SECURITIZATION RESEARCH 9 Se ptember 2011 PLEASE SEE ANALYST CERTIFICATIONS AND IMPORTANT DISCLOSURES STARTING AFTER PAGE 40 SECURITIZED PRODUCTS WEEKLY Trends and issues 3 President Barack Obama’s 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 Bernanke’s 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 America’s exiting the correspondent busin ess 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%. Year- to-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%. Views on a Page 2 Ajay Rajadhyaksha +1 212 412 7669 [email protected]www.barcap.com
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
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 Administration’s American Job’s act,1
the administration’s 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, largersteps 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. FNCL5s 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.
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 refinancingprogram (HARP) to help more borrowers benefit from today’s historically low interest rates. This has the potential tonot only help these borrowers, but their communities and the American taxpayer, by keeping borrowers in theirhomes and reducing risk to Fannie Mae and Freddie Mac.” http://www.whitehouse.gov/the-press-office/2011/09/08/fact-sheet-american-jobs-act
Figure 1: Sharp widening in the coupon stack (ticks) Figure 2: GN/FN swaps march higher (ticks)
-25
-20
-15
-10
-5
0
FNMA
3.5s
FNMA
4s
FNMA
4.5s
FNMA
5s
FNMA
5.5s
FNMA
6s
vs 10y swap vs 10y Tsy vs Tsy curve and vol
50
55
60
65
70
75
80
85
90
1-
Jun
15-
Jun
29-
Jun
13-
Jul
27-
Jul
10-
Aug
24-
Aug
7-
Sep
GN/FN 4s GN/FN 4.5s GN/FN 5s
Note: Performance from Sep 1 to Sep 8 close. Source: Barclays Capital Source: Barclays Capital
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 week’s
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 year’s refi episode as well, as newly incentivized borrower’s 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 large-
scale 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 bycoupon, 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 tofaster 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
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. Thoughthere 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
Note: Prices as of Sep 7, 2011, for cash bonds and September 8, 2011, for synthetic indices. Weekly changes areWednesday-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 underwroteand 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
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 Trade recommendation Rationale
Overall view Favorable Higher yields than other risky assets.
Strong demand, cheap leverage, supply coming
from strong sources that can afford to slowdown if needed.
Outright longs Leveraged or unleveragedalt-A FRM/PrimeX.FRM1
High yields across scenarios, steady carry,limited downside potential.
Enough support to protect principal, strongyields in stress scenarios, lower Libor forwards.
Recoveryoriented
2006-07 subprime pro rataLCF
Subprime 03-04 perma-failmezz M1/M2
High near-term severities hasten crossover,longer-term declines in severities benefit pricesat recent lows. Limited further worseningpossible because of supply pressures.
Enough support to protect principal, strongyield profiles; buy on any dips.
Collateralstories
Favor MTA negams versushybrid negams
Favor deals with amortizingIO loans in jumbo
MTAs amortizing versus hybrids will likely havesmaller payment shocks and, eventually, lower
severities than hybrid option ARMs.IO loans underperforming by 1.5-2.0x, butamortizing IOs will likely have lower paymentshocks and outperform.
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 theweek 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 market’s 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
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 Moody’s 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 beliquidated 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 2007-
GG9 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.
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 pay-
downs 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 self-
sustainability. 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.
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 mortgagerate 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
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
Cpn Vintage
Cohort
($bn) BoA ($bn)
BoA Corr
($bn) Pct BoA Corr Cohort CPR Cohort CRR BoA CPR
Cohort CPR
impact
4.5 2010 120 25 12 10% 19 19.2 19.8 1.1
2009 248 58 24 10% 27 26.2 23.3 1.2
5 2010 60 16 4 6% 17 16.6 12.7 0.4
2009 64 14 6 10% 29 27.5 25.1 1.3
2008 50 15 4 8% 44 41.6 35.4 1.5
5.5 2008 75 21 7 9% 38 34.4 32.1 1.6
6 2008 37 8 3 8% 31 24.5 27.8 1.0
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 marketwhere 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
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, whichwould 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 put-
back 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 notchange 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 notcaused 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
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 77-
year 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. Anychange (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.
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
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 anumber 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.
2 We discuss this in more detail in the CMBS Strategy Weekly , July 15, 2011
Figure 3: Liquidations with severities exceeding 100%
0
5
10
15
20
25
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 trustreceives 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/non-
recoverable advances, severities typically will be reported at or above 100%.
Deemed non-recoverable interest or Advances (prior shortfall)
Deemed non-recoverable interest or Advances (paid from trustprincipal)
Liquidation Expenses 1 through 4
Net Proceeds Proceeds-Liquidation Expenses
Source: Barclays Capital
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 non-
recoverable 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 Boscov’s
Monroeville Mall loan, in BACM 2006-3 was also liquidated in August with a severity
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 atzero, 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 notenough 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.
3 Reimbursements can be made from principal or interest depending on the PSA language
Source: Moody’s Investors Service, Standard & Poor’s, Fitch Ratings, Barclays Capital
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, pre-
crisis 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. Notableexceptions include WOART, the enhancement for which was increased to 9.45% by S&P
because of the agency’s increased cumulative net loss expectation; and VALET and TAOT,
both of which had the enhancement reduced slightly for 2011 transactions.
Figure 3: Average initial credit enhancement, by vintage (%)
Issuer Pre-2009 2009 2010 2011
CFAST 8.50% 13.93% 18.30% NA
HART 9.75% 13.60% 10.80% 10.55%
ALLYA 6.25% 6.13% 9.88% 7.45%
BAAT 8.30% 8.85% 7.98% NA
CARMX 9.26% 15.42% 7.92% 6.40%
FORDO 5.50% 6.10% 6.06% 6.00%
WOART 6.75% 11.75% 5.90% 9.45%
VALET 2.35% 5.60% 5.60% 3.60%
TAOT NA NA 5.33% 3.65%
NAROT 4.75% 7.30% 4.50% 4.50%
HAROT 3.15% 4.05% 3.33% 2.75%
USAOT 3.25% 3.00% 2.25% NA
Source: Moody’s Investors Service, Standard & Poor’s, 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 insecuritized 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
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 2009-
2011. 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 enhancementand 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
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%.
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 07-
1 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
Note: The performance is from Friday, September 2, 2011, close, to Thursday, September 8, 2011, close. Cash non-agency positions are marked on Wednesday andindices and cash CMBS positions are marked on Thursday. Source: Barclays Capital
20 Long PrimeX ARM.1 48 1.9 24 103.5 104.1 6/23/11
21 Long 2006 AM (duration hedged) 5.3 1 5.0 1 S + 580 S + 600 8/8/11
22 Long 2007 DUPER ( duration hedged) 4.9 1 5.0 1 S + 320 S + 375 8/8/11
Equity invested 100.4
Cash available 26.4
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, toThursday, 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
28 Long CMBX.AM2 vs short CMBX.AM.5 10 1 5/10/10 4.1 4/6/11 3.1 (464) (4.64)
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 atthe 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 markfor 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
Analyst Certification(s) We, Aaron Haan, Sandipan Deb, Ajay Rajadhyaksha, Nicholas Strand, Sandeep Bordia, Dennis Lee, Jasraj Vaidya, Keerthi Raghavan, Julia Tcherkassova, JosepAstorina, Derek Chen, Wei-Ang Lee and Siddarth Ramkumar, hereby certify (1) that the views expressed in this research report accurately reflect our personviews 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 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 researreport 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 thspecific 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 ResearCompliance, 745 Seventh Avenue, 17th Floor, New York, NY 10019 or refer to https://ecommerce.barcap.com/research/cgi-bin/all/disclosuresSearch.pl 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 Capmay 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 aaffiliate thereof (the "firm") regularly trades, generally deals as principal and generally provides liquidity (as market maker or otherwise) in the debt securitithat are the subject of this research report (and related derivatives thereof). The firm's proprietary trading accounts may have either a long and / or shoposition in such securities and / or derivative instruments, which may pose a conflict with the interests of investing customers. Where permitted and subjeto appropriate information barrier restrictions, the firm's fixed income research analysts regularly interact with its trading desk personnel to determine curreprices of fixed income securities. The firm's fixed income research analyst(s) receive compensation based on various factors including, but not limited to, tquality of their work, the overall performance of the firm (including the profitability of the investment banking department), the profitability and revenues the Fixed Income Division and the outstanding principal amount and trading value of, the profitability of, and the potential interest of the firms investinclients in research with respect to, the asset class covered by the analyst. To the extent that any historical pricing information was obtained from BarclaCapital trading desks, the firm makes no representation that it is accurate or complete. All levels, prices and spreads are historical and do not represecurrent market levels, prices or spreads, some or all of which may have changed since the publication of this document. Barclays Capital produces a variety research products including, but not limited to, fundamental analysis, equity-linked analysis, quantitative analysis, and trade ideas. Recommendatiocontained in one type of research product may differ from recommendations contained in other types of research products, whether as a result of differitime 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 LiquidAssessment Review. As such, Barclays Capital research on Allied Irish Banks plc and The Bank of Ireland is restricted and ratings and estimates suspendBarclays Capital is providing investment banking services to Ocwen Financial Corp in its potential acquisition of Litton Loan Servicing LP from GoldmSachs. .
DisclaimerThis publication has been prepared by Barclays Capital, the investment banking division of Barclays Bank PLC, and/or one or more of its affiliates as providebelow. It is provided to our clients for information purposes only, and Barclays Capital makes no express or implied warranties, and expressly disclaims warranties of merchantability or fitness for a particular purpose or use with respect to any data included in this publication. Barclays Capital will not treunauthorized 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 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, nany of their respective officers, directors, partners, or employees have any liability for (a) any special, punitive, indirect, or consequential damages; or (b) anlost profits, lost revenue, loss of anticipated savings or loss of opportunity or other financial loss, even if notified of the possibility of such damages, arisinfrom any use of this publication or its contents. Other than disclosures relating to Barclays Capital, the information contained in this publication has beeobtained from sources that Barclays Capital believes to be reliable, but Barclays Capital does not represent or warrant that it is accurate or complete. Tviews in this publication are those of Barclays Capital and are subject to change, and Barclays Capital has no obligation to update its opinions or tinformation 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 aother interests, including those of Barclays Capital and/or its affiliates. This publication does not constitute personal investment advice or take into accouthe individual financial circumstances or objectives of the clients who receive it. The securities discussed herein may not be suitable for all investors. BarclaCapital recommends that investors independently evaluate each issuer, security or instrument discussed herein and consult any independent advisors thbelieve necessary. The value of and income from any investment may fluctuate from day to day as a result of changes in relevant economic marke(including changes in market liquidity). The information herein is not intended to predict actual results, which may differ substantially from those reflectPast 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 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 whave professional experience in matters relating to investments. The investments to which it relates are available only to such persons and will be entered inonly 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 connectitherewith accepts responsibility for its contents. Any U.S. person wishing to effect a transaction in any security discussed herein should do so only bcontacting 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 regulationpermit 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 BarclayCapital policy on the dissemination of research reports, please go to http://www.barcap.com/Client+offering/Research/Research+Policy. Subject to tconditions of this publication as set out above, Absa Capital, the Investment Banking Division of Absa Bank Limited, an authorised financial services provid(Registration No.: 1986/004794/06), is distributing this material in South Africa. Absa Bank Limited is regulated by the South African Reserve Bank. Th
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, 3of 2002, or any other financial, investment, trading, tax, legal, accounting, retirement, actuarial or other professional advice or service whatsoever. Any SouAfrican person or entity wishing to effect a transaction in any security discussed herein should do so only by contacting a representative of Absa CapitalSouth Africa, 15 Alice Lane, Sandton, Johannesburg, Gauteng 2196. Absa Capital is an affiliate of Barclays Capital. In Japan, foreign exchange research repoare prepared and distributed by Barclays Bank PLC Tokyo Branch. Other research reports are distributed to institutional investors in Japan by Barclays Capi
Japan Limited. Barclays Capital Japan Limited is a joint-stock company incorporated in Japan with registered office of 6-10-1 Roppongi, Minato-ku, Tok106-6131, Japan. It is a subsidiary of Barclays Bank PLC and a registered financial instruments firm regulated by the Financial Services Agency of JapaRegistered Number: Kanto Zaimukyokucho (kinsho) No. 143. Barclays Bank PLC, Hong Kong Branch is distributing this material in Hong Kong as authorised institution regulated by the Hong Kong Monetary Authority. Registered Office: 41/F, Cheung Kong Center, 2 Queen's Road Central, Hong Kon
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 'recommendation' in Taiwan. Barclays Capital Securities Taiwan Limited does not accept orders from clients to trade in such securities. This material may nbe distributed to the public media or used by the public media without prior written consent of Barclays Capital. All equity research material is distributed India by Barclays Securities (India) Private Limited (SEBI Registration No: INB/INF 231292732 (NSE), INB/INF 011292738 (BSE), Registered Office: 208Ceejay House | Dr. Annie Besant Road | Shivsagar Estate | Worli | Mumbai - 400 018 | India, Phone: + 91 22 67196363). Other research reports are distributein India by Barclays Bank PLC, India Branch. Barclays Bank PLC Frankfurt Branch distributes this material in Germany under the supervision of Bundesanstafür Finanzdienstleistungsaufsicht (BaFin). This material is distributed in Malaysia by Barclays Capital Markets Malaysia Sdn Bhd. This material is distributedBrazil 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 Cent(Registered No. 0060) is regulated by the Dubai Financial Services Authority (DFSA). Barclays Bank PLC-DIFC Branch, may only undertake the financservices 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 andlicensed to conduct business activities as a branch of a commercial bank incorporated outside the UAE in Dubai (Licence No.: 13/1844/2008, RegisterOffice: Building No. 6, Burj Dubai Business Hub, Sheikh Zayed Road, Dubai City) and Abu Dhabi (Licence No.: 13/952/2008, Registered Office: Al JaziTowers, Hamdan Street, PO Box 2734, Abu Dhabi). Barclays Bank PLC in the Qatar Financial Centre (Registered No. 00018) is authorised by the QaFinancial Centre Regulatory Authority (QFCRA). Barclays Bank PLC-QFC Branch may only undertake the regulated activities that fall within the scope of existing QFCRA licence. Principal place of business in Qatar: Qatar Financial Centre, Office 1002, 10th Floor, QFC Tower, Diplomatic Area, West Bay, PO Bo15891, Doha, Qatar. This material is distributed in Dubai, the UAE and Qatar by Barclays Bank PLC. Related financial products or services are only available
Professional Clients as defined by the DFSA, and Business Customers as defined by the QFCRA. This material is distributed in Saudi Arabia by Barclays SauArabia ('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 placefuture. Barclays Saudi Arabia is a Closed Joint Stock Company, (CMA License No. 09141-37). Registered office Al Faisaliah Tower | Level 18 | Riyadh 1131Kingdom of Saudi Arabia. Authorised and regulated by the Capital Market Authority, Commercial Registration Number: 1010283024. This material distributed in Russia by OOO Barclays Capital, affiliated company of Barclays Bank PLC, registered and regulated in Russia by the FSFM. Broker License #1711850-100000; Dealer License #177-11855-010000. Registered address in Russia: 125047 Moscow, 1st Tverskaya-Yamskaya str. 21. This material distributed in Singapore by the Singapore branch of Barclays Bank PLC, a bank licensed in Singapore by the Monetary Authority of Singapore. For mattersconnection with this report, recipients in Singapore may contact the Singapore branch of Barclays Bank PLC, whose registered address is One Raffles QuaLevel 28, South Tower, Singapore 048583. Barclays Bank PLC, Australia Branch (ARBN 062 449 585, AFSL 246617) is distributing this material in Australia.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 construeto 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 bused, 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 thtransactions or other matters addressed herein. Accordingly, you should seek advice based on your particular circumstances from an independent tadvisor.