previous levels. It is assumed to remain in effect for a year for all Fannie, Freddie and FHA fixed rate loans that are current and have not had a 20-day delinquency in the past 12 months. The CBO also caps origination fees at 1% of the balance or $1000. It then used this custom model to predict prepayment rates on various cohorts bucketed by coupon, vintage, product and FICO, in the base case and under the new program.

Costs of the program mostly overwhelm the benefits


On the benefits side, the CBO model projects about $7.4bn of direct cash flow savings for borrowers in the first year. This is driven by the lower monthly payments. The GSEs and FHA are expected to save about $3.9bn in guarantee losses due to better performance. Additionally, there is some small potential for upside from lower default and foreclosure levels that is not quantified. The fee upside for originators and other parties in the origination chain is also assumed to be small, given the low cap on origination costs. However, there are multiple costs that more or less overwhelm the benefits. First, the GSEs, the Fed and Treasury are expected to lose about $4.5bn on their portfolio holdings due to faster prepayments on premiums. Second, private investors are expected to lose $13-15bn on their holdings due to faster prepayments. Finally, the CBO also assumes $100mn in losses to the agencies due to the non-enforceability of rep and warranties on the refinanced loans that eventually default. Thus, there is no overall economic benefit from the program, as borrower savings are more than offset by losses elsewhere. The CBO argues that some of that could be mitigated by the fact that most of the costs are being borne by government entities and foreign investors. This could keep some of the domestic stimulative effects intact. Likewise on the
9 September 2011 5

Barclays Capital | Securitized Products Weekly

housing front, there is unlikely to be significant upside. Underwriting for purchase loans is unlikely to be eased and refinancing by current borrowers will not lead to incremental demand for homes. There is only a marginal benefit from slightly lower level of defaults.

No upside even with alternative programs


The CBO also presented a couple of alternative scenarios, including one in which delinquent borrowers are also refinanced and the other in which origination costs are higher. Though there is some reallocation of costs, the overall conclusion remains the same. Additionally, it is worth mentioning that some of the CBO assumptions lean towards the benign side. First, the rep and warranty relief cost of $100mn seems extremely low. The potential opposition of banks to refis due to rep and warranty concerns, as well as estimates from various lawsuits, clearly shows that the numbers involved are orders of magnitude higher. Second, the analysis does not incorporate the potential for much higher mortgage rates if investors are forced to stomach large losses. Third, the origination cost assumption is much lower than what is evident in the market. Finally, the implicit revenue loss for GSEs and FHA from not realizing higher g-fees or MIP from borrowers who would refinance anyway is not incorporated. The key takeaway is that the CBO does not expect any upside from these large-scale refinancing programs, even using some benign assumptions. As a non-partisan agency that carries significant weight in government circles, this report will work to further weaken the case for such programs. In Refi realities, August 26, 2011 we examined various refinancing approaches based on mortgage valuations. This report takes the analysis to the macroeconomic level and concludes that there is no silver bullet hidden in most of these proposals.

Residential credit: Markets remain volatile


Equity and credit markets continued to seesaw this week as markets continued to react to the weak jobs report last Friday and the whims of European sovereign credit sentiment. Prices in non-agencies were largely stable, with a slight rise in alt-A and jumbo fixed bonds (0.5pt), while the rest of the market was flat. Volume picked up modestly as market participants returned after very light August trading. ABX prices fell by around 0.5pt, mostly in line with equity moves. PrimeX fell marginally on the week, with FRMs down 0.25pt and ARMs nearly flat. We continue to believe that non-agencies look attractive on both a relative and an outright basis and, in our opinion, are likely to outperform other risky assets, such as credit, across a range of scenarios. We continue to favor high-coupon alt-A and jumbo FRMs, with or without leverage, and subprime LCF/locked-out PAAAs that are near crossover, though we caution that the latter trade is much more sensitive to the macro environment. We also find value in seasoned mezzanine subprime bonds from deals in which triggers will continue to fail and also like FHA/VA re-performer deals, which we think offer stable return profiles.

9 September 2011

Barclays Capital | Securitized Products Weekly

Figure 3: Non-agency prices


Price Jumbo fix SS AAA Jumbo hyb SS AAA Alt-A fix SS AAA Alt-A hyb SS AAA MTA SS AAA ABX 06-2 PAAA ABX 06-2 AAA ABX 07-1 PAAA ABX 07-1 AAA PrimeX FRM.2 PrimeX ARM.2 Re-REMIC SSNR AAA Re-REMIC SSNR A Current 90 78 79 58 54 73 47 46 37 99 94 S + 200-220 S + 260-275 1 week 0.5 0 0.5 0 0 (0.7) (0.2) (0.3) (0.7) (0.2) (0.0) 0 0 1 month 0 (1) (1) (2) (2) 1.7 (1.4) (0.6) (0.0) (1.4) 0.5 0 (20) 3 month 1 (1) 1 (2) (0) 1.3 (0.1) 0.2 0.4 (3.8) (0.7) (10) (30)

Note: Prices as of Sep 7, 2011, for cash bonds and September 8, 2011, for synthetic indices. Weekly changes are Wednesday-Wednesday for cash bonds and Thursday-Thursday for the synthetic indices. Source: Barclays Capital

FHFA sues 17 banks for Securities Act violations


On September 2, 2011, the Federal Housing Finance Agency (FHFA) filed lawsuits against 17 financial institutions and some of their senior officers for allegedly having violated securities law when they sold approximately $196bn of private-label RMBS to Fannie Mae and Freddie Mac between 2005 and 2008. Although there are several causes of action listed by the FHFA in each of the suits filed against the banks, the primary allegations involve violations against the Securities Act of 1933 and state securities laws in Virginia and the District of Columbia, as well as common law fraud and negligent misrepresentation. The FHFA is claiming that by misrepresenting the quality of the loans that were placed into the securitization trusts, the banks violated the Securities Act of 1933 when they underwrote and sold these securities. As such, it is demanding that the banks compensate the GSEs for damages incurred on those bonds.

Signs of life on the AG Settlement?


News reports suggest that there has been some movement in the settlement talks between banks and the state AGs and that banks have been offered a deal. The reports also claim that although there has been some agreement over future standards of servicing, there has been little agreement on the monetary aspects of the plan or on the legal liabilities from which the banks want to be released. Details remain scant, but we continue to expect a deal to be reached over the next 6-12 months and that liquidation timelines will start contracting as that happens. There has already been a pickup in the rate of foreclosure filing from loans that are 60+ days delinquent and we expect the foreclosure to REO timelines to start shortening soon after a deal is reached. These timelines should also go down as the recently filed foreclosures flow through the system since they are expected to be cleaner than prior filings and will likely see fewer legal challenges. It remains to be seen what the monetary component of this plan amounts to and how much of that is likely to flow through to non-agency deals in the form of partial prepayments on borrowers that get debt forgiveness/other mods as a result of the settlement. Overall, unless some adverse details emerge, this plan should be a marginal positive for non-agency valuations.
9 September 2011 7

Barclays Capital | Securitized Products Weekly

Trustee rep and warranty lawsuits pickup


After US Bank filed a lawsuit on the HVMLT deal last week, another deal (MABS 2006-HE3) was the subject of a rep and warranty related lawsuit filed by US Bank as trustee. In our view, lawsuits are likely being driven by investors/investor groups that are alleging issues with mortgage underwriting. It is unclear whether this is the beginning of a trend in which US Bank as trustee systematically starts pursuing these lawsuits, or if these are one-off cases. In any case, we expect such lawsuits/other repurchase activity/settlements to pick up in the coming months, but these are unlikely to have a blanket effect on prices in the space. We continue to believe that the proportion of deals with cash flows from such repurchases/claims is likely to be small enough that the only way to profit from this as a strategy is to purchase securities backed by organizations likely to pay, where the trustee is more amenable to providing loan files and actively pursuing rep and warranty claims. Figure 4: Top trade recommendations
Type Overall view Trade recommendation Favorable Rationale Higher yields than other risky assets. Strong demand, cheap leverage, supply coming from strong sources that can afford to slow down if needed. High yields across scenarios, steady carry, limited downside potential. Enough support to protect principal, strong yields in stress scenarios, lower Libor forwards. High near-term severities hasten crossover, longer-term declines in severities benefit prices at recent lows. Limited further worsening possible because of supply pressures. Enough support to protect principal, strong yield profiles; buy on any dips. MTAs amortizing versus hybrids will likely have smaller payment shocks and, eventually, lower severities than hybrid option ARMs. IO loans underperforming by 1.5-2.0x, but amortizing IOs will likely have lower payment shocks and outperform.

Outright longs

Leveraged or unleveraged alt-A FRM/PrimeX.FRM1

Recovery oriented

2006-07 subprime pro rata LCF

Subprime 03-04 perma-fail mezz M1/M2 Collateral stories Favor MTA negams versus hybrid negams Favor deals with amortizing IO loans in jumbo
Source: Barclays Capital

CMBS: Spreads flat over the week on low volume


Spreads bounced around during the week in line with broader markets but ended Thursday roughly unchanged. Disappointing payroll numbers drew the benchmark GG10s wider by nearly 25bp on Tuesday, before tightening gradually toward the end of the short week. Overall spreads on 2007 LCF stayed at 320bp over swaps, and 2007 AMs were unchanged as well, at S+710. Supply was light this week, and demand from insurance companies and bank portfolios helped older vintage LCF and AMs tighten marginally. 2006 AMs ended the week 5-10bp tighter, to finish at S+580. On the new issue side, despite reports of a new CMBS conduit deal pricing in the coming days, spreads on 2.0 paper in the 10y AAA sector also tightened slightly. Absent major upside surprises in economic data, we expect much of the near-term spread action to be dependant on the markets expectation of any potential actions by the Fed. In a speech on Thursday, Chairman Bernanke once again indicated that the Fed would do all that it can to help restore high rates of growth and employment in a context of price stability, and our economics team believes that the FOMC is likely to take accommodative steps at its September meeting.

9 September 2011

Barclays Capital | Securitized Products Weekly

New CMBS conduit deal in the market


According to a Bloomberg report published on Thursday, a$1.5bn CMBS conduit deal (MSC 2011-C3) is being marketed. The pre-sale report published by Moodys suggests that the A1 through A4 30% enhanced classes would be in the public format while the 20% enhanced AJs and below would continue to be issued under SEC rule 144A. This is similar to the previous deal, DBUBS 2011-LC3, in which the senior classes were also issued in the public format. As we argued in CMBS Strategy Weekly, July 22, 2011, many investors might have limited allocation for 144A paper, and as such, issuing in public format is critical in stimulating demand. CMBS 2.0 spreads have widened over the prior months along with the rest of the market, with 10y AAA paper now trading close to S+225 after touching the S+100 mark in April/May. The deal contains two GGP properties that were recently refinanced Park City Center and Oxmoor Center ( see GGP announces refinancing of five malls, June 7, 2011). As such, retail remains the dominant property type, as is the case with most CMBS.2 issuance, with more than 46% of the balance backed by retail properties.

CDO liquidation scheduled for next week


The Newbury Street CDO, consisting of close to $240mn in CMBS bonds, is scheduled to be liquidated next week, after shortfalls on the most senior bonds in recent months likely pushed investors to vote in favor of the liquidation. Most of the underlying bonds are mezzanine classes from the 2005-07 vintage, with one AJ position from the GCCFC 2007GG9 deal. While more seasoned mezzanines have attracted some interest in recent weeks, trading volumes in the 06-07 mezzanine space have remained low as investors stay near the top of the capital structure.

Dealer CMBX long protection positions continue to fall


As expected, weekly DTCC data showed another drop in dealer long protection positions on the CMBX.AAA. The net long dollar notional on the AAA declined to $2.2bn in the week ending September 2 from $2.6bn a week earlier. We track this data on a weekly basis in the synthetic supplement section of the CMBS Strategy Weekly. We expect dealer long protection positions in the CMBX indices to drop further when the conduit deals in the pipeline close and warehouse hedges are taken off.

Beacon Seattle 1616 North Fort Myer Drive sold


This is a reprint of an Instant Insight published on September 6, 2011 Based on a report published in the Washington Business Journal, TIAA-CREF has acquired the 1616 N. Fort Myer Drive property from Beacon Capital Partners for $145.5mn. This is the latest in a series of property sales from the modified Beacon-Seattle portfolio securitized in six CMBS conduit deals and one CRE CDO (see CMBS Strategy Weekly, July 8, 2011, for the pari passu structure and details about the modification). Previously, five properties had been released from the portfolio, with the latest Liberty Place release reported in the July remittance (see Liberty Place released, July 11, 2011). As discussed in our July 8 publication, the release price from a property sale is not the same as the sale price; as such, the securitized trusts could see a lower amount flowing in as deal cash flows. For previous property sales from the Beacon Seattle portfolio, release prices have been 31-75% of the sale price. As a result, we expect some interest shortfall reimbursements and principal paydowns to hit the deals in the coming remittance periods.

9 September 2011

Barclays Capital | Securitized Products Weekly

According to the modification template, excess proceeds from the sale of properties will be used to build up the master reserve account until it reaches a cap of $100mn. Once this target has been reached, the excess proceeds will go to hyper-amortize the principal balance of the loan. After the Liberty Place release, the reserve account stood at almost $92mn. With the sale and potential release of the 1616 N. Fort Myer Drive property, the reserve cap is likely to be reached and the excess proceeds could be used to pay down the loan. Figure 5: 2007 LCF CMBS spreads
bp 540 490 440 390 340 290 240 190 20 0 -20 Daily spread chg (rhs) Level bp 100 80 60 40

-40 140 Dec-09 Feb-10 Apr-10 Jun-10 Aug-10 Oct-10 Dec-10 Feb-11 Apr-11 Jun-11 Aug-11
Source: Barclays Capital

Consumer ABS
Bond selection rises in importance in prime retail auto loan ABS
The auto ABS new issue machine continues to chug along, with auto-related ABS primary market transaction volume totalling $42.4bn year-to-date. Of this, $30.0bn has been backed by retail auto loans, 66% of which were prime quality in nature. Recent vintage transactions (eg, 2009-2011) have had credit enhancement requirements reduced since the depths of the credit crisis. Subsequently, these transactions have been posting improved collateral performance trends, most notably in the area of cumulative net losses. In Relative value among prime retail auto loan ABS on page 27, we assess the collateral performance and credit enhancement of the largest prime retail auto loan ABS issuers. We identify relative value within the prime retail auto ABS sector among issuers and across vintages.

Market action
Primary market activity bounced back this week with three new auto transactions hitting the market as investors returned to their desks after the Labor Day holiday. No credit card or student loan transactions were marketed this week. The auto transactions from ALLYA, AMCAR, and SDART were priced on Thursday. First up, AmeriCredit priced its fourth transaction of the year, AMCAR 2011-4, totalling $900bn in senior and subordinate offered notes. Shortly after, Santander priced its third retail auto transaction, the $900bn SDART 2011-3 senior/subordinate offering. Rounding out the week, Ally priced a $1.35bn transaction, ALLYA 2011-4, of which only the senior classes totalling $1.26bn were offered. Because of the lack of issuance (the last consumer ABS transaction priced in mid-August) all transactions were greeted with strong demand for the senior classes, as well as healthy investor interest in the subordinates.

9 September 2011

10

Barclays Capital | Securitized Products Weekly

Secondary market activity was somewhat muted in a continuation of the summer trend. Despite most investors returning to their seats after the Labor Day weekend, trading volumes were relatively slow as investors likely concentrated more heavily on the new issue calendar. FFELP secondary spreads continued to grind tighter, coming in about 5bp in the 3y, 5y, and 7y average life tenors. Secondary spreads of prime auto and fixed and floating rate credit card transactions remained unchanged w/w.

Recommendations
We maintain our overweight on FFELP-backed student loan ABS because we believe the tightening trend has further to run. S&P continues to review transactions directly linked to the rating of the US government. However, we believe investors are rightly focusing on the actual credit quality of the transactions, as opposed to S&Ps opinion (ie, rating). We also remain overweight on retail auto and other auto-related sectors, including subordinates, dealer floorplan and rental fleet securitizations. Despite continued supply in the sector (really, the only game in town for significant volume of primary consumer ABS issuance), as discussed in Relative value among prime retail auto loan ABS on page 27, we believe credit quality has improved to levels that, at current spreads, represent attractive value relative to other short-term consumer ABS. However, we are neutral on bank credit card ABS, given tight spread levels, and our belief that further upside potential is more limited in this asset class than others.

Figure 6: Retail auto ABS spreads (bp)


600 500 400 300 200 100 0 Sep-09 3yr AAA 3yr A 3yr BBB

Figure 7: Retail auto ABS spreads (bp), 12m detail


200 180 160 140 120 100 80 60 40 20 0 Sep-10 3yr AAA 3yr A 3yr BBB

Mar-10

Sep-10

Mar-11

Sep-11

Dec-10

Mar-11

Jun-11

Sep-11

Source: Barclays Capital

Source: Barclays Capital

9 September 2011

11

Barclays Capital | Securitized Products Weekly

Figure 8: Credit card ABS spreads (bp)


450 400 350 300 250 200 150 100 50 0 Sep-09 Mar-10 Sep-10 Mar-11 Sep-11 3yr AAA 3yr A 3yr BBB

Figure 9: Credit card ABS spreads (bp), 12m detail


200 180 160 140 120 100 80 60 40 20 0 Sep-10 Dec-10 Mar-11 Jun-11 Sep-11 3yr AAA 3yr A 3yr BBB

Source: Barclays Capital

Source: Barclays Capital

Figure 10: FFELP student loan ABS spreads (bp)


120 100 80 60 40 20 0 Sep-09 3yr AAA 5yr AAA 7yr AAA

Figure 11: FFELP student loan ABS spreads (bp), 12m detail
120 100 80 60 40 20 0 Sep-10 3yr AAA 5yr AAA 7yr AAA

Mar-10

Sep-10

Mar-11

Sep-11

Dec-10

Mar-11

Jun-11

Sep-11

Source: Barclays Capital

Source: Barclays Capital

9 September 2011

12

Barclays Capital | Securitized Products Weekly

PREPAYMENT COMMENTARY

Re-assessing refinancing risks


Derek Chen +1 212 412 2857 derek.chen@barcap.com Wei-Ang Lee +1 212 412 5356 weiang.lee@barcap.com Nicholas Strand +1 212 412 2057 nicholas.strand@barcap.com

This is a reprint of the Instant Insight published on September 7, 2011. FNCL speeds came in slightly below our expectation as the sharp rally in rates since the beginning of August did not seem to have filtered into prepayments at all. Total paydowns increased 13% m/m, to $29.4bn (16.8 CPR), almost entirely because of the three-day increase in day count. Aggregate prepayments are slightly below the August 10 report, when driving rates were more than 30bp higher. As lower rates continue to filter into prepayments, speeds should inch higher over the next few months. However, we believe 2010 highs will be reached only if rates rally below 4.1% (another 25bp decline from today's levels and 35bp lower than last year's low). The 2009/10 4.5s increased by 2.0/1.8 CPR, to 15.4/10 CPR (Figure 1), while the 2008/09/10 5s picked up by 2.4/2.1/0.9 CPR, reaching 25.7/16.9/10.9 CPR. The 2008/09 speed differential increased, to 6.7 CPR for 4.5s (from 4.7 CPR last month) and 8.8 CPR for 5s (from 8.3 CPR last month). Newly HARP-able collateral (ie, FNCL Q2 09 origination) did not show a particularly strong pick-up (Figure 3). Higher coupons rose by 2-3 CPR for 5.5/6s because of the jump in day-count. However, on a day-count neutral basis, these coupons remain the most rate-insensitive in the stack. Chase-serviced pools, which have had elevated FNCL speeds since March, experienced a stronger pickup than other servicers (Figure 2). At an estimated zero-point rate of 4.33%, we expect the FNCL 2009 4.5/5s to peak at 26.3/23.1 CPR and the 2010 4.5/5s to reach 19.5/16.2 CPR. After briefly touching 3900, the MBA refinance index has lost significant ground in recent weeks, despite the no-point rate remaining below 4.4%. This highlights the difficulty of sustaining refinancing momentum and should lead to lower total pay-downs versus the 2010 high. However, certain cohorts, specifically new vintage 3.5-4.5s, could reach or exceed last year's peak because they can take advantage of the new low in mortgage rates. Bank of America's exit from the correspondent business should slow conventional speeds modestly. More important, it signals continued risk aversion and decreases the likelihood of a sudden embracing of HARP by BoA, something Chase did early this year. President Barack Obama is unlikely to announce a sweeping refinancing program in his speech on Thursday, but we do believe policymakers are adopting a more serious stance towards resolving roadblocks to refinancing. However, any such plan must sufficiently address put-back risk to be effective an enormous and complicated task with no clear solution, in our view. GNMA seems less exposed to policy risks, given FHA's struggle to maintain selfsustainability. Any threat to that is, in our view, politically unpalatable and would only undermine FHA's ability to continue offering affordable loans to underserved borrowers. Furthermore, a possible exit of overseas investors, the largest source of demand for GNs, will likely make policymakers think twice before implementing any major changes.

9 September 2011

13

Barclays Capital | Securitized Products Weekly

Prepayment outlook: Aint no rate low enough


Recent data are consistent with our view that refinancing risk remains very much contained, despite the persistent rally in rates. As the zero-point mortgage rate dipped to an all-time low, the MBA Refinancing Index initially picked up, briefly reaching 3915. However, as rates stagnated around 4.3-4.4%, the refi index quickly lost its momentum, tumbling 19% over the following three weeks. This resembles last year's experience, when mortgage rates declined by 21bp between August 12 and October 28 while the refi index fell (Figure 4). The pattern highlights today's extremely muted refinancing responsiveness, which stems from exceptionally tight underwriting standards, onerous refinancing frictions and a diminished media effect. When rates drop to new lows, a limited number of borrowers who are both interested and able to refinance emerges. But as the fresh pool of borrowers depletes, refinancing activity diminishes quickly.

Figure 1: Benchmark conventional speeds


FNCL CPR Coupon Vintage 3.5 4 4.5 2010 2010 2009 2010 2009 2008 2003 2010 2009 2008 2007 2006 2005 2004 2003 2009 2008 2007 2006 2005 2004 2003 2002 2008 2007 2006 2005 2004 2003 2002 2008 2007 2006 2002 2008 2007 2006 Diff 1.5 1.7 1.6 1.8 2.0 4.0 3.5 0.9 2.1 2.4 1.6 4.8 2.4 2.6 2.4 2.7 2.3 2.3 3.1 2.7 1.5 2.9 3.7 2.1 1.6 2.8 1.4 2.2 3.0 2.3 0.2 2.2 1.1 2.8 1.9 1.4 0.4 Prev 2.7 4.8 8.7 8.2 13.4 18.1 16.6 10.0 14.9 23.2 20.8 19.2 19.4 18.9 20.2 13.9 23.5 22.8 21.9 17.8 18.6 20.6 22.1 22.8 22.2 21.2 17.1 16.7 16.8 20.6 21.5 20.4 20.5 18.1 21.3 20.8 20.3 Curr 4.3 6.6 10.3 10.0 15.4 22.1 20.0 10.9 16.9 25.7 22.3 24.0 21.8 21.5 22.7 16.6 25.8 25.1 24.9 20.4 20.1 23.5 25.9 24.9 23.8 24.0 18.5 18.9 19.9 22.9 21.7 22.6 21.7 20.9 23.1 22.2 20.6 Proj 4.0 7.6 11.9 12.3 18.4 26.2 21.0 12.5 18.3 28.7 24.4 22.5 22.4 22.6 24.2 15.5 26.1 25.0 24.1 19.2 20.3 22.7 24.8 24.5 23.7 22.6 18.0 17.8 18.4 22.2 22.5 21.3 21.4 19.3 22.0 21.1 20.7 Diff 0.7 1.5 1.3 1.9 2.1 -2.3 3.0 1.6 1.7 3.2 1.9 3.4 2.5 2.2 1.6 2.5 2.4 2.2 2.2 3.0 2.7 3.5 5.2 1.7 1.7 3.1 2.5 2.0 3.1 1.7 2.4 1.1 1.0 3.2 -3.4 -0.9 -4.1 Prev 2.9 4.9 8.8 9.3 12.3 22.8 17.0 10.2 14.7 23.3 18.3 19.1 19.3 20.1 19.9 14.9 22.6 21.2 21.4 18.0 19.1 18.6 18.4 22.4 20.2 20.7 16.1 17.9 15.5 17.2 20.1 20.2 19.3 14.6 24.0 23.0 19.9 Gold CPR Curr 3.6 6.5 10.1 11.2 14.4 20.5 19.9 11.8 16.3 26.5 20.1 22.5 21.8 22.3 21.5 17.3 25.0 23.4 23.6 21.0 21.8 22.1 23.6 24.0 21.9 23.9 18.6 19.9 18.6 18.9 22.6 21.2 20.3 17.8 20.6 22.1 15.8 Gold-FNCL -0.7 -0.1 -0.1 1.2 -1.0 -1.6 -0.1 0.9 -0.6 0.9 -2.2 -1.6 0.0 0.8 -1.2 0.8 -0.8 -1.7 -1.3 0.5 1.7 -1.4 -2.2 -0.9 -1.9 -0.2 0.2 1.0 -1.3 -4.0 0.9 -1.3 -1.3 -3.1 -2.5 -0.1 -4.8

5.5

6.5

Source: Fannie Mae, Freddie Mac, Barclays Capital

9 September 2011

14

Barclays Capital | Securitized Products Weekly

The primary-secondary spread tells the same story. As discussed before, in today's environment, this spread has become a gauge for the level of refinancing activity, rather than a driver of it, given the capacity constraints faced by originators. Figure 5 shows that this spread has tracked the refi index quite closely, widening when refinancing activity picks up and narrowing when it subsides. The fact that the spread has narrowed substantially over recent weeks suggests that refinancing applications have tapered off. The past couple of years' experience and the recent behavior of the refi index and primarysecondary spread suggest the following: Mortgage rates have to keep rallying substantially week after week to keep the refi index above 4000. Since the refi index and primary-secondary spread are much lower than last year's levels, total pay-downs should be materially lower than last year's peak. Given these observations, we have updated our peak 1m CPR forecasts for major FNCL cohorts under various mortgage rate scenarios (Figure 6). Specifically, for an unchanged no-point mortgage rate of 4.33, we expect the refi index to sustain a level close to 3900, similar to the September 2010 prepayment report but much lower than the last year's peak (December 2010 report).

Figure 2: FNCL speeds by issuer


Chase BoA Citi Wells Other Cpn Orig Curr CPR M/m chg Curr CPR M/m chg Curr CPR M/m chg Curr CPR M/m chg Curr CPR M/m chg 5 2010 11.7 1.7 7.7 0.3 11.3 4.8 11.8 0.6 10.4 0.8 2009 19.3 2.8 15.2 1.3 18.7 3.0 20.5 0.9 15.7 2.3 2008 42.9 5.9 20.3 1.5 25.1 3.6 29.2 -0.2 24.0 2.5 2007 33.6 0.9 20.5 2.7 23.1 1.8 33.5 -2.9 20.4 1.3 2006 42.2 18.7 24.8 5.4 25.6 3.7 32.0 9.4 21.1 3.3 2005 29.4 -0.2 19.5 2.1 22.7 4.4 29.2 3.8 21.4 2.1 2004 32.6 8.6 18.9 3.0 20.7 2.9 27.0 -3.6 22.6 2.5 2003 30.7 2.9 18.0 1.8 19.5 0.0 24.9 2.2 23.3 2.7 5.5 2009 25.6 5.3 12.0 1.3 21.0 3.3 17.9 3.6 16.0 2.4 2008 40.7 5.4 20.9 0.6 28.0 4.9 31.1 1.4 23.7 2.5 2007 36.9 2.8 22.9 2.5 25.2 2.4 29.2 3.0 23.7 1.8 2006 35.7 7.1 23.7 3.3 27.3 3.6 29.2 2.3 23.7 2.9 2005 32.2 4.7 18.2 1.3 20.2 2.9 25.7 3.4 20.7 3.0 2004 33.5 4.7 17.4 0.4 18.7 3.1 29.6 5.8 21.2 1.8 2003 33.1 6.0 19.3 1.8 21.9 6.4 23.2 4.9 24.5 2.9 2002 33.0 5.0 23.1 4.3 20.4 4.0 0.2 0.2 26.3 3.5 6 2008 35.1 -0.2 20.9 0.6 25.1 4.0 29.4 1.6 23.9 3.2 2007 35.6 1.4 22.0 0.6 22.8 1.1 29.3 2.7 22.7 2.3 2006 37.6 6.7 21.7 0.6 25.1 3.4 30.3 2.0 23.5 3.4 2005 29.7 2.1 16.7 0.7 20.5 4.2 15.6 -2.9 18.5 1.5 2004 35.7 7.3 17.8 1.9 16.5 -1.4 18.4 5.5 19.3 2.4 2003 28.2 7.5 16.2 1.3 14.8 -1.5 11.8 -2.2 21.5 3.8 2002 32.6 2.9 19.7 2.8 21.1 4.9 6.0 -32.0 23.3 2.4 6.5 2008 33.5 3.0 18.6 -0.3 21.0 0.0 24.7 -0.5 20.8 0.3 2007 35.9 3.9 20.6 2.1 25.0 1.1 33.0 11.8 21.0 1.4 2006 33.8 -1.0 19.4 -0.5 23.0 3.4 25.5 -6.7 21.5 2.0 2002 23.2 -1.4 19.3 3.8 14.4 -5.7 0.5 0.1 21.7 3.4
Source: Fannie Mae, Barclays Capital

9 September 2011

15

Barclays Capital | Securitized Products Weekly

Figure 3: The HARP effect on 4.5/5s


6 5 M/m chg in CPR 4 3 2 1 0 -1 Mar-09 Nov-09 Jan-09 Feb-09 Apr-09 May-09 Aug-09 Dec-09 Sep-09 Oct-09 Jun-09 Jul-09 4.5 5

Figure 4: MBA refi index versus the NP mortgage rate


7000 6000 5000 4000 3000 2000 1000 0 Jan-09 May-09 Sep-09 Jan-10 May-10 Sep-10 MBA refi NP 30yr Rate (Inverted) 4.00 4.25 4.50 4.75 5.00 5.25 5.50 5.75

Origination month
Source: Fannie Mae, Barclays Capital

Source: MBA, Freddie Mac, Barclays Capital

That said, given today's rate (4.33%) 4.5s, 4s and 3.5s should all prepay close to or faster than last year's peak given their pristine credit, large loan size, minimum burnout and a mortgage rate below last year's low. Combined with a lower expected overall prepayment volume, this means that higher coupons should be much slower than late last year.

Figure 5: MBA refinance index versus the primary-secondary spread


6000 5000 4000 3000 2000 1000 0 Jan-10 140 130 120 110 100 90 80 70 60 Apr-10 Jul-10 Oct-10 MBA refi
Source: MBA, Freddie Mac, Barclays Capital

Jan-11 Prim-sec spread

Apr-11

Jul-11

50 Oct-11

9 September 2011

16

Barclays Capital | Securitized Products Weekly

Figure 6: Projected peak CPR for major FNCL cohorts by mortgage rate scenario
Actual CPR by report date Coupon 4 4 4.5 4.5 4.5 4.5 5 5 5 5 5 5 5 5 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.5 6.0 6 6 6 6 6 6 6.5 6.5 6.5 Vintage WAC Bal ($bn) 2010 4.50 121 2009 4.58 102 2010 4.94 122 2009 4.94 226 2008 5.18 6 2003 5.06 16 2010 5.35 58 2009 5.43 57 2008 5.65 40 2007 5.75 9 2006 5.77 6 2005 5.63 54 2004 5.53 29 2003 5.49 69 2009 5.95 7 2008 6.03 60 2007 6.13 55 2006 6.15 25 2005 5.98 49 2004 5.93 38 2003 5.93 67 2002 6.00 14 2008 6.52 30 2007 6.56 65 2006 6.55 48 2005 6.49 12 2004 6.42 15 2003 6.47 12 2002 6.50 13 2008 6.98 8 2007 7.05 17 2006 7.01 19 No-point m ortgage rate Day count MBA Refi Index Sep 2010 Dec 2010 Aug 2011 4.4 6.9 4.8 7.3 14.4 8.7 11.8 19.2 8.2 18.8 26.6 13.4 36.2 46.4 18.1 23.1 33.5 16.6 11.2 16.6 10.0 26.3 28.6 14.9 39.2 43.7 23.2 29.3 35.9 20.8 27.8 33.6 19.2 27.4 32.9 19.4 27.3 34.7 18.9 28.0 35.8 20.2 21.3 21.7 13.9 34.1 38.1 23.5 31.6 34.4 22.8 29.7 33.5 21.9 22.7 26.1 17.8 23.2 27.7 18.6 25.9 31.6 20.6 30.7 36.5 22.2 30.1 31.3 22.8 28.0 28.4 22.2 26.6 27.7 21.3 19.5 20.6 17.1 19.8 22.2 16.7 19.4 21.4 16.9 23.6 26.9 20.7 27.3 25.6 21.7 26.3 24.4 20.4 23.6 23.2 20.6 4.75 4.45 4.68 22 20 20 3964 4590 2619 5.00 3.7 6.0 5.1 8.6 14.3 11.4 7.0 10.7 19.5 17.8 17.0 15.5 15.1 15.7 12.0 22.7 22.7 23.0 16.7 16.7 18.2 19.7 25.7 24.3 23.7 18.2 17.6 18.1 19.9 23.7 21.9 21.6 5.00 21 *2200 Projected Peak CPR by mortgage rate 4.50 9.9 12.3 14.1 21.7 31.8 22.2 13.1 20.1 32.3 26.5 25.0 23.5 24.4 25.8 17.6 30.1 28.3 27.8 21.4 21.9 24.8 28.3 27.5 26.1 25.0 19.5 19.9 21.1 24.0 24.2 23.6 22.7 4.50 21 *3500 4.33 12.8 17.3 19.5 26.3 39.5 27.2 16.2 23.1 36.4 29.3 26.9 26.1 27.4 29.1 19.1 31.0 29.0 28.4 21.4 22.5 25.6 29.3 27.8 26.3 25.1 19.5 19.9 21.3 24.2 24.5 23.6 22.7 4.33 21 *3900 4.20 16.4 22.1 23.0 30.7 45.8 31.7 18.8 26.6 40.4 32.5 29.8 29.1 30.9 32.8 20.8 33.5 31.1 30.2 22.9 24.5 27.9 32.2 28.8 27.3 25.9 20.2 20.6 22.5 25.8 25.2 23.6 22.9 4.20 21 *4200 4.00 24.1 31.8 29.2 38.4 51.9 39.9 23.3 32.8 45.5 38.3 34.8 34.4 37.0 39.5 23.6 37.8 34.4 33.2 25.5 27.6 31.7 37.1 30.2 28.8 27.0 21.2 21.9 24.4 28.3 26.0 23.8 23.5 4.00 21 *5500

Note: * estimated. Source: Fannie Mae, Barclays Capital

BoAs exit from the correspondent channel should depress speeds further
Early this year, an abrupt jump in the speeds of Chase-serviced FNMA pools alarmed investors with the possibility of other big servicers, such as Bank of America, suddenly turning more aggressive toward HARP refinances and pushing up speeds. We argued (Securitized Products Weekly, June 17, 2011) that such a development seemed rather unlikely because: The poorer quality of the loans that BoA inherited from CountryWide makes them an unfit target for the HARP program, given the new rep & warrant risks associated with any refinancing transaction. The bulk settlement between BoA and Freddie Mac early this year has largely relieved the lender of the put-back risks on its legacy Countrywide loans. Because refinancing these loans comes with new put-back risks, BoA has no incentive to do so.

9 September 2011

17

Barclays Capital | Securitized Products Weekly

The recent announcement by BoA to exit the correspondent channel has made a sudden jump in its speeds even more unlikely. Rather, it should act to depress BoA's speeds further. In short, correspondents are conduits that underwrite and fund loans only to sell them to large banks such as BoA, who then rep & warrant the loans and deliver them to the GSEs. Typically, there is a recourse clause such that if a loan is put back by the GSEs, the correspondent has to reimburse the bank for the loss. However, if the correspondent is out of business at the time of the put-back, the bank has to take the loss. Given the record level of put-backs by the GSEs, the lack of control over how correspondents underwrite loans, and the large of number of correspondents going out of business in recent years, we think BoA is exiting this channel to protect itself from further credit losses. In theory, if a correspondent cannot sell loans to BoA, it can still originate and sell them to other banks such as Wells Fargo or Chase. However, selling to a different bank would likely require complying with a new set of underwriting standards, which would probably disqualify some existing loans from refinancing. In addition, a drop in total production of correspondent loans should also reduce the refinance-ability of homeowners. Figure 7 illustrates the possible effect on FNCL speeds. For recent vintages, roughly 25% of the UPB is serviced by BoA, of which 23-48% is from the correspondent channel. All told, correspondent loans serviced by BoA account for 6-10% of recent FNCL vintages. If we assume that BoA's correspondent loans prepay similarly to average BoA pools but will slow 50% in the future, total FNCL speeds should drop 0.5-1.5 CPR. More important, the exit and other recent developments related to BoA signal that the lender is focusing on strengthening its defence against GSE put-backs, rather than increasing production or market share. With that mindset, it is unlikely to become more aggressive with HARP refinances because that would bring more put-back risks. Figure 7: CPR effect from exit of BoA correspondent business
Cohort ($bn) 120 248 60 64 50 75 37 BoA Corr ($bn) 12 24 4 6 4 7 3 Cohort CPR impact 1.1 1.2 0.4 1.3 1.5 1.6 1.0

Cpn 4.5 5

Vintage 2010 2009 2010 2009 2008

BoA ($bn) 25 58 16 14 15 21 8

Pct BoA Corr Cohort CPR 10% 10% 6% 10% 8% 9% 8% 19 27 17 29 44 38 31

Cohort CRR 19.2 26.2 16.6 27.5 41.6 34.4 24.5

BoA CPR 19.8 23.3 12.7 25.1 35.4 32.1 27.8

5.5 6

2008 2008

Note: Based on the December 2010 report. Source: Fannie Mae, Barclays Capital

For now, a government-induced refinancing wave still seems remote


Although refinancing activity has been muted, the MBS market is worried that a new government program would soon push up speeds significantly and wreak havoc in a market where almost every pool is priced above $102. It has been reported that a program could be announced as early as tomorrow, when President Obama addresses Congress. In Refi Realities, August 26, 2011, we discussed in detail the various options that might be considered by policymakers. In short: Although such drastic measures as removing the rep & warrant risk for lenders and offering a universal 4% mortgage rate to all homeowners would be the most effective in helping struggling borrowers, the complications and implied taxpayer burden make them very unlikely.
9 September 2011 18

Barclays Capital | Securitized Products Weekly

A more likely measure, in our view, is fine-tuning the existing HARP program by considering such options as increasing the LTV limit and removing the upfront delivery fees. However, this represents only a moderate increase in refinance-ability and would not pose a substantial increase in prepayment risk, in our view. Given the recent cheapening in MBS, these changes have already been priced into the market. Another possibility is to relax or eliminate the HARP origination date requirement, which would push up the speeds of 2009 and later vintages significantly. However, we think the likelihood is lower than what is being priced into the market: 1. Since such a change would benefit only loans originated after May 2009, it would not help struggling borrowers those who are on the brink of default or owe more than their homes are worth to refinance, a central concern that has been voiced repeatedly by policymakers and the media. 2. There is a reason why HARP had an origination date requirement to begin with. By granting partial mortgage insurance (MI) waivers, HARP went against the GSE charter. FHFA bypassed the conflict by framing HARP refinances as loss mitigations: pre-emptive measures to help loans on the verge of default. While it is easier to make a case that loans made before 2009 are collectively under stress and, therefore, that mass loss mitigations are appropriate, it is not as clear that such is the case for newer production. All in all, we do not expect a sweeping refinancing program to be included in President Obama's speech tomorrow. That said, we believe that policymakers are aware of how putback risks are preventing homeowners from refinancing and are looking at ways to alleviate this at a minimum cost. In this regard, we see two possibilities: The Treasury brokers a deal with the MI companies and lenders such that they pay a certain amount to the GSEs to cover all existing and potential loss claims. After that, loans refinanced through the HARP program are no longer required to carry MI and are exempt from put-back risks. The settlement can be paid in many instalments to minimize the immediate hit to banks and MI companies, and one could argue that such a deal does not change the eventual total loss to banks and MI companies but merely puts a closure to it. After that, the interests of all parties should be realigned: the GSEs and banks are both encouraged to refinance, while MI companies will not stand in the way. Although this is conceptually feasible, it would face tremendous hurdles. Given the significant difference in loan quality across lenders and MI companies, and intertwined exposures among lenders, MI companies and the GSEs, negotiations could take years. As discussed earlier, a major impediment is that every refinancing results in a new loan that comes with fresh reps & warranties. Were it to default shortly thereafter, the GSEs and MI companies would have an easier time faulting the lender for underwriting defects. Hence, it is in the lender's interest not to refinance the loan: if it defaults, it will have made several years' timely payments, bolstering the case that the default was not caused by underwriting defects. However, there is no reason for a refinance transaction to result in new rep & warrant requirements since no new risk has been generated. Therefore, the GSEs could announce that from now on, if a loan is refinanced, the lender is subject only to the rep & warrant associated with the original loan but does not take on any new put-back risk. For example, if a loan is refinanced at 36 WALA but defaults seven months later, the GSE would decide whether to issue a repurchase request based on the underwriting documents
9 September 2011 19

Barclays Capital | Securitized Products Weekly

of the original loan, and whether to take into consideration that the loan had made 43 payments before the default. This way, the lender will have more incentive to refinance higher-risk loans to reduce the probability of default. The problem is that MI companies will have to grant similar relief on put-backs for the program to deliver the full effect. In summary, although we believe that the chance of a government-induced refinancing wave remains very low, the fear will likely persist as long as MBS dollar prices are high, prepayment speeds slow, and economic conditions weak. Therefore, investors should keep an eye on possible developments along the lines discussed above.

Why GNMA MBS are less prone to policy risks


Arguably, the GNMA/FHA program has done a good job of being countercyclical and providing critical support to the housing market during times of distress. Being pretty much the only line of credit available to homeowners with a less than perfect credit profile, it accounts for about half of all purchase loans made today. Throughout its 77year history, it has never asked for money from taxpayers (in stark contrast to the GSEs), and maintaining this self-sustainability is, in our view, critical for it to continue providing support for the housing market. However, despite the recent tightening in underwriting standards and increases in the insurance premiums, FHA is struggling to do so. Any change (such as rolling back the insurance premium or relaxing underwriting) that threats the self-sustainability this would, in our view, be politically unpalatable and only undermine FHA's ability to continue offering affordable loans to underserved borrowers. FHA is already shouldering more than its fair share of the housing market, and the current loan limits are much too big for its intended goal of supporting lower-income homeowners. It seems unlikely that the program will be expanded. FHA/VA still has arguably some of the most streamlined refinancing programs, something that is missing in conventional space. The possible exit of overseas investors, the largest source of demand for GNMA MBS, will likely make the administration think twice before implementing any major changes. Thus, we believe that any administration initiative is more likely to be channelled through the GSEs, which are already on public support, rather than the FHA.

9 September 2011

20

Barclays Capital | Securitized Products Weekly

Figure 8: FNMA 30y short-term Barclays Capital forecasts


Cpn 3.5 4 4.5 Vintage WAC 2010 4.14 2010 4.50 2009 4.58 2010 4.94 2009 4.94 2008 5.17 2003 5.07 2010 5.36 2009 5.42 2008 5.65 2007 5.75 2006 5.77 2005 5.63 2004 5.53 2003 5.49 2009 5.94 2008 6.03 2007 6.13 2006 6.15 2005 5.98 2004 5.93 2003 5.93 2002 6.00 2008 6.52 2007 6.56 2006 6.55 2005 6.49 2004 6.43 2003 6.47 2002 6.50 2008 6.98 2007 7.05 2006 7.01 2002 6.96 2008 7.49 2007 7.64 2006 7.61 No-point m ortgage rate Day count Actual FNM A Bal ($bn) CPR 3M 22.5 3.3 119.8 5.3 101.0 8.7 120.7 8.4 222.9 13.4 6.0 19.2 16.6 17.4 57.0 9.9 55.8 15.4 39.3 24.1 9.2 20.7 5.5 21.0 53.1 19.9 28.4 20.0 67.9 21.2 7.3 15.0 58.7 24.4 53.2 23.8 24.1 23.3 47.6 18.9 37.5 19.1 65.0 21.7 13.9 23.6 29.5 24.0 63.3 23.2 47.2 23.0 12.0 17.9 14.3 17.9 11.7 18.9 12.5 21.6 7.9 22.2 16.6 21.6 18.7 21.2 6.0 19.5 2.7 22.5 4.8 21.8 3.3 20.6 4.75 22 Sep-11 4.3 6.6 10.2 10.0 15.4 22.2 20.0 10.9 16.9 25.7 22.3 24.1 21.8 21.5 22.7 16.6 25.8 25.1 24.9 20.4 20.1 23.5 25.9 25.0 23.8 24.0 18.5 18.9 19.9 23.0 21.8 22.6 21.7 21.2 23.3 22.2 20.7 4.71 23 Projected (report month) Oct-11 Nov-11 Dec-11 5.1 5.6 6.1 9.4 10.9 12.1 13.0 14.7 16.0 14.3 16.7 18.5 20.1 22.8 24.8 30.2 34.4 37.5 22.5 24.1 25.4 13.0 14.2 15.3 19.1 20.6 21.8 30.0 32.5 34.5 24.7 26.3 27.6 24.3 24.8 25.3 22.8 23.7 24.5 23.3 24.6 25.7 24.7 26.1 27.3 16.3 16.8 17.9 25.9 27.0 29.3 24.9 25.7 27.4 24.5 25.2 26.8 19.4 19.5 20.0 19.5 19.9 21.0 22.6 22.8 23.9 25.3 25.9 27.6 24.4 24.9 26.2 23.2 23.7 24.9 22.9 23.0 23.6 17.7 17.9 18.3 17.9 18.0 18.5 18.9 19.1 19.9 21.8 21.8 22.7 21.3 21.9 23.1 21.7 21.8 22.3 20.7 20.8 21.3 20.0 20.0 20.7 22.6 22.8 23.2 21.7 21.9 22.2 19.9 20.1 20.5 4.52 4.33 4.30 21 20 20

5.5

6.0

6.5

7.0

Note: As of September 9, 2011, open mortgage rate at 4.30% (10y at 2.00%). Source: Barclays Capital

9 September 2011

21

Barclays Capital | Securitized Products Weekly

Figure 9: FNMA 15y short-term Barclays Capital forecasts


Coupon 3 3.5 4 Vintage 2010 2010 2010 2009 2008 2003 2010 2009 2008 2007 2006 2005 2004 2003 2009 2008 2007 2006 2005 2004 2003 2002 2008 2007 2006 2005 2008 2007 2006 M ortgage Rate Day-Count WAC 3.60 3.93 4.41 4.48 4.58 4.55 4.84 4.89 5.05 5.25 5.19 5.12 4.96 4.96 5.48 5.56 5.71 5.67 5.49 5.42 5.44 5.50 6.01 6.07 6.01 5.91 6.52 6.52 6.48 Actual FNM A Bal ($bn) CPR 3M 6.1 4.2 42.7 9.4 38.4 15.1 37.1 22.3 1.2 23.8 8.1 18.7 8.4 14.8 15.8 21.8 8.0 31.2 0.4 26.7 0.4 20.0 3.4 21.6 7.5 18.4 26.7 18.7 1.0 20.3 6.4 27.1 2.1 27.3 1.3 24.0 5.5 19.1 5.5 17.6 17.1 17.6 7.1 20.0 2.6 25.3 3.5 23.0 3.2 21.9 2.0 16.9 0.9 18.9 2.6 20.9 2.4 20.0 3.92 22 Sep-11 5.0 12.2 17.1 27.6 26.4 19.9 16.9 25.1 34.6 24.7 22.2 23.3 20.1 20.5 22.0 28.6 29.5 26.0 21.6 18.9 19.1 21.9 24.9 24.7 23.1 19.4 21.1 22.3 21.5 3.85 23 Oct-11 8.1 17.5 22.4 33.4 33.7 21.7 21.1 30.7 38.4 29.1 21.9 24.1 21.7 23.4 23.1 30.3 30.5 25.9 20.0 17.5 18.6 21.2 26.8 25.1 22.8 18.2 19.2 20.9 20.4 3.67 21 Projected (report month) Nov-11 8.9 18.8 23.5 34.6 35.3 22.0 22.0 31.8 38.9 30.0 21.5 24.0 21.9 23.9 23.1 30.3 30.3 25.5 19.3 16.9 18.2 20.7 27.0 24.9 22.5 17.6 18.4 20.2 19.9 3.50 20 D ec-11 10.0 20.7 25.6 37.0 38.2 22.9 23.7 34.1 40.9 31.9 21.8 24.7 22.8 25.2 23.9 31.5 31.3 26.0 19.2 16.8 18.3 20.9 28.1 25.5 22.8 17.6 18.2 20.2 19.9 3.48 20

4.5

5.5

Note: As of September 9, 2011, open mortgage rate at 3.48% (10y at 2.00%). Source: Barclays Capital

Figure 10: GNMA 30y short-term Barclays Capital forecasts


G NM AI Prepayments Actual Coupon 4 4.5 Vintage 2010 2009 2010 2009 2008 2003 5 2010 2009 2008 2007 2006 2005 2004 2003 5.5 2009 2008 2007 2006 2005 2004 2003 2002 6.0 2008 2007 2006 2005 2004 2003 2002 6.5 2008 2007 2006 2002 2001 Mortgage Rate Day-Count WAC 4.5 4.5 5.00 5.00 5.00 5.00 5.50 5.50 5.50 5.50 5.50 5.50 5.50 5.50 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.50 6.50 6.50 6.50 6.50 6.50 6.50 7.00 7.00 7.00 7.00 7.00 B al ($bn) 34.2 14.1 57.6 106.0 0.4 1.9 15.2 70.4 10.5 0.6 0.6 4.6 2.8 8.7 8.2 21.6 3.1 2.7 4.8 4.8 9.8 1.8 15.4 5.0 4.6 1.1 2.0 2.7 2.5 2.6 1.5 1.4 1.4 1.2 3-month CPR 4.4 6.8 7.5 8.5 10.9 12.5 10.4 11.6 16.4 17.1 17.7 15.7 14.4 14.1 16.3 18.4 21.1 19.4 17.0 15.4 15.1 17.2 19.6 20.3 20.7 14.0 14.1 12.8 15.2 20.5 18.8 19.4 13.9 13.7 4.75 22 Sep 3.8 6 7 7.8 8.3 11.6 9.1 10.8 16.4 19.2 14.8 14.8 13.1 13.5 15.0 17.0 19.6 17.4 16.4 14.5 15.1 16.3 17.9 19.5 20.4 12.5 13.5 12.1 14.9 19.0 16.9 20.0 13.6 12.5 4.71 23 Projected (report month) Oct 5.5 9.3 10.2 13.7 22.8 14.5 14.9 18.6 24.7 24.9 19.5 19.1 16.0 16.4 23.4 26.4 26.8 25.0 20.9 18.0 18.2 20.2 26.8 26.2 25.7 17.3 17.1 15.5 18.3 25.6 23.2 24.4 15.0 15.4 4.52 21 N ov 6.0 9.8 10.7 14.5 23.9 15.0 15.8 20.2 26.0 25.4 20.2 19.9 16.8 17.2 25.1 28.0 28.0 26.4 21.6 18.5 19.1 21.2 28.0 27.1 26.3 18.0 17.9 16.2 18.9 26.3 24.3 24.7 15.6 16.0 4.33 20 Dec 6.1 9.7 10.7 14.4 24.0 14.9 15.9 20.3 26.1 25.3 20.2 19.8 16.7 17.1 25.3 28.2 28.0 26.5 21.6 18.5 19.1 21.2 28.1 27.2 26.3 18.0 17.8 16.1 18.8 26.3 24.4 24.7 15.5 15.9 4.30 20 WAC 4.38 4.46 4.87 4.92 4.98 5.11 5.29 5.35 5.52 5.54 5.64 5.60 5.57 5.55 5.87 5.94 6.02 6.06 5.97 5.93 6.01 6.27 6.44 6.46 6.46 6.40 6.37 6.43 6.76 6.87 6.90 6.87 7.23 7.27 Cbal 34.1 2.5 80.6 46.0 0.3 0.5 33.3 60.7 2.7 0.5 0.8 4.9 2.4 3.9 6.4 18.8 3.5 2.9 5.7 5.4 4.3 0.7 10.0 7.5 4.3 1.4 2.5 1.4 1.0 4.9 2.5 1.2 0.4 0.7 Actual 3-month CPR 4 5.5 7.1 7.4 12.4 14.8 10.9 9.9 17.3 18.4 15.7 16.0 14.9 14.1 15.4 18.0 20.2 20.5 16.2 15.9 13.9 15.8 20.4 20.3 21.5 19.0 15.3 14.9 15.0 20.9 19.6 22.1 12.7 14.7 4.75 22 Sep 3.6 4.6 6.7 6.7 10 13.3 9.7 9.4 17.9 20.8 14.5 15.8 14.6 13.7 13.9 17.2 19.8 20.2 15.7 15.7 13.6 14.3 19.3 18.8 20.5 17.9 14.1 14.9 15.2 19.5 18.7 21.5 12.4 12.8 4.71 23 G NM AII Prepayments Projected (report month) Oct 5.1 8.3 8.6 11.9 16.5 16.9 13.8 16.0 24.0 27.6 20.4 21.4 17.1 16.9 20.7 24.7 26.3 24.2 20.8 17.3 16.0 19.7 26.1 27.9 25.0 20.6 17.5 16.5 16.2 25.2 26.0 24.5 13.4 15.6 4.52 21 Nov 5.6 8.8 9.0 12.6 17.3 17.5 14.5 17.3 25.2 28.1 21.2 22.3 17.8 17.7 22.1 26.2 27.3 25.5 21.6 17.9 16.7 20.6 27.3 28.8 25.6 21.5 18.3 17.2 16.7 25.9 27.1 24.9 13.9 16.2 4.33 20 Dec 5.6 8.6 8.9 12.5 17.3 17.4 14.6 17.3 25.3 28.1 21.1 22.2 17.8 17.6 22.2 26.3 27.4 25.6 21.6 17.8 16.6 20.6 27.4 28.9 25.5 21.5 18.2 17.2 16.6 25.9 27.2 24.9 13.8 16.1 4.30 20

Note: Incorporates BoA delinquency cleanup over three months. As of September 9, 2011, open mortgage rate at 4.30% (10y at 2.00%). Source: Barclays Capital

9 September 2011

22

Barclays Capital | Securitized Products Weekly

CMBS

Trust implications of super-high loss severities


Julia Tcherkassova +1 212 412 5977 julia.tcherkassova@barcap.com Keerthi Raghavan +1 212 412 7947 keerthi.raghavan@barcap.com

CMBS monthly liquidation volumes have stayed above the $1bn mark for most of 2011, a sharp step up from earlier years (Figure 1). Liquidation volume has been boosted by a number of note auctions in recent months, in which special servicers look to dispose of smaller loans backed by distressed properties, as an alternative to going through a potentially long and expensive foreclosure and REO process. At the same time delinquent larger loans in Tier 1 centers have a lower probability of liquidation; usually receiving some sort of workout, possibly through a modification or extension (Figure 2). 2 A combined effect of these two trends had been a pick-up in loss severities. Fixed costs arising out of legal, foreclosure, and other fees are a higher share of the liquidation proceeds on smaller loans, leading to correspondingly higher severities. Combined with the need to reimburse outstanding advances and ASERs at the time of loan liquidation, higher expenses result in some loans reporting severities greater than 100%. Such instances were relatively uncommon for much of 2007 to 2009, averaging about five cases per quarter. Since 2010, however, the number of loans reporting severities in excess of 100% has grown to 20 per quarter (we have been highlighting some of these instances in our monthly CMBS Credit Handbook). In some cases, these have had fairly significant effects on the deal cash flow dynamics resulting in interest shortfalls reaching up to the originally AAA-rated tranches, and/or a diversion of principal cash flows. Overall we identified 175 loans that were liquidated with severities exceeding 100% in the conduit universe since 2007. As Figure 3 shows, the number has grown in recent years with 120 of these liquidations taking place in 2010 and 2011. A bulk of these have been concentrated in loans that are $25mn or below; three of the largest loans reporting liquidation severities at 100% were the $48m Holiday Inn Portfolio in GSMS 2007-GG10, the $46m Eastland Mall in CASC 1998-D7, and the $31mn Connecticut Health in GMAC 1997-C1. In addition to the Holiday Inn portfolio, the latest August remittance saw two other such instances (Please see CMBS Credit Handbook for details). We expect liquidations to remain elevated in the coming months and as such, the number of such instances of loans taking losses north of 100% will likely increase.

Figure 1: Historical liquidations ($bn)


1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Jan-08 Jul-08 Jan-09 Jul-09 Jan-10 Jul-10 Jan-11 Jul-11
Source: Intex, Barclays Capital
2

Figure 2: Modified vs. liquidated (Share of specially serviced)


45% 40% 35% 30% 25% 20% 15% 10% 5% 0% <$15mm
Source: Intex, Barclays Capital

Modified

Liquidated

$15-$100mm

>$100mm

We discuss this in more detail in the CMBS Strategy Weekly, July 15, 2011

9 September 2011

23

Barclays Capital | Securitized Products Weekly

Figure 3: Liquidations with severities exceeding 100%


25 20 15 10 5 0 07Q1 07Q3 08Q1 08Q3 09Q2 09Q4 10Q2 10Q4 11Q2 Number of loans with sev>100%

Source: Intex, Barclays Capital

Components of severity
Before discussing implications on deal and bond level cash flows, we start our analysis with a more general question: why could severities on some loans be above 100%? In Figure 4, we show a typical breakdown of overall losses realized on a liquidated loan. When a loan is disposed of, liquidation fees (usually at about 100bp) are typically charged as a percentage of net proceeds. Other fees, including special and master servicing fees, are calculated on the outstanding balance of the loan. Deal remittances also report a rather non-descriptive other fees column, which includes a range of expenses from legal costs to broker fees and would typically include a large fixed component. Consequently, they form a higher share of overall expenses on smaller loans Once the fees have been reimbursed, liquidation proceeds are used to pay back the master servicer for any P&I advanced on the loan. After the servicer has been reimbursed, the trust receives any unpaid advances through reimbursements of ASERs and non-recoverable advances. Only then does the trust receive any principal recovery on the loan. Therefore, if liquidation proceeds do not cover the sum of expenses, servicer advances and ASER/nonrecoverable advances, severities typically will be reported at or above 100%.

9 September 2011

24

Barclays Capital | Securitized Products Weekly

Figure 4: Components of severity


1) Servicer/Trustee fees Liquidation Fee to SS on SS loan Workout Fee on Corrected Mtg. loan Master Servicing Fees (prior period) Special Servicing Fees Trustee Fees Other Fees 2) Advances P & I Advances (net advanced) Interest on Advances- P & I 3) Amounts held back for future payment Other Unpaid Fees and expenses Other amounts 4) Unadvanced interest Cumulative Aser Amount Deemed non-recoverable interest or Advances (prior shortfall) Deemed non-recoverable interest or Advances (paid from trust principal) Liquidation Expenses Net Proceeds
Source: Barclays Capital

1 through 4 Proceeds-Liquidation Expenses

Some examples
In some instances, liquidations with severities exceeding 100% might lead to interest shortfalls. This would typically happen if liquidation proceeds do not cover expenses and servicer advances. A recent example was the 17320 Gale Avenue loan in COMM 2006-C8. The $13.6mn loan had accrued $337k in P&I advances and $625k in ASERs after being in foreclosure since May 2009. When it was finally liquidated in August 2011, liquidation proceeds were reported as only $10k. Since the proceeds did not cover expenses and servicer advances, the servicer reimbursed a total of $339k from the general trust interest cash flows resulting in interest shortfalls to the trust reaching up to the originally AA rated tranche C. In most cases where liquidation proceeds are this low, the special servicer would have already deemed the loan as non-recoverable and stopped advancing any P&I prior to the liquidation event. Nearly 85% of all loans reporting severities exceeding 100% since 2008 had been categorized as non-recoverable by the special servicer at some point prior to their liquidation. The Gale Avenue loan was unusual in that sense, as it had not been classified as nonrecoverable and the servicer was still advancing $32k of interest every month. This resulted in a build-up of advances and an eventual large reimbursement from general interest proceeds, leading to interest shortfalls. In contrast to the above example, liquidations with severities exceeding 100% could also lead to interest shortfalls reimbursements in some instances. The $19m Boscovs Monroeville Mall loan, in BACM 2006-3 was also liquidated in August with a severity
9 September 2011 25

Barclays Capital | Securitized Products Weekly

reported in excess of 100%. The loan had been delinquent for nearly three years and had been deemed non-recoverable since December 2009. Subsequently, P&I advances made by the servicer were recovered from general interest proceeds in May and June of 2010 3. As a result, outstanding advances dropped to zero. When the loan was finally liquidated in August, there were some proceeds remaining after paying various fixed expenses and servicer fees. Since ASERs and advances were already at zero, these were applied as a reimbursement of non-recoverable interest. This led to interest shortfalls being paid down on the highest short-falling tranches in BACM 2006-3, despite overall severities being greater than 100%.

Deal level effect of interest shortfalls resulting from high severities


Typically the loans liquidated with severities exceeding 100% are relatively small. However, the interest shortfalls triggered by such liquidations could reach relatively high up in the capital structure. We have already discussed the originally AA rated C tranche in COMM 2006-C8 that took a shortfall this month as a result of the Gale Avenue liquidation. In June 2011, MSC 2006-IQ12 AJ took interest shortfalls as a result of the liquidation of the Chatham II loan. Though the loan had been deemed non-recoverable, P&I advances had not been reimbursed earlier by the servicer. As a result, the proceeds from liquidation were not enough to reimburse the servicer fees and the advances, resulting in a shortfall. On the flip side, since these shortfalls are linked to a single liquidation, they are usually temporary. Using our example with IQ12, the AJ tranche recovered its shortfalls in the next month following the liquidation. As such, we believe that when a shortfall on a tranche positioned relatively high in the capital structure is caused specifically by liquidation of a loan with severity exceeding 100%, it might present an opportunity to take a long position, since such a shortfall is likely to be recovered in the short term. However, these opportunities should be viewed as very deal specific because they depend on performance of other loans securitized in the same pool.

Reimbursements can be made from principal or interest depending on the PSA language

9 September 2011

26

Barclays Capital | Securitized Products Weekly

CONSUMER ABS

Relative value among prime retail auto loan ABS


Joseph Astorina +1 212 412 5435 joseph.astorina@barcap.com

The auto ABS issuance machine continues to chug along. Auto-related ABS primary market transaction volume totals $42.4bn year-to-date, of which $30.0bn has been backed by retail auto loans. Credit enhancement levels, after rising immediately post the Lehman failure, began declining in 2010 and have now stabilized. Fittingly, transactions completed from 2009 to present have demonstrated improving collateral performance as measured by cumulative net losses. The majority of retail auto loan ABS traded in the secondary market were issued in 2009-2011. We assess collateral performance and credit enhancement of these post credit-crisis transactions across vintages and identify relative value within the prime retail auto ABS sector.

Retail auto ABS issuance remains strong


Year-to-date, issuance of retail auto loan ABS totals close to $30bn, down 13% from the first nine months of 2010 ($34.0bn). Including auto-related ABS transactions total volume has reached $42.4bn in 2011 versus $50.5bn for the same period in 2010. By comparison, retail auto loan securitization volume in 2010 totaled $41.1bn and total auto related was $61.7bn. Despite the y/y decline, auto ABS issuance continues to outpace primary market volume of other non-mortgage asset classes by a wide margin. Figure 1: Retail auto loan ABS issuance ($bn)
20 18 16 14 12 10 8 6 4 2 0 Q1 FORDO NAROT BAAT CFAST HAROT CARMX ALLYA MBART SDART USAOT AMCAR VALET HART WOART TAOT Other

Q2 2009

Q3

Q4

Q1

Q2 2010

Q3

Q4

Q1

Q2 2011

Q3

Source: IFR Markets, Barclays Capital

Of the year-to-date total retail auto loan ABS volume, about 87% has been sold by a dozen issuers, entities that comprise frequent, generally programmatic securitizers of auto ABS (Figure 1). Together, these issues accounted for $25.7bn in new issue transactions through September 9, 2011. Approximately two-thirds of the retail auto issuance has been backed by prime quality collateral, with the remaining third supported by non-prime collateral. Interestingly, two issuers, AMCAR and SDART, are responsible for 75% of the non-prime issuance. For the balance of this article, we will focus on the prime segment of the retail auto ABS market, leaving non-prime for a future analysis.
9 September 2011 27

Barclays Capital | Securitized Products Weekly

Credit enhancement declines have stabilized in 2011


Figure 2 details by quarter of issuance the initial credit enhancement levels for the top 12 issuers, by volume since 2009, of prime retail auto loan ABS. Also depicted, to provide historical perspective, are enhancement levels for transactions from the same issuers completed prior to the failure of Lehman Brothers in September 2008. We excluded non-prime deals because credit enhancement prior to September 2008 was generally provided by a monoline insurance policy, whereas post-crisis deals have been structured with senior/subordinate enhancement. As such, hard credit enhancement for pre-2009 non-prime transactions was structured to a BBB level, with the monoline wrap bringing the rating to AAA. In contrast senior/subordinate non-prime auto loan deals have a much higher hard credit enhancement requirement, making comparisons between pre- and post-crisis transactions tenuous. We also excluded from the chart issuers that had no pre-crisis transactions (e.g., MBART). Figure 2: Credit enhancement trends
21% 18% 15% 12% 9% 6% 3% 0% Pre-09 Q1 09 Q2 09 Q3 09 Q4 09 Q1 10 Q2 10 Q3 10 Q4 10 Q1 11 Q2 11 Q3 11
Source: Moodys Investors Service, Standard & Poors, Fitch Ratings, Barclays Capital

FORDO NAROT

BAAT CFAST

HAROT CARMX

ALLYA USAOT

HART VALET

TAOT WOART

Enhancement requirements for issuers such as ALLYA, CARMX, and NAROT have been reduced by 350-450bp since hitting highs in Q4 09. Others, including FORDO, HAROT, and USAOT, have experienced more modest declines. Regardless, the trend in Figue 2 is clear credit enhancement has declined from Q3 09 highs and has stabilized close to, or below, precrisis levels for most issuers in 2011. In Consumer ABS Strategy Update: Assessing retail auto loan ABS credit enhancement trends, November 19, 2010, we determined that most of the reductions in credit enhancement levels were likely justified given collateral composition and improved cumulative net loss performance metrics of recent transactions. Figure 3 shows the average credit enhancement requirements for issuers in the pre-crisis period and from 2009 to 2011. Issuers are listed in descending order of average initial credit enhancement in 2010 and, in most instances, the rankings are the same for 2011. Notable exceptions include WOART, the enhancement for which was increased to 9.45% by S&P because of the agencys increased cumulative net loss expectation; and VALET and TAOT, both of which had the enhancement reduced slightly for 2011 transactions.

9 September 2011

28

Barclays Capital | Securitized Products Weekly

Figure 3: Average initial credit enhancement, by vintage (%)


Issuer CFAST HART ALLYA BAAT CARMX FORDO WOART VALET TAOT NAROT HAROT USAOT Pre-2009 8.50% 9.75% 6.25% 8.30% 9.26% 5.50% 6.75% 2.35% NA 4.75% 3.15% 3.25% 2009 13.93% 13.60% 6.13% 8.85% 15.42% 6.10% 11.75% 5.60% NA 7.30% 4.05% 3.00% 2010 18.30% 10.80% 9.88% 7.98% 7.92% 6.06% 5.90% 5.60% 5.33% 4.50% 3.33% 2.25% 2011 NA 10.55% 7.45% NA 6.40% 6.00% 9.45% 3.60% 3.65% 4.50% 2.75% NA

Source: Moodys Investors Service, Standard & Poors, Fitch Ratings, Barclays Capital

We note that many of the 2009 and 2010 transactions have deleveraged significantly, resulting in current credit enhancement that is much higher than the original level. However, we believe a comparison of transactions based on original enhancement requirements is more consistent and meaningful. In our view, the relative rankings based on initial credit enhancement provide the first piece to the relative value puzzle.

Recent transaction collateral performance is improved


The second piece is transaction collateral performance. Cumulative net losses provide a clear, observable comparison of performance trends among different transactions from the same issuer, as well as across issuers, after controlling for variations in collateral pools. Figures 4 to 15 detail the cumulative net loss performance for transactions from the largest prime retail auto ABS issuers in 2009-11. We look at performance of recent transactions, as well as deals issued as far back as 2007, to gain insight and perspective on changes in securitized collateral performance through and after the depths of the credit crisis. In general, we find that many shelves posted stable to slightly deteriorating credit performance in 2007 and 2008 vintage transactions. However, post-crisis (i.e., 2009 and later) deals are performing well, with successive deals from the same issuer reporting lower cumulative net losses than previous transactions. In addition, cumulative net losses on recent vintage deals are generally much better than 2007-08 vintages. Specifically, the last CARAT transaction from 2008 is reporting cumulative net losses in line with the early 2007 deals (Figure 4). However, the post-crisis (i.e., 2009 and later) ALLYA deals are performing well, with cumulative net losses much lower than 2007-08 vintages. Similarly, 2009-11 CARMX transactions (Figure 6) report improved performance over early deals, as do those of CFAST (Figure 7). FORDO (Figure 8), HAROT (Figure 9), HART (Figure 10), and NAROT (Figure 11) are all reporting improved cumulative net losses in 2009-11 transactions as well. TAOT performance has always been exceptionally strong, with less than 50bp in cumulative net losses historically (Figure 12). Each successive USAOT, VALET, and WOART transaction since 2008 has reported lower cumulative net losses than the previous (Figure 13 to Figure 15).

9 September 2011

29

Barclays Capital | Securitized Products Weekly

Figure 4: CARAT/ALLYA cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2007-1 2007-4 2009-A 2010-2 2010-5 2011-3 2007-2 2008-1 2009-B 2010-3 2011-1 2007-3 2008-2 2010-1 2010-4 2011-2

Figure 5: BAAT cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2008-1 2009-3 2009-1 2010-1 2009-2 2010-2

Source: Intex, Barclays Capital

Source: Intex, Barclays Capital

Figure 6: CARMX cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2007-1 2008-1 2009-1 2010-1 2011-1 2007-2 2008-2 2009-2 2010-2 2007-3 2008-A 2009-A 2010-3

Figure 7: DCAT/CFAST cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2007-A 2009-A 2008-A 2009-B 2008-B 2010-A

Source: Intex, Barclays Capital

Source: Intex, Barclays Capital

Figure 8: FORDO cumulative net loss, %


2007-A 2008-B 2009-B 2009-E 2011-A 2007-B 2008-C 2009-C 2010-A 2011-B 2008-A 2009-A 2009-D 2010-B

Figure 9: HAROT cumulative net loss, %


7.0 6.0 5.0 2008-1 2009-2 2010-2 2008-2 2009-3 2010-3 2009-1 2010-1 2011-1

7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6

4.0 3.0 2.0 1.0 0.0 11 16 21 26 31 36 41 46 1 6 11 16 21 26 31 36 41 46

Source: Intex, Barclays Capital

Source: Intex, Barclays Capital

9 September 2011

30

Barclays Capital | Securitized Products Weekly

Figure 10: HART cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2007-A 2010-B 2008-A 2011-A 2009-A 2011-B 2010-A

Figure 11: NAROT cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2007-B 2009-A 2008-A 2010-A 2008-B 2011-A 2009-1

Source: Intex, Barclays Capital

Source: Intex, Barclays Capital

Figure 12: TAOT cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2010-A 2010-B 2010-C 2011-A

Figure 13: USAOT cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2008-1 2009-1 2008-2 2009-2 2008-3 2010-1

Source: Intex, Barclays Capital

Source: Intex, Barclays Capital

Figure 14: VALET cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2008-1 2008-2 2010-1 2011-1

Figure 15: WOART cumulative net loss, %


7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1 6 11 16 21 26 31 36 41 46 2007-B 2010-A 2008-A 2011-A 2009-A

Source: Intex, Barclays Capital

Source: Intex, Barclays Capital

9 September 2011

31

Barclays Capital | Securitized Products Weekly

Putting it all together


To complete our analysis, we meld the credit enhancement of specific transactions with collateral performance and assess both against secondary trading levels. We measure relative performance of transactions within a vintage by comparing cumulative net losses at a given deal age. For example, for 2009 vintage transactions, we assess cumulative net losses at 24 months of deal age, or if a transaction is not seasoned that much, at 18 months. For 2011 transactions, we rank deals by cumulative net losses at six months of deal age, or current cumulative net losses for deals not aged at least six months. The majority of retail auto loan ABS traded in the secondary market were issued in 20092011. Most 2009 transactions have paid down to such an extent that there are only one or two senior classes outstanding. The last cash flow class of 2009 deals has generally rolled to become a 1y average life bond, while the penultimate payer is less than 0.4y in most instances. Of outstanding 2009 vintage prime retail auto ABS transactions, we find the first deals that year from CARMX, FORDO, and HART, as well as the second CFAST deal, to be among the highest in cumulative net losses at 24 (or 18) months of deal age. However, these transactions also have the highest levels of credit enhancement in this vintage. Given relatively similar trading levels (i.e., within 2-3bp of each other) of bonds in this vintage, the short average life, and the amount of credit enhancement relative to cumulative net losses, we are generally indifferent among the major issuers of prime transactions in this vintage. Transactions from 2010 generally have two or three senior classes outstanding, with the last cash flow having rolled down to a 2y average life, and the penultimate to a 1y bond. Cumulative net losses for transactions in this vintage are below 1%, with most deals reporting under 0.5% cumulative net losses at 12 or 18 months of deal age. We find penultimate and last cash flow ABS issued in 2010 from HART and ALLYA are attractive given the double-digit initial credit enhancement and the highest spreads available among the 2010 vintage. In contrast to the 2009 and 2010 vintages, most 2011 prime auto ABS issues have all senior classes outstanding (either three or four depending on issuer). Given credit enhancement and cumulative net loss levels in this vintage, at current spread levels we view ALLYA, CARMX, and HART penultimate and last cash flow ABS as attractive. We also like FORDO and WOART last cash flow ABS.

9 September 2011

32

Barclays Capital | Securitized Products Weekly

CONVEXITY PORTFOLIO

Wider basis leads to losses


Mortgages widened sharply over the short week, and our long basis positions underperformed. FNCL 4s lost 21bp versus 2y and 10y Treasuries, and FNCI 3.5s lost 16bp. Our 15s/30s trade offset some of these losses, as DW 3.5s gained 10.5bp versus CL 4.5s. Bank demand remains strong in 15y 3s and 3.5s. Overall, the convexity portfolio was down 68bp on the week and is down 18bp year-to-date. ROE since initiation stands at 105.63%. Figure 1: Convexity portfolio trade performance
Portfolio return statistics YTD P/L Initial equity ($mn) ($mn) Total P/L ($mn) YTD % ROE Total % ROE 1-week P/L ($mn) 1-week % ROE

Convexity Portfolio

100
Total P/L (bp) 1 wk P/L (bp)

-0.37
Equity ($mn)

105.63
Avg. equity ($mn)

-0.18%

105.63%
Total P/L ($000)

-1.41
1 wk P/L ($000)

-0.68%

Convexity trades

Leverage

% ROE

Start date

Long FNCL 4s vs Tsy curve Long FNCI 3.5s vs Tsy curve Long DW 3.5s/FNCL 4.5s

-14.1 7.0 6.1

-21.1 -16.5 10.5

30 30 40

30 30 40

20 20 20

-842 425 495

-2.8 1.4 1.2

-1,264 -986 840

8/26/11 8/26/11 8/26/11

Note: The performance is from Thursday, September 1, 2011, close, to Thursday, September 8, 2011, close. Source: Barclays Capital

Figure 2: Current convexity portfolio trades


Trade 1 2 3 4 Long 30y 4s vs Tsy curve Long 15y 3.5 vs Tsy curve Long DW 3.5s/FNCL 4.5s Long GN 4.5/4 swap Cash
Note: Pricing is as of the close on September 8, 2011. Source: Barclays Capital

Long in face value terms 600 FNCL 4s 600 FNCI 3.5s 800 15y 3.5s, 23 10s 450 GN 4.5s, 8 10s

Short in face value terms 290 2y Tsy, 230 10y Tsy 316 2y Tsy, 123 10y Tsy 770 30y 4.5s, 106 2s

Hedge details Curve-neutral Curve-neutral Curve-neutral

Equity Lever ($mn) age Notional 30 30 40 30 105.6 20 20 20 15 600 600 800 450

Initiation level 105.6bp ZV sprd 85.6bp ZV sprd 104-03 vs 105-16+ 2-04+

Current level 117.9bp ZV sprd 91.2bp ZV sprd 104-13+ vs 105-27

Initiation date 8/26/11 8/26/11 8/26/11 9/9/11

315 GN 4.5s, 16 2s Curve-neutral

9 September 2011

33

Barclays Capital | Securitized Products Weekly

Figure 3: Retired trades


Leveraged portfolio statistics P/L (bp) total 106 94 198 782 48 242 14 25 -5 -16 40 113 41 2 88 55 -11 18 67 -17 80 116 106 -14 -86 -1.8 12.9 -34.1 16.3 -41.4 107.8 93.8 17.2 336.0 64.8 23.44 28.9 72.2 4.7 -54.3 Equity ($mn) 10 20 10 20 10 10 20 20 10 20 10 10 10 20 20 10 20 20 20 20 5 8.4 10 20 20 10 20 20 20 20 20 20 20 20 20 20 20 20 20 10 Avg Equity 10 12 10 17 10 10 19 20 10 20 10 10 10 20 15 10 20 20 20 20 5 8.4 10 20 20 10 20 20 20 20 20 20 20 20 20 20 20 20 20 10 P/L ($mn) 1.07 2.24 1.98 2.66 0.49 0.27 0.54 0.99 -0.05 -0.64 0.80 2.26 0.81 0.07 2.72 1.10 -0.42 0.71 2.67 -0.68 0.2 1.16 2.14 0.53 -3.43 0.28 0.53 2.04 0.65 -1.64 8.63 3.75 1.38 11.29 2.60 1.88 1.16 1.45 0.19 -0.16

Retired trades 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 FN 4.5s vs Treasury GN2/FN 5.0 swaps FN 6.0 LLB vs 3y Treasury FNT 379 vs TBA, 2s10s swaps FN 5.5/5.0 Inverse IO vs 2s10s swaps 2005 FNCL 5.5 vs TBA FNCI 4.0 vs FNMA 4.5 FN 5.5/4.5 GN/FN 4.5 swaps Short FN 5s vs 10y Treasury Short FN 4.5s vs 10y Treasury Long FN 4.5s vs 5.0s GN2/FN 4.5s swaps FN 5.5/4.5 Short FN 4.5s vs 10y Treasury Long GN/FN 5.0 swaps Long FNCI 4.0s vs FNCL 4.5s Long 4.5s vs 10y swaps Long FN 4.5 butterfly Synthetic vs TBA 5.0 Long GN inverse IO Long FN 6.0/5.0 swap Short FN 5.0 butterfly Short FN 5.5 vs 10y Treasury Long FNS 399 vs 2s10s Treasury Short GN/FN 6.0 swap Short FN 4.5 vs Treasury Long FN 6/4.5 swap FNCI 4.5s vs FNCL 5.0s Short FN 6.5 Mar-Apr roll Long Gold/FN 6.5 swap Short Gold/FN 6.0 swap (May settle) Long FH 6.5/4.5 swap Long FN 6.5 Apr/May roll Short FN 5.5 May/Jun roll Short GN/FH 6.0 swap Long synthetic 5.0s vs TBA Short Gold/FN 6.5 swap Long FNS 379 vs TBA

Start date 1/16/09 1/16/09 1/16/09 1/16/09 1/29/09 2/12/09 2/26/09 3/26/09 4/2/09 4/16/09 5/8/09 5/27/09 6/10/09 6/18/09 7/9/09 8/5/09 8/6/09 8/20/09 9/10/09 9/10/09 10/15/09 9/17/09 9/10/09 9/24/09 11/19/09 12/3/09 12/3/09 1/21/10 1/21/10 9/24/09 2/10/10 2/10/10 2/11/10 2/10/10 3/12/10 3/12/10 2/10/10 3/12/10 3/19/10 2/11/10

End date 3/19/09 8/6/09 3/12/09 9/10/09 4/2/09 6/11/09 6/4/09 4/16/09 5/7/09 6/18/09 6/25/09 6/25/09 6/25/09 8/6/09 9/10/09 8/20/09 9/24/09 9/24/09 10/29/09 11/19/09 12/3/09 12/3/09 1/21/10 1/21/10 1/21/10 2/10/10 2/10/10 2/10/10 2/10/10 2/11/10 3/5/10 3/5/10 3/5/10 3/12/10 3/25/10 3/25/10 4/2/10 4/16/10 4/30/10 5/21/10

Leverage 10 20 10 2 10 1 20 20 10 20 20 20 20 20 20 20 20 20 20 20 1 1 20 20 20 2 20 20 20 20 20 20 20 20 20 40 20 10 20 3

% ROE 10.7 18.0 19.8 15.9 4.9 2.7 2.9 4.9 -0.5 -3.2 8.0 22.6 8.1 0.4 17.6 11.0 -2.1 3.5 13.3 -3.4 4.0 13.8 21.4 -2.7 -17.2 2.8 2.6 10.2 3.3 -8.2 43.1 18.8 6.9 67.2 13.0 9.4 5.8 7.2 1.0 -1.6

Source: Barclays Capital

9 September 2011

34

Barclays Capital | Securitized Products Weekly

Figure 3: Retired trades (contd.)


Leveraged portfolio statistics P/L (bp) total -5.7 7.8 43.2 0.8 192.3 26.2 3.6 -0.9 73.8 269 -53.9 -64.5 69.5 -62.5 -2.3 84.0 -18.3 71.2 96.1 47.1 54.7 86.7 81.6 99.9 286.2 -2.3 11.4 -67.5 -77.8 -31.8 -40.4 65.9 79.8 -20.7 16.7 11.4 -9.4 -27.2 -57.4 -54.3 -33.6 10.2 239.2 -32.2 Equity ($mn) 20 20 40 20 10 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 15 20 15 15 15 15 15 15 15 15 20 15 15 20 15 5 20 15 Avg Equity 20 20 30 20 10 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 15 20 18 15 15 15 15 15 15 15 20 15 15 20 15 5 20 15 P/L ($mn) -0.45 0.63 2.60 0.07 3.85 1.05 0.15 -0.02 2.96 10.76 -4.31 -5.15 2.78 -2.47 -0.18 6.72 -1.46 5.71 4.81 2.36 4.38 4.34 4.90 4.02 11.47 -0.14 0.34 -2.69 -3.27 -0.95 -1.21 1.98 2.40 -0.62 0.50 0.34 -1.09 -0.81 -1.71 -2.17 -1.00 0.10 9.58 -0.96

Retired trades 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 Long FN 2005 5.5s versus TBA 5.5s Short GN/FN 5.0s Long FN 4.5s vs 2s10s Long GN2/GN1 4.5s Long credit impaired IIO vs short 6.5 IOS FNCL 5/4.5 down-in-coupon swap Long FNCI 4.5s vs FNCL 5s Long FN 4.5s vs payer CMM Long IOS 5.5% vs down-in-coupon 5.5/5.0 Long IOS 5 vs down-in-coupon 5/4.5 swap Short GN/FN 5.5 Long FNCL 5.5 fly Short FN 4s hedged with Tsy Short GN/FN 6.5 swap Up-in-coupon 5.5/5 swap Short FNCL 5.0 fly Long GN 5.5 fly Long FN 4s hedged with swaps MBX 6s vs MBS 5s Long FNCL 5s vs FNCI 4.5s Long GN/FN 4s vs GN/FN 4.5s Long FN 5.5/4 swap Long FNCL 5.5s versus swaps Down-in-coupon FN 15y 4.5/3.5 Long IOS 5s vs TBA 4s Short FN 5 fly Down-in-coupon 5/4.5 swap FN 4s + IOS 4.5 vs FN 4.5s Long FN 4s vs swaps Long FN 3.5s vs swaps Down-in-coupon FNCL 5.5/4.5 Long FN 4.5s vs swaps Long FN 5s vs swaps Down-in-coupon DW 4.5/4 swap Down-in-coupon FNCL 4.5/4 swap Long FN 4s versus swaps Long Gold/FN 5.5s Down-in-coupon FNCL 5.5/4 swap Short GN/FN 4s Down-in-coupon FNCL 5/4 GN/FN 4/5 box trade Down-in-coupon DW 4.5/4 swap Long IOS 6s vs TBA 4s Down-in-coupon DW 4.5/3.5 swap

Start date 3/12/10 5/27/10 4/2/10 6/4/10 6/4/10 6/25/10 6/18/10 4/30/10 6/11/10 8/6/10 7/9/10 7/9/10 8/10/10 4/30/10 8/27/10 8/6/10 8/20/10 8/20/10 9/10/10 10/8/10 9/10/10 10/8/10 10/15/10 7/30/10 8/5/10 12/3/10 1/7/11 11/5/10 11/12/10 1/14/11 2/4/11 2/11/11 2/11/11 1/28/11 3/11/11 3/18/11 2/11/11 3/25/11 1/28/11 3/21/11 2/25/11 4/15/11 1/14/11 3/18/11

End date 5/21/10 6/10/10 6/24/10 7/1/10 7/1/10 7/8/10 7/23/10 8/5/10 8/5/10 8/13/10 8/20/10 8/20/10 8/20/10 9/10/10 9/16/10 9/23/10 10/7/10 10/15/10 10/15/10 10/29/10 11/5/10 11/12/10 11/12/10 1/7/11 1/14/11 1/14/11 2/4/11 2/11/11 2/11/11 2/11/11 2/18/11 3/4/11 3/18/11 3/18/11 3/18/11 3/21/11 3/24/11 4/8/11 4/15/11 4/29/11 5/6/11 5/6/11 5/13/11 5/20/11

Leverage 40 40 20 40 2 20 20 20 20 20 40 40 20 20 20 40 40 40 25 20 40 30 30 20 20 30 20 20 20 20 20 20 20 20 20 20 40 20 20 20 20 20 20 20

% ROE -2.2 3.1 8.7 0.3 38.5 5.2 0.8 -0.1 14.8 53.8 -21.5 -25.7 13.9 -12.4 -0.9 33.6 -7.3 28.5 24.0 11.8 21.9 21.7 24.5 20.1 57.4 -0.7 2.3 -13.4 -17.7 -6.3 -8.1 13.2 16.0 -4.1 3.4 2.3 -5.4 -5.4 -11.4 -10.8 -6.7 2.1 47.9 -6.4

Source: Barclays Capital

9 September 2011

35

Barclays Capital | Securitized Products Weekly

Figure 3: Retired trades (contd.)


Leveraged portfolio statistics P/L (bp) total 16.4 -4.1 256.6 -10.5 52.5 -61.5 2.0 36.9 25.4 32.4 60.2 -1.8 1.6 -83.6 -99.5 Equity ($mn) 20 20 15 20 20 20 20 20 20 20 20 15 40 50 50 Avg Equity 20 20 15 20 20 20 20 20 20 20 20 15 40 50 50 P/L ($mn) 1.32 -0.16 5.78 -0.84 2.11 -2.46 0.09 1.48 1.02 2.60 2.41 -0.04 0.13 9.38 9.09

Retired trades 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 Long Gold/FN 5s Long 30y 4.5s/15y 4s Long IOS 2010 5s vs TBA 4s Long GN2/GN1 4.5s Long IOS 5.5 03 vs 08 Long synthetic 4.5s vs swaps Long 30y 5s vs swaps Long FNCL 5/4 swap Short FNCL 5.5/5 swap Long GN2/GN1 4s Short FN 4.5 fly Long IOS 5 09 vs IOS 4.5 09 Short DW 4s/FNCL 4.5s Long FN 4.5s vs 10y Tsy Long FN 5s vs 10y Tsy

Start date 3/25/11 5/6/11 4/1/11 4/15/11 4/15/11 6/3/11 5/6/11 5/6/11 6/10/11 6/24/11 7/15/11 6/17/11 8/5/11 8/10/11 8/10/11

End date 5/20/11 6/9/11 6/17/11 6/24/11 6/24/11 6/30/11 7/15/11 8/5/11 8/5/11 8/11/11 8/18/11 8/25/11 8/25/11 8/25/11 8/25/11

Leverage 40 20 15 40 20 20 20 20 20 40 20 15 20 15 15

% ROE 6.6 -0.8 38.5 -4.2 10.5 -12.3 0.4 7.4 5.1 13.0 12.1 -0.2 0.3 -18.8 -18.2

Source: Barclays Capital

Figure 4: RoE since inception


140 120 100 80 60 40 20 0 -20 Jan-09 Apr-09 Jul-09 Oct-09 Jan-10 Apr-10 Jul-10 Oct-10 Jan-11 Apr-11 Jul-11
Source: Barclays Capital

9 September 2011

36

Barclays Capital | Securitized Products Weekly

CREDIT PORTFOLIO

Portfolio stays level


Our portfolio gained 22bp this week as both CMBS and non-agency markets stayed stable. Non-agency cash bonds gained modestly, with alt-A and jumbo fixed bonds up 0.5pt and jumbo and alt-A ARMs flat. PrimeX fell marginally, with decreases of around 1/4pt. ABX 071 AAA fell 3/4pts, in line with equities. Our CMBS cash positions were flat, with only carry gains coming into the portfolio. Our hedged alt-A positions gained, as alt-A increased modestly and credit hedges rose. The total return on the credit portfolio since inception is 49.5%, and the year-to-date return is -0.78%. Figure 1: Credit portfolio trade performance,
YTD P/L ($mn) -1.17 Total P/L (bp) 1-wk P/L (bp) Equity ($mn) Total P/L ($mn) 49.5 Avg. EQ ($mn) YTD % ROE -0.78 1 week P/L ($ 000) Total % ROE 49.5 Total P/L ($ 000) 1 week P/L ($mn) 0.33 % ROE 1 week % ROE 0.22 Start date

Portfolio return statistics Credit portfolio Credit trades Long 2007 Alt-A Hybrid MEZZ Long Alt-A Fixed WAC pass-through SSNR vs basket of shorts Long Alt-A ARM WAC pass-through SSNR vs basket of shorts Long 2007 Alt-A Hybrid SSNR 2 Long 2007 Jumbo FIX WAC pass-though SSNR (repo-levered) Long Snr mezz off Prime/Alt-A re-REMIC Long PrimeX FRM.2 Long PrimeX FRM.1 Long Alt-A Fixed SSNR Long back end re-remics off 2007 GG10 A4s Long 2007 CMBS AMs Long 2007 CMBS PAC IOs Long 2007 CMBS AMs (2) Long 2007 Alt-A Fixed Levered Long ABX 07-1 AAA unlevered Long 2007 AJ unlevered Long CMBX AA 2 Long ABX 07-1 AAA unlevered (2) Long PrimeX ARM.1 Long 2006 AM Long 2007 DUPER

8,780 5,102 3,429 4,136 6,304 2,603 5,720 13,210 3,059 (5,176) (8,373) 9,308 (8,914) 2,026 (1,517) (3,816) (3,249) (1) 225 46 280

11 300 219 12 106 9 (140) (382) 82 63 57 328 57 312 (156) 14 44 (156) (107) 41 11

2.8 6.2 6.9 4.7 5.2 5.0 3.8 4.3 4.5 1.0 3.0 1.0 3.0 7.2 7.4 10.0 5.0 7.4 1.9 5.0 5.0

3.0 7.0 7.5 4.9 4.6 5.0 4.4 4.9 4.8 1.0 3.0 1.0 3.0 7.3 7.4 10.0 5.0 7.5 2.0 5.0 5.0

3 187 151 5 55 4 (54) (165) 37 6 17 33 17 225 (115) 14 22 (116) (21) 20 6

2,624 3,612 2,636 2,026 2,873 1,301 2,503 6,425 1,465 (518) (2,512) 931 (2,674) 1,487 (1,126) (3,816) (1,625) (0) 44 23 140

87.80 51.02 34.29 41.36 63.04 26.03 57.20 132.10 30.59 (51.76) (83.73) 93.08 (89.14) 20.26 (15.17) (38.16) (32.49) (0.01) 2.25 0.46 2.80

7/10/2009 7/31/2009 7/31/2009 11/13/2009 1/7/2010 3/12/2009 5/6/2010 5/21/2010 6/3/2010 2/4/2011 2/4/2011 2/4/2011 2/24/2011 3/17/2011 3/17/2011 3/17/2011 4/8/2011 6/8/2011 6/23/2011 8/8/2011 8/8/2011

Note: The performance is from Friday, September 2, 2011, close, to Thursday, September 8, 2011, close. Cash non-agency positions are marked on Wednesday and indices and cash CMBS positions are marked on Thursday. Source: Barclays Capital

9 September 2011

37

Barclays Capital | Securitized Products Weekly

Figure 2: Active credit portfolio


Long notional (mn) 23 7.7 9.3 9.6 23 6.25 50 50 7.7 5 15.1 500 14.9 36.6 17.3 12.7 6.27 20.38 48 5.3 4.9 1 1 1 1 1 1 1 12.9 15.5 Short notional (mn) Hedge ratio Equity ($mn) 2.8 6.2 6.9 4.7 5.2 5 3.8 4.3 4.5 1 3 1 3 7.2 7.4 10 5 7.5 1.9 5.0 5.0 100.4 26.4 Long Long Short Short current initiatio current initiation Initiation Leverage level date n level level level 1 1 1 1 1.7 1 10 22 1 5 5 5 5 4 1 1 1 1 24 1 1 8 79.5 57.6 57.6 90.2 91.5 99.0 105.8 69.5 S + 515 S + 710 T + 450 S + 710 79.5 36.9 S+1530 53.7 36.9 103.5 S + 580 S + 320 11 63 54 54 86 80 100 105.5 65 S + 270 S + 250 T + 400 S + 235 82 43.25 S+800 79.75 36.8 104.1 S + 600 S + 375 86.3* 86.3* 88.5* 88.5* 7/10/09 7/30/09 7/30/09 11/13/0 9 1/8/10 3/12/10 5/6/10 5/21/10 6/3/10 2/4/11 2/4/11 2/4/11 2/24/11 3/17/11 3/17/11 3/17/11 4/8/11 6/8/11 6/23/11 8/8/11 8/8/11

Trade 1 2 3 4 5 6 7 9 Long 2007 Alt-A hybrid MEZZ Long Alt-A fixed WAC pass-through SSNR vs basket of shorts Long Alt-A hybrid WAC pass-through SSNR vs basket of shorts Long Alt-A hybrid WAC pass-through SSNR Long 2007 jumbo FIX WAC pass-through SSNR (repo-leveraged) Long sr mezz of prime/alt-A re-REMIC Long PrimeX FRM.2 Long PrimeX FRM.1

10 Long 2007 Alt-A fixed SSNR 11 Long back end re-REMICs off 2007 GG10 A4s (repo-leveraged and duration hedged) 12 Long 2007 AMs (repo-leveraged and duration hedged) 13 Long 2007 PAC IOs (repo-leveraged and duration hedged) 14 Long 2007 AMs (repo-leveraged and duration hedged) 15 Long 2007 Alt-A fixed (repo-levered) 16 Long ABX 2007-1 AAA (cash basis) 17 Long 2007 AJ (duration hedged) 18 Long CMBX.AA.2 19 Long ABX 2007-1 AAA (cash basis) 20 Long PrimeX ARM.1 21 Long 2006 AM (duration hedged) 22 Long 2007 DUPER ( duration hedged) Equity invested Cash available

Note: * Levels are calculated for a basket of shorts consisting of 1 part CMBX AJ.4 and 3 parts CDX.HY. The performance is from Friday, September 2, 2011, close, to Thursday, September 8, 2011, close. Cash non-agency positions are marked on Wednesday and indices and cash CMBS positions are marked on Thursday. Source: Barclays Capital

9 September 2011

38

Barclays Capital | Securitized Products Weekly

Figure 3: Retired trades


Trade 1 2 3 4 5 6 7 8 9 Long ABX 07-1 PAAA vs 07-1 AAA Long 2007 Alt-A FRM SSNR Long 2007 Alt-A FRM sr mezz Long 1H06 and prior CMBS AM Long 2006 option ARM SSNRs from XS OC deals Long 2007 subprime 3rd CF from imp pro-rata deals Short CMBX.BBB-.1 Long 2005 vintage LCF super duper (hedged with swaps) Long recent vintage CMBS A2 classes 7.79 7.5 9 4.8 3.4 10 5 5 7.9 10 3 10 11.7 10 10 3.5 1.9 1.6 3.9 10 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 12.4 12.7 13.1 12.1 1 1/15/09 3/5/09 3/12/09 5/28/09 9/25/09 10/22/09 2/9/10 11/20/09 7/30/09 7/16/09 1/7/10 5/28/09 7/15/10 4/29/09 8/5/10 6/3/09 7/7/09 8/13/09 9/2/09 5/10/10 78 60 34 48 32.2 28.5 16.4 5.3 21.5 0.4 36.3 59 85.1 7.5 74.8 100 100 100 100 4.1 7/14/09 7/15/09 7/23/09 10/2/09 10/26/09 2/12/09 4/16/10 5/10/10 6/24/10 6/24/10 7/01/10 8/25/10 9/29/10 10/13/10 10/20/10 2/16/11 2/16/11 2/16/11 2/16/11 4/6/11 96 64 49 60 21.3 24.75 8.9 16.7 27.04 3.13 40.83 80 94.4 7.7 86.3 115.2 118.9 116.9 116.4 3.1 2295 1022 4580 1538 545 1193 1193 1076 (957) 852 1,131 3,140 1,291 254 1539 1545 1196 998 2536 (464) 24.4 13.1 50.9 31.9 16.0 11.93 14.9 21.5 (9.57) 8.52 11.31 49.38 12.91 2.54 15.39 43.1 60.8 62.8 65.2 (4.64) 10 Long ABX 06-2 PAAA 11 Long 2006 option ARM SSNRs from XS OC deals Mar 2009 12 Long 2007 alt-A ARM SSNR 13 Short CMBX.A.3 2 14 Long ABX 06-2 AAA vs PAAA 15 Long CMBX.AJ.2, short CMBX.AM.2 16 Long CMBX.AJ2 vs short CMBX.AJ.5 17 Short CMBX.A.3 18 Short CMBX.A.4 vs CMBX A.3 19 Long ABX 07-2 AAA 20 Long 2007 jumbo hybrid SSNR 21 Long 2006 CMBS AM 22 Long PrimeX ARM.1 vs short ARM.2 23 Long 2006 CMBS AJ 24 Long Basket TALF-eligible consumer ABS 25 Long Basket 2009 TALF-eligible ABS 2 26 Long basket 2009 TALF-eligible ABS 3 27 Long basket 2009 TALF-eligible ABS 4 28 Long CMBX.AM2 vs short CMBX.AM.5 Equity ($mn) 0.9 10 2.5 4.3 9 6.6 10 8.7 Leverage 1 1 1 1 1 1 1 1 Initiation date 1/15/09 3/4/09 3/4/09 3/24/09 1/15/09 1/15/09 5/15/09 6/29/09 Initiation Closing level Closing date level 2.5 43 13 41.3 44 32 22 S+495 3/5/09 4/30/09 4/30/09 5/7/09 6/4/09 6/4/09 6/18/09 7/10/09 6.5 58 20 55.2 41 31 15 S+300 bp Total PL ($000) 400 3702 1566 1225 (251) (381) 690 1057 Total ROE (%) 44.49 37.02 62.65 28.32 (2.8) (5.8) 8.85 12.14

Note: The initiation long price levels are the offer-side marks for the cash bonds as of the initiation date; the long current level is the bid-side mark for the cash bond at the current reporting date. Similarly, the initiation short price is the offer-side mark for protection at the initiation date, and the short current level is the bid-side mark for protection at the current reporting date. We will assume appropriate bid offers as observed in the market for our trade initiation and termination dates. Source: Barclays Capital

Figure 4: ROE since inception (%)


70 60 50 40 30 20 10 0 -10 Jan-09 Apr-09 Jul-09 Oct-09 Jan-10 Apr-10 Jul-10 Oct-10 Jan-11 Apr-11 Jul-11
Source: Barclays Capital

9 September 2011

39

Barclays Capital | Securitized Products Weekly

US SECURITISATION STRATEGY ANALYSTS


Ajay Rajadhyaksha Head of US Fixed Income and Securitised Products Strategy +1 212 412 7669 ajay.rajadhyaksha@barcap.com Joseph Astorina ABS Strategy +1 212 412 5435 joseph.astorina@barcap.com Sandipan Deb US RMBS and CMBS Strategy +1 212 412 2956 sandipan.deb@barcap.com Dennis Lee US RMBS and CMBS Strategy +1 212 412 2009 dennis.lee2@barcap.com Nicholas Strand Agency MBS Strategy +1 212 412 2057 nicholas.strand@barcap.com Sandeep Bordia US RMBS and CMBS Strategy +1 212 412 2099 sandeep.bordia@barcap.com Aaron Haan US RMBS +1 212 412 2099 aaron.haan@barcap.com Wei-Ang Lee Agency MBS Strategy +1 212 412 5356 weiang.lee@barcap.com Robert Tayon ABS Strategy +1 212 412 2512 robert.tayon@barcap.com Mukul Chhabra Agency MBS Strategy +1 212 412 2089 mukul.chhabra@barcap.com Sarah Johns ABS Strategy +1 212 412 2099 sarah.johns@barcap.com Keerthi Raghavan US RMBS and CMBS Strategy +1 212 412 7947 keerthi.raghavan@barcap.com Julia Tcherkassova US RMBS and CMBS Strategy +1 212 412 5977 julia. tcherkassova @barcap.com Derek Chen Agency MBS Strategy +1 212 412 2857 derek.chen@barcap.com Shweta Kapadia Agency MBS Strategy +1 212 412 2099 shweta.kapadia@barcap.com Siddarth Ramkumar Agency MBS Strategy +1 212 412 7581 siddarth.ramkumar@barcap.com Jasraj P. Vaidya US RMBS and CMBS Strategy +1 212 412 2265 jasraj.vaidya@barcap.com

9 September 2011

40

Analyst Certification(s) We, Aaron Haan, Sandipan Deb, Ajay Rajadhyaksha, Nicholas Strand, Sandeep Bordia, Dennis Lee, Jasraj Vaidya, Keerthi Raghavan, Julia Tcherkassova, Joseph Astorina, Derek Chen, Wei-Ang Lee and Siddarth Ramkumar, hereby certify (1) that the views expressed in this research report accurately reflect our personal views about any or all of the subject securities or issuers referred to in this research report and (2) no part of our compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this research report. To the extent that any of the conclusions are based on a quantitative model, Barclays Capital hereby certifies (1) that the views expressed in this research report accurately reflect the firm's quantitative research model and (2) no part of the firm's compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this research report. Important Disclosures For current important disclosures regarding companies that are the subject of this research report, please send a written request to: Barclays Capital Research Compliance, 745 Seventh Avenue, 17th Floor, New York, NY 10019 or refer to https://ecommerce.barcap.com/research/cgi-bin/all/disclosuresSearch.pl or call 212-526-1072. Barclays Capital does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that Barclays Capital may have a conflict of interest that could affect the objectivity of this report. Any reference to Barclays Capital includes its affiliates. Barclays Capital and/or an affiliate thereof (the "firm") regularly trades, generally deals as principal and generally provides liquidity (as market maker or otherwise) in the debt securities that are the subject of this research report (and related derivatives thereof). The firm's proprietary trading accounts may have either a long and / or short position in such securities and / or derivative instruments, which may pose a conflict with the interests of investing customers. Where permitted and subject to appropriate information barrier restrictions, the firm's fixed income research analysts regularly interact with its trading desk personnel to determine current prices of fixed income securities. The firm's fixed income research analyst(s) receive compensation based on various factors including, but not limited to, the quality of their work, the overall performance of the firm (including the profitability of the investment banking department), the profitability and revenues of the Fixed Income Division and the outstanding principal amount and trading value of, the profitability of, and the potential interest of the firms investing clients in research with respect to, the asset class covered by the analyst. To the extent that any historical pricing information was obtained from Barclays Capital trading desks, the firm makes no representation that it is accurate or complete. All levels, prices and spreads are historical and do not represent current market levels, prices or spreads, some or all of which may have changed since the publication of this document. Barclays Capital produces a variety of research products including, but not limited to, fundamental analysis, equity-linked analysis, quantitative analysis, and trade ideas. Recommendations contained in one type of research product may differ from recommendations contained in other types of research products, whether as a result of differing time horizons, methodologies, or otherwise. Barclays Capital is acting as financial advisor to the Central Bank of Ireland in relation to their Prudential Capital Assessment Review and Prudential Liquidity Assessment Review. As such, Barclays Capital research on Allied Irish Banks plc and The Bank of Ireland is restricted and ratings and estimates suspended. Barclays Capital is providing investment banking services to Ocwen Financial Corp in its potential acquisition of Litton Loan Servicing LP from Goldman Sachs. .

Disclaimer This publication has been prepared by Barclays Capital, the investment banking division of Barclays Bank PLC, and/or one or more of its affiliates as provided below. It is provided to our clients for information purposes only, and Barclays Capital makes no express or implied warranties, and expressly disclaims all warranties of merchantability or fitness for a particular purpose or use with respect to any data included in this publication. Barclays Capital will not treat unauthorized recipients of this report as its clients. Prices shown are indicative and Barclays Capital is not offering to buy or sell or soliciting offers to buy or sell any financial instrument. Without limiting any of the foregoing and to the extent permitted by law, in no event shall Barclays Capital, nor any affiliate, nor any of their respective officers, directors, partners, or employees have any liability for (a) any special, punitive, indirect, or consequential damages; or (b) any lost profits, lost revenue, loss of anticipated savings or loss of opportunity or other financial loss, even if notified of the possibility of such damages, arising from any use of this publication or its contents. Other than disclosures relating to Barclays Capital, the information contained in this publication has been obtained from sources that Barclays Capital believes to be reliable, but Barclays Capital does not represent or warrant that it is accurate or complete. The views in this publication are those of Barclays Capital and are subject to change, and Barclays Capital has no obligation to update its opinions or the information in this publication. The analyst recommendations in this publication reflect solely and exclusively those of the author(s), and such opinions were prepared independently of any other interests, including those of Barclays Capital and/or its affiliates. This publication does not constitute personal investment advice or take into account the individual financial circumstances or objectives of the clients who receive it. The securities discussed herein may not be suitable for all investors. Barclays Capital recommends that investors independently evaluate each issuer, security or instrument discussed herein and consult any independent advisors they believe necessary. The value of and income from any investment may fluctuate from day to day as a result of changes in relevant economic markets (including changes in market liquidity). The information herein is not intended to predict actual results, which may differ substantially from those reflected. Past performance is not necessarily indicative of future results. This communication is being made available in the UK and Europe primarily to persons who are investment professionals as that term is defined in Article 19 of the Financial Services and Markets Act 2000 (Financial Promotion Order) 2005. It is directed at, and therefore should only be relied upon by, persons who have professional experience in matters relating to investments. The investments to which it relates are available only to such persons and will be entered into only with such persons. Barclays Capital is authorized and regulated by the Financial Services Authority ('FSA') and member of the London Stock Exchange. Barclays Capital Inc., U.S. registered broker/dealer and member of FINRA (www.finra.org), is distributing this material in the United States and, in connection therewith accepts responsibility for its contents. Any U.S. person wishing to effect a transaction in any security discussed herein should do so only by contacting a representative of Barclays Capital Inc. in the U.S. at 745 Seventh Avenue, New York, New York 10019. Non-U.S. persons should contact and execute transactions through a Barclays Bank PLC branch or affiliate in their home jurisdiction unless local regulations permit otherwise. This material is distributed in Canada by Barclays Capital Canada Inc., a registered investment dealer and member of IIROC (www.iiroc.ca). To access Barclays Capital policy on the dissemination of research reports, please go to http://www.barcap.com/Client+offering/Research/Research+Policy. Subject to the conditions of this publication as set out above, Absa Capital, the Investment Banking Division of Absa Bank Limited, an authorised financial services provider (Registration No.: 1986/004794/06), is distributing this material in South Africa. Absa Bank Limited is regulated by the South African Reserve Bank. This

publication is not, nor is it intended to be, advice as defined and/or contemplated in the (South African) Financial Advisory and Intermediary Services Act, 37 of 2002, or any other financial, investment, trading, tax, legal, accounting, retirement, actuarial or other professional advice or service whatsoever. Any South African person or entity wishing to effect a transaction in any security discussed herein should do so only by contacting a representative of Absa Capital in South Africa, 15 Alice Lane, Sandton, Johannesburg, Gauteng 2196. Absa Capital is an affiliate of Barclays Capital. In Japan, foreign exchange research reports are prepared and distributed by Barclays Bank PLC Tokyo Branch. Other research reports are distributed to institutional investors in Japan by Barclays Capital Japan Limited. Barclays Capital Japan Limited is a joint-stock company incorporated in Japan with registered office of 6-10-1 Roppongi, Minato-ku, Tokyo 106-6131, Japan. It is a subsidiary of Barclays Bank PLC and a registered financial instruments firm regulated by the Financial Services Agency of Japan. Registered Number: Kanto Zaimukyokucho (kinsho) No. 143. Barclays Bank PLC, Hong Kong Branch is distributing this material in Hong Kong as an authorised institution regulated by the Hong Kong Monetary Authority. Registered Office: 41/F, Cheung Kong Center, 2 Queen's Road Central, Hong Kong. This material is issued in Taiwan by Barclays Capital Securities Taiwan Limited. This material on securities not traded in Taiwan is not to be construed as 'recommendation' in Taiwan. Barclays Capital Securities Taiwan Limited does not accept orders from clients to trade in such securities. This material may not be distributed to the public media or used by the public media without prior written consent of Barclays Capital. All equity research material is distributed in India by Barclays Securities (India) Private Limited (SEBI Registration No: INB/INF 231292732 (NSE), INB/INF 011292738 (BSE), Registered Office: 208 | Ceejay House | Dr. Annie Besant Road | Shivsagar Estate | Worli | Mumbai - 400 018 | India, Phone: + 91 22 67196363). Other research reports are distributed in India by Barclays Bank PLC, India Branch. Barclays Bank PLC Frankfurt Branch distributes this material in Germany under the supervision of Bundesanstalt fr Finanzdienstleistungsaufsicht (BaFin). This material is distributed in Malaysia by Barclays Capital Markets Malaysia Sdn Bhd. This material is distributed in Brazil by Banco Barclays S.A. This material is distributed in Mexico by Barclays Bank Mexico, S.A. Barclays Bank PLC in the Dubai International Financial Centre (Registered No. 0060) is regulated by the Dubai Financial Services Authority (DFSA). Barclays Bank PLC-DIFC Branch, may only undertake the financial services activities that fall within the scope of its existing DFSA licence. Barclays Bank PLC in the UAE is regulated by the Central Bank of the UAE and is licensed to conduct business activities as a branch of a commercial bank incorporated outside the UAE in Dubai (Licence No.: 13/1844/2008, Registered Office: Building No. 6, Burj Dubai Business Hub, Sheikh Zayed Road, Dubai City) and Abu Dhabi (Licence No.: 13/952/2008, Registered Office: Al Jazira Towers, Hamdan Street, PO Box 2734, Abu Dhabi). Barclays Bank PLC in the Qatar Financial Centre (Registered No. 00018) is authorised by the Qatar Financial Centre Regulatory Authority (QFCRA). Barclays Bank PLC-QFC Branch may only undertake the regulated activities that fall within the scope of its existing QFCRA licence. Principal place of business in Qatar: Qatar Financial Centre, Office 1002, 10th Floor, QFC Tower, Diplomatic Area, West Bay, PO Box 15891, Doha, Qatar. This material is distributed in Dubai, the UAE and Qatar by Barclays Bank PLC. Related financial products or services are only available to Professional Clients as defined by the DFSA, and Business Customers as defined by the QFCRA. This material is distributed in Saudi Arabia by Barclays Saudi Arabia ('BSA'). It is not the intention of the Publication to be used or deemed as recommendation, option or advice for any action (s) that may take place in future. Barclays Saudi Arabia is a Closed Joint Stock Company, (CMA License No. 09141-37). Registered office Al Faisaliah Tower | Level 18 | Riyadh 11311 | Kingdom of Saudi Arabia. Authorised and regulated by the Capital Market Authority, Commercial Registration Number: 1010283024. This material is distributed in Russia by OOO Barclays Capital, affiliated company of Barclays Bank PLC, registered and regulated in Russia by the FSFM. Broker License #17711850-100000; Dealer License #177-11855-010000. Registered address in Russia: 125047 Moscow, 1st Tverskaya-Yamskaya str. 21. This material is distributed in Singapore by the Singapore branch of Barclays Bank PLC, a bank licensed in Singapore by the Monetary Authority of Singapore. For matters in connection with this report, recipients in Singapore may contact the Singapore branch of Barclays Bank PLC, whose registered address is One Raffles Quay Level 28, South Tower, Singapore 048583. Barclays Bank PLC, Australia Branch (ARBN 062 449 585, AFSL 246617) is distributing this material in Australia. It is directed at 'wholesale clients' as defined by Australian Corporations Act 2001. IRS Circular 230 Prepared Materials Disclaimer: Barclays Capital and its affiliates do not provide tax advice and nothing contained herein should be construed to be tax advice. Please be advised that any discussion of U.S. tax matters contained herein (including any attachments) (i) is not intended or written to be used, and cannot be used, by you for the purpose of avoiding U.S. tax-related penalties; and (ii) was written to support the promotion or marketing of the transactions or other matters addressed herein. Accordingly, you should seek advice based on your particular circumstances from an independent tax advisor. Barclays Capital is not responsible for, and makes no warranties whatsoever as to, the content of any third-party web site accessed via a hyperlink in this publication and such information is not incorporated by reference. Copyright Barclays Bank PLC (2011). All rights reserved. No part of this publication may be reproduced in any manner without the prior written permission of Barclays Capital or any of its affiliates. Barclays Bank PLC is registered in England No. 1026167. Registered office 1 Churchill Place, London, E14 5HP. Additional information regarding this publication will be furnished upon request.