2008 db fixed income outlook (12!14!07)

Global

14 December 2007

Fixed Income Outlook 2008

Deutsche Bank Securities Inc.

All prices are those current at the end of the previous trading session unless otherwise indicated. Prices are sourced from local exchanges via Reuters, Bloomberg and other vendors. Data is sourced from Deutsche Bank and subject companies.

Deutsche Bank does and seeks to do business with companies covered in its research reports. Thus, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report.

Investors should consider this report as only a single factor in making their investment decision.

Independent, third-party research (IR) on certain companies covered by DBSI's research is available to customers of DBSI in the United States at no cost. Customers can access this IR at http://gm.db.com, or call 1-877-208-6300 to request that a copy of the IR be sent to them.

DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1

Compendium

Table of Contents Bond Market Strategy ...................................... Page 05US..................................................................... Page 09Euroland............................................................ Page 29UK..................................................................... Page 56EMEA ............................................................... Page 63Japan ................................................................ Page 72Asia................................................................... Page 76Dollar Bloc Strategy.......................................... Page 77Linkers .............................................................. Page 84

Global Views Level Slope Volatility USD Short Steeper NeutralEUR Neutral Steeper LongJPY Neutral Neutral NeutralGBP Neutral Steeper LongAUD Neutral Neutral NeutralCAD Neutral Steeper NeutralNZD Neutral Flatter Neutral

Research Team

Mustafa Chowdhury Research Analyst ( ) 212 250-7540 [email protected]

Ralf Preusser Strategist (+44) 20 754-52469 [email protected]

Francis Yared Strategist (+44) 020 754-54017 [email protected]

Fixe

d In

com

e G

lob

al M

arke

ts R

esea

rch

Fi

xed

Inco

me

Rel

ativ

e V

alu

e

The end of easy leverage: 2008, a year of rising risk premia

14 December 2007 Fixed Income Weekly


EXECUTIVE SUMMARY Reduced financial leverage, lower credit availability and negative wealth effects

should weigh on the consumer and growth, especially in the US

The good shape of the corporate sector, an aggressive Fed, a depreciating dollar and reasonable equity valuations should help steer the US economy away from a recession

Downside risks to this view remain, especially in light of the scale of the credit contraction but the current process of recognizing and writing-down impaired assets is a very positive step in ensuring that the current slow-down does not translate into a deflationary recession

Despite the uncertain economic outlook, the pricing inconsistencies between the US and EUR markets, the vulnerability of the UK economy and the strength of the factors supporting rising risk premia have helped us develop high conviction investment strategies for 2008

We like to enter the year positioned for higher risk premia globally, which will be driven by higher long term rates in the US, but lower policy rates in Euroland and the UK

We recommend curve steepeners in USD, GBP and EUR. We are underweight duration in the US (target 4.50% on 10Y UST), and are long the front-end in EUR and GBP. We are long EUR 5Yx5Y against USD 5Yx5Y

In USD we are buyers of tail volatility and like 2Y-10Y ASW disinversion trades

In EUR we are buyers of forward volatility and like payer flies on 1Yx2Y

For US Agencies, we do not expect the crucial OFHEO 30% capital surplus requirement to be relaxed in the first half of the year. If the 30% requirement is lifted in the second half, increased supply could put pressure on spreads

The EUR Jumbo covered bond market has started differentiating according to issuer specific (cover pool credit quality, issue structure, bank credit quality) and market specific topics (housing market risk, country risk). We expect that differentiation to continue in 2008

Three key global factors will drive rates returns in EMEA in 2008: high non-core inflation, the global growth slowdown and the continued effects of the credit crunch. We are bearish on the outlook for Poland and Hungary and bullish on Turkey and Israel. Over the course of 2008 we expect most rates curves in EMEA to steepen with the ZAR curve likely to disinvert by 100bp

In Japan, we recommend cutting remaining outright shorts. We like defensive butterfly positions, paying the belly in 5Y-7Y-10Y

We agree with market pricing for the RBA and RBNZ, while that for the BoC looks out of step with the strength of the economy. We find 10Y CAN expensive versus the US. We look for 10Y ACGB/UST spreads to compress

Inflation markets remain stuck between near-term price pressures from commodities and prospects of slowing growth, with the focus being skewed towards the former in Europe and the latter in the US. We see headline inflation moderating next year in both, but continue to believe that given differences in monetary policy and FX trends, current US/Europe spreads offer attractive opportunities in real rate and inflation space

The next regular Fixed Income Weekly will be published 11 January 2008. We wish all our clients and readers very Happy Holidays!



Global Trades

EUR Expressed with

bond/swap Entry date Closing Date Entry level

Target level at inception

Current/Closing Level

EUR - 5/10 steepener Bond 21-Aug-07 14bp 20bp

EUR - Long 2Y Bond 25-Sep-07 4.00% 4.03%

EUR - Receive 3M March Eonia Swap 25-Sep-07 4.02% 3.77% 4.16%

USD Expressed with




USD -2Y-10Y steepener Bonds 14-Sep-07 96bp 150bp 96bp

USD - Short 10Y Swap 14-Sep-07 5.07% 5.50% 4.85%

USD - 2/10 Swap spread Steepener Swap 21-Sep-07 2.5bp -22bp

GBP Expressed with




GBP - 2Y-10Y steepener Swap 26-Jul-07 -41bp -30bp -26bp

GBP - Long 2Y GIlts Bond 7-Dec-07 4.49% 4.62%

JPY Expressed with




JPY - YEH8- YEH9 Steepener 26-Oct-07 25bp 9.5bp

JPY - Long 10Y B/E Bond 25-May-07

JPY - Bull Steepener 3m fwd 4Y-20Y Swap 9-Nov-07 1.07% 1.13%

Cross market Expressed with




Buy Tips VsOATei(Tips 32 and OATei 32) Bond 10-Nov-06 -34bp -45bp -13bp

Receive EUR 5Yx5Y, pay USD 5Yx5Y Swap 14 Dec 07 54 bp 85 bp 54 bp

Asia Expressed with




INR - Rec 5Y Outright Swap 3-Aug-07 7.54% 7.25% 7.10%

HKD-2/10 Steepener Swap 23-Aug-07 44.5bp 80bp 56bp

SGD-2/10 Steepener Swap 23-Aug-07 39bp 120bp 69bp

$-Bloc Expressed with

bond/swap Entry date Closing Date Entry level Target level at

inception Current/Closing

Level

NZD - 6M fwd 1Y/2Y flattener Swap 10-Aug-07 -21bp -40bp -20bp

AUD - Receive 2020 BE Inflation Bond 17-Oct-07 360bp 330bp 365bp

AUD - 2Y/10Y AUD/USD basis Swap spread steepener Swap 14-Nov-07 -2bp 2bp -1bp

Receive 3Y Vs March 08 futures Swap 7-Dec-07 91bp 70bp 91bp

Receive body of AUD 2Y/5Y/10Y Butterfly Swap 29-Nov-07 36bp 10bp 31bp

EMEA Expressed with




PLN - sell 3X6 FRA FRAs 14-Dec-07 6.10% 5.60% 6.10%

PLN - long CPI-linked Aug-16, pay PLN 10Y bond/swap 14-Dec-07 300bp 350bp 300bp

HUF - 2Y-10Y steepener swap 14-Dec-07 -60bp -20bp -60bp

CZK - pay CZK 5Y, receive EUR 5Y swap 23-Nov-07 -15bp 30bp -26bp

SKK - pay SKK 2Y, receive EUR 2Y swap 26-Feb-07 -10bp 20bp -18bp

ZAR - pay ZAR 5Y swap 23-Nov-07 9.64% 10.00% 9.67%

TRY - long CPI-linked Feb-12 bond 2-Nov-07 109 113 112 Source: Deutsche Bank



Bond Market Strategy 2008

Reduced financial leverage, lower credit availability and negative wealth effects will weigh on the consumer and growth, especially in the US

The good shape of the corporate sector, an aggressive Fed, a depreciating dollar and reasonable equity valuations should help steer the US economy away from a recession

Downside risks to this view remain, especially in light of the scale of the credit contraction but the current process of recognizing and writing-down impaired assets is a very positive step in ensuring that the current slow-down does not translate into a deflationary recession

Despite the uncertain economic outlook, the pricing inconsistencies between the US and EUR markets, the vulnerability of the UK economy and the strength of the factors supporting rising risk premia have helped us develop high conviction investment strategies for 2008

We like to enter the year positioned for higher risk premia globally, which will be driven by higher long term rates in the US, but lower policy rates in Euroland and the UK

A year of rising risk premia

Tighter credit conditions will lead to higher risk premia and steeper curves The US economy is entering uncharted territory of credit rationing, reduction of financial leverage and falling house prices. The combination of lower credit availability and negative wealth effects will weigh on the US consumer and US growth. The good shape of the corporate sector, an aggressive Fed, a depreciating dollar and reasonable equity valuations should help steer the US economy away from a recession. However downside risks to this view remain, especially in light of the scale of the credit contraction.

Despite the uncertain economic outlook, the pricing inconsistencies between the US and EUR markets, the vulnerability of the UK economy and the strength of the factors supporting rising risk premia have helped us develop high conviction investment strategies for 2008. We like to enter the year positioned for higher risk premia globally which will be driven by higher long term rates in the US but lower policy rates in Euroland and the UK.

Painful deleveraging under way The second half of 2007 has turned-out to be the period when the excess leverage in the financial system finally began unwinding. Years of low real rates, recycling of large savings from emerging markets, loose lending standards and low economic volatility had compressed risk premia and led investors to increase leverage.

As corporate leverage was actually declining during most of the period, the appetite for debt and yield pick up had to be satisfied by other forms of debt creation: financial innovation (ABCP, CDO squared, SIVs etc. ) and increased consumer leverage (subprime in the US and mortgage markets in general).

The bursting of the US housing market bubble triggered the liquidity put implicit or explicit in several of the ABCP and SIV structures. The competition for banks’ balance sheet increased as previously off balance-sheet vehicles, warehoused loans (leveraged loans, new mortgage originations, CMBS etc.), and provision for losses started crowding out other lending activities. As a result lending standards in the US have started to tighten rapidly.

The tightening of credit standards has not been limited to the US market. Indeed, as evidenced by the latest ECB lending survey, credit standards tightened in Europe for both mortgage and corporate loans. In the UK, the tightening of lending criteria is also apparent through the sharp drop in mortgage approvals and is further exacerbated by the combination of an overextended mortgage market and the dependence of some institutions on wholesale funding. We believe that the credit rationing initiated in 2007 is likely to continue well into 2008.

Asset absorption more than half way through While a great deal of progress has been made in terms of absorbing assets on banks’ balance sheets (and in the case of Northern Rock, IKB etc.. public balance sheets), we estimate that around one-third of the SIV and ABCP conduits which included some form of credit term structure arbitrage still need to be absorbed by the system. Beyond year-end effects, liquidity is likely to remain strained until a solution for these assets has been engineered. In that respect, Wednesday’s coordinated effort from the world’s central banks is an important first step in the right direction. While the amounts involved remain modest relative to the size of the problem, one



can be hopeful that the revamped discount rate window will alleviate some of the funding concerns in the US and the UK.

Even if banks manage to find the short term liquidity to absorb the assets on balance sheet, they have yet to permanently resolve their funding issues. Indeed the maturity mismatch between the assets (average of 4.5 years for SIV for instance) and liabilities (mostly short term borrowing from central banks), will eventually need to be resolved through either longer term funding (long term debt issuance or increase in the deposit base) or a sale of the assets. While the asset/liability mismatch persists (i.e. until the market for these assets starts functioning again, or banks are able to issue long term debt in size), it is likely that the liquidity situation will not fully normalize.

But repairing of the capital bases will constrain lending Beyond the current liquidity crisis, there is little doubt that the capital for financial institutions is in need of serious repair. The combination of large losses (estimated anywhere between $200 and $400bn) and the accumulation of assets on balance sheet (leveraged loans, SIV, ABCP etc..) will require a considerable increase in capital. We note that several successful capital injections have been completed so far. However, the price at which these transactions have been completed is a reflection of the high insurance premium that financial institutions had to pay for emergency capital. These capital injections are therefore most likely only one part of a wider strategy to shore up capital and are likely to be supplemented by other means including a scaling down of lending activities. The process of rebuilding the capital base, recognizing the losses, funding the assets and restoring confidence is likely to curb lending activity in 2008.

For instance, the capital base of monoline insurers would be seriously eroded if rating agencies were to revise upward their assumptions for cumulative losses for 2005-2007 subprime vintages towards current analysts’ estimates. As highlighted by our securitization team, rating agencies have so far been relatively slow in adjusting their assumptions (and in downgrading subprime related securities). Thus, we should expect the pressures on the capital of monoline insurers to stay in place until the process of convergence between rating agencies assumptions and actual expected losses is completed. In the meantime, and as some of the monolines suggested, we should expect them to reduce their activities in order to improve their capital positions. This will erode one of the key elements in the creation of leverage in the structured finance market.

Similarly Spanish and UK banks are finding it increasingly difficult to fund new mortgage originations as the covered bond, securitisation and wholesale funding markets are all under stress. As a results, these institutions have had to increase the price/reduce the availability of mortgage loans which is likely to negatively impact frothy markets.

However, it is obvious that the current, though painful, process of recognizing and writing-down impaired assets is actually a very positive step in ensuring that the current slow-down does not translate into a Japanese-style deflationary recession.

Slowdown is (over)priced in the US: short duration

Neutral the front end Our analysis of the credit rationing outlook suggests that the current environment is more about availability of credit than price. As a result, the effectiveness of rate cuts will be reduced, in the same way as the aggressive lending practices in 2004-2006 delayed the effectiveness of the rate hikes.

This suggests that from a central bank perspective the solution to the current crisis should, together with rate cuts, involve providing more long term liquidity to the market, for instance by further expanding the TAF initiative. The remaining inflation risk in the US (see the discussion below), raises question marks over excessive Fed easing to cure the liquidity problem.

On the other hand, the macroeconomic impact of credit rationing and the downside risks to growth that it poses, justify further rate cuts. However, given that the corporate sector and equity market valuations enter the crisis in reasonable shape, and given the 10% depreciation of the USD trade weighted index year to date, it is difficult to argue at this point that there is scope for a much more aggressive Fed than what is already reflected in the forwards.

Thus, given the volatility of the front end of the curve and the uncertainty around the path that the Fed will choose to cure the liquidity crisis, we prefer to remain neutral on US front end pricing and assume for the rest of our analysis that the forwards will get realised.

But short the long end Assuming that the forwards in the front-end get realised, we recommend maintaining short duration position in the US for the following reasons.

The current level of bond risk premia does not reflect the current upside risks to inflation. Indeed, when the Fed initiated its rate cut process, it itself acknowledged inflationary pressures. This was in marked contrast to previous rate cut cycles, when the FOMC described



inflation as being contained. Rising inflation risks are also reflected in Consensus Forecasts’ long term inflation forecasts, which rose following the first rate cut (unlike previous rate cut cycles). In fact, the 10% depreciation of the trade weighted dollar over the last 12 month is already being reflected in higher import prices and should impact inflation in 2008. Finally, we note that in recent history, the only time when 10Y rates remained below 4% for a sustained period (with a minimum of about 3.1%) was in late 2002, early 2003 when the Fed Funds target rate was at 1% and the market was concerned about deflation risks. As mentioned above, we view the risks of a deflationary recession as limited, given the current process of loss recognition and bank recapitalization. In fact the current level of 10Y yields can only be explained by a sizable flight to quality premium which is evidenced by the widening spread between same maturity OIS and government bonds.

As highlighted in the US overview section, mortgage technicals are likely to add to steepening pressure on the long end of the curve. The duration of the mortgage universe will extend as the weaker housing market and tighter lending criteria lead to lower mortgage prepayment speeds and increased fixed rate originations.

Thus the combined force of rising inflation risk premia, mortgage technicals and a shift in asset allocation from sovereign wealth funds away from sovereign credit and towards equities (discussed below) should contribute to support long term rates.

Decoupling is (over)priced in Euroland: long the front end

Long the front end While the market is pricing another 100bp of cuts from the Fed and 50bp from the BoE, hawkish rhetoric from the ECB is leading the front end of the EUR curve to price a 40% chance of a hike. While Euroland’s economy is likely to be less impacted by the credit rationing than the US or UK economies, its growth outlook will nonetheless suffer both directly and indirectly from the credit crisis. We remain receivers of EONIA.

First, and as mentioned in the discussion above, the competition for balance sheet usage and tightening of lending criteria impacts financial institutions not only in the US but in Euroland and the UK as well. Second, as our economists have highlighted, it is unlikely that consumption growth in Asia will compensate for a slowdown of consumption in the US. Third, given the more proactive approach from the BoE and the Fed, the EUR is likely to face continued upside pressures. Further appreciation of the currency (or even consolidation at

current levels), will both help contain inflation which we expect to start falling in Q1 of next year (see Global Linkers update for more details), and also reduce the competitiveness of the Euro area. We note for instance that the good performance of the US economy in Q3 has been driven by exports, which is more than likely to occur at the expense of the Euroland economy given that global growth will be constant at best. Third, most of the key forward looking indicators in Euroland have started to turn indicating that the peak of the cycle has been reached already.

In fact, we would expect the ECB to become more explicit about rate cuts in Q1 of next year once (a) the important annual wage negotiations are completed, (b) inflation starts to subside as we expect and (c) the impact of the credit rationing becomes more obvious and the downward trend in the forward looking indicators gets confirmed by the data.

But neutral the long end The long end of the EUR curve will face conflicting forces that lead us to be neutral. On one hand, if we are right about the front end of the EUR curve, we would expect the 10Y rate to rally. This rally could be further supported by some potential shift in international reserve allocations towards Europe as currency pressures lead reserve fund managers to hedge their USD exposure. On the other hand, EUR rates are highly correlated to their US counterparts and react with a high beta to movements in USD rates. We therefore prefer to maintain a neutral duration view in Euroland.

Short USD 5Y5Y v. EUR 5Y5Y Being short the USD 5Y5Y vs. EUR 5Y5Y naturally falls out from our analysis. The analysis of the inflation and real rate components is also supportive of the trade. First, looking at the inflation component, we concur with the view reflected in Consensus Forecasts’ expectation of higher inflation risks in the US relative to Europe. This should support higher breakevens and steeper breakeven curves in the US. Second, looking at the real rate component, we note that long term forward real rates in the US have materially outperformed their EUR counterpart, so that the spread between the two is now negative (see Global Linkers for more details).

Risks in the UK skewed to the downside

High leverage… Analysing current events through the lens of an unwind of leverage highlights the vulnerability of the UK economy. Indeed, rising house prices towards unsustainable levels and low savings rates have pushed the UK consumer leverage to all time highs. Moreover, the UK economy



itself is heavily dependent on the service sector (75% of GDP) and especially the financial sector (10% of GDP). The financial and service sectors are clearly the most exposed to the current credit crisis as evidenced by the rapid fall of the services PMI.

…an over-extended housing market… The UK has benefited from an extremely rapid house price appreciation, even relative to the US. As a result, most house price models would tend to highlight the fact that UK houses are overvalued. This view would be confirmed by standard affordability measures such as house prices to earnings. The high level of house prices is often justified by supply/demand imbalances in the UK. We would tend to disagree with this view as a rental yield on buy-to-let properties below the mortgage rate is more likely a reflection of speculative behaviour and over-supply of rental properties.

…and tighter credit conditions will require aggressive rate cuts High libor fixings, dysfunctional covered bond and securitization markets have increased the funding pressures on UK mortgage originators. Mortgage lenders have naturally reacted by pulling back their lending criteria and/or increasing the cost of credit especially for buy-to-let and subprime mortgages. At this stage, only aggressive easing by the BoE could prevent a more brutal adjustment to house prices. We still see value in the UK front-end.

Risk premia set-to continue to rise: maintain steepeners

Our highest conviction trade for 2008 is that risk premia will continue to increase for several reasons.

First, the continued contraction in credit availability and the forced deleveraging of the system will naturally result in rising risk premia as asset prices adjust to lower levels of financial leverage and as risk is priced more conservatively when retained on balance sheet.

Second, central banks benefited from an exceptional macroeconomic environment during the last tightening cycle. The cycle started from an extremely low level of real rates while with the benefit of globalization, inflation levels remained subdued. Thus central banks had the opportunity to gradually normalize interest rates and the luxury to be exceptionally transparent in their forward looking statements. This very benign macro environment enabled central banks to be more predictable which reduced volatility in the front end of the curve and risk premia more generally. However, now that inflation is close to, or above central banks’ comfort zones and that

the trade-off between growth and inflation is more acute, central banks are bound to be more agnostic themselves about the direction of interest rates and therefore become less predictable. This will reintroduce volatility in the front end of the curve and support rising risk premia.

Third, unless a more permanent funding solution is found for the assets that moved back on banks’ balance sheet, we see still some risk of potentially disorderly liquidation of the assets underlying the SIV and ABCP structures.

Finally, and as we have argued in “The Unwind of the Bond Market Conundrum” (Fixed Income Special Report 20 June 2007) and the latest edition of Global Economic Perspectives, we believe that there are compelling opportunity costs, currency hedging and asset/liability management arguments for shifting SWF asset allocation to equities relative to fixed income. The rapid increase in reserves and the simultaneous improvement in the Equity Risk Premium have resulted in an increasing opportunity cost for holding fixed income securities rather than equities.

One can also make the argument that the increased allocation towards equities makes sense from an ALM perspective. As oil exporters have used early windfalls to significantly repair their balance sheets (as can be seen in the evolution of Russia’s external debt and Saudi’s domestic debt ratios) there are less ALM arguments to maintain a high asset allocation into debt. Finally, relative to Treasuries, investment in the equity of large US multinationals offers a good hedge against potential USD depreciation, since these will benefit at the margin from an improvement in US competitiveness.

We can already witness this trend in the behaviour of SWF money. Such a shift away from sovereign credit to equities should also contribute to an unwind of the bond market conundrum and a bear steepening of the curves especially in the US.

We therefore recommend maintaining steepeners in the US, UK and EUR and also selective long volatility positions (see the US and EUR derivatives sections).

Ralf Preusser (44) 20 7545 2469

Francis Yared (44) 20 7545 4017



US

US Overview

Repairing the breach We view there are two aspects of the current

crisis that need to be addressed: a capital crisis and a funding crisis. Government support and disclosure of losses are helping with the funding crisis, while the bank recapitalization is gradually resolving the capital crisis.

We also look at 4 drivers of the twin crises: mortgage supply, past Fed tightening, the fall in housing prices, and the hidden leverage of the securitization markets.

Our base scenario is for moderate GDP growth and 10% housing price declines. But the risk scenarios of recession or inflation raise the issue of increased systemic risks and high volatility.

We recommend being underweight duration for the base case, but across the scenarios, the constant theme is for implementing yield curve steepeners, entering swap spread curve disinversion trades, and buying out-of-the-money put and call volatility.

Capital vs. funding crisis In the current financial crisis, we see two distinct aspects of the crisis that need to be addressed in the coming year. The first is the bank capital crisis, rooted in anxiety about the effect of losses on bank (and other financial institutions’) balance sheets and their solvency. The second is the funding crisis, which is based on the lack of trust among banks due to the lack of transparency regarding losses. Each aspect has had some distinct origins as well as some overlapping ones, and will be solved with different tools, with the solutions occurring at different paces. For example, at one extreme there are the GSEs, which are funding at extremely good levels, while they are forced to raise capital to cover credit losses. On the other side of the coin are institutions such as Northern Rock that are on the verge of bankruptcy because they can’t find funding, while they have had little in the way of US subprime exposures and related losses. Most other financial institutions are somewhere in the middle of that spectrum.

LIBOR-OIS spread a measure of the funding crisis

0

20

40

60

80

100

120

Jan-07 Mar-07 May-07 Jul-07 Sep-07 Nov-07

Libor-OIS spread

Source: Deutsche Bank

Four underlying drivers There have been 4 fundamental factors that have interacted in various ways to cause these two crisis aspects. The first was the Fed tightening monetary policy beyond a likely neutral rate. The steady increase in the fed funds rate by 425 bp from 2004 through 2006 changed the economics of strategies that banks and consumers had adopted in the lower-rate environment. The second was the decline in housing prices and expectations of further declines. The Case-Shiller indices have already reported a 5% fall in home prices on a nationwide basis since mid-2006, and the housing-price futures market is pricing in another 8% drop over the next year. The third factor was hidden leverage caused by the creation of ABCP-financed SIVs, which were vulnerable due to an asset-liability mismatch. Total outstanding ABCP issuance had been nearly $1.2 trillion at the peak, but there is little likelihood that the ABCP-financed SIV business model will survive in its current form, given the exposure to correlated risks. The last, but not the least, was the pickup in volume of ARM resets and the resulting increase in fixed-rate mortgage supply. ARM resets and prepayments are likely to amount to $400 bn per quarter in 2008, adding substantially to both the mortgage duration supply and credit losses.

How did these drivers create the capital and funding crises? The sharp increase in mortgage supply kept consistent widening pressure on the mortgage market, forcing spreads wider. At the same time, this occurred against a



background of weak demand, as banks, which had been the primary buyers of mortgages since 2003, stopped buying. Low net interest margins, a result of the Fed tightening, led to no bid for the mortgage assets that were coming to the market. The resulting markdowns in mortgage prices started the capital crisis and contributed substantially to it.

Declining housing prices also contributed to the capital crisis, by leading to subprime CDO losses, first brought to light by the Bear Stearns hedge fund losses in June, and requirements to mark down the value of positions. The CDO losses were exacerbated by the increased market anticipation of forced selling or hedging of positions. The CDO losses brought rating agencies into the limelight, with the market questioning the models, correlation assumptions etc that the agencies have used to arrive at their rating decisions. These CDO losses, and anticipation of ratings downgrades, also created transparency issues, and a lack of trust among market participants, as CDO holders resisted marking their positions to market and disclosing the losses. This in turn led to a general unwillingness to lend, contributing to the funding crisis.

ABX price history

0

20

40

60

80

100

120


AAA (06-2)

BBB (06-2)


Finally, the recent structural innovations in the financial sector, particularly SIVs, and the hidden leverage that they represented, created a strong need to borrow when the traditional means of funding through ABCP was cut off. as first happened to IKB and Countrywide in August. The existence of such massive hidden leverage and ALM mismatches came as a surprise to the broader markets. So the reaction was equally massive, especially in LIBOR resets and swap spreads. As it became clear that there would be need for bank financial support or the shift of SIV assets to bank balance sheets, the spiraling bank funding crisis ensued.

Asset backed commercial paper outstanding

700

800

900

1000

1100

1200

1300


ABCP outstanding

Source: Federal Reserve

The methods of addressing these two aspects of the crisis have differed somewhat. For the capital crisis, recapitalization of banks (or other financial institutions such as GSEs or monoline insurers) was facilitated by the investments of sovereign wealth funds, or the public issuance of preferred stock in the public or private markets. However, the recapitalization activities were made possible first by a decline in bank equity prices, which enabled equity risk capital to enter the picture. So far, the interest in buying capital securities has been very strong as shown by the magnitude of oversubscription of GSE preferred issuances. In addition, Fed rate cuts lowered the anxiety level somewhat, kept the economy from sinking into a recession that would have further hurt the capital positions, and made it easier for banks to hold securities on their balance sheets.

In addition, the monoline insurers have began to enter into large reinsurance contracts, albeit at a high price, mitigating some of their risk.

For the funding crisis, government or quasi-government substitution for private funding went a good way toward restoring confidence. The Fed cut the fed funds target rate by 100 bp so far, and has added funds aggressively whenever shortages developed. In addition, the launch of the Term Auction Facility sought to overcome the stigma of the discount window and lend funds on a collateralized basis to a wide variety of banks. Another significant source of funding has been the Federal Home Loan Bank System, which has lent the banking system nearly an additional $200 bn in home loan advances during the crisis. At the same time the FHLBs themselves are funding at a very favorable rate. So, in a way, the FHLB advances served as a complement to the Fed discount window, which many banks are reluctant to use. Finally, the increase in transparency, through pre-announcing losses based on an actual mark-to-market value against indicators such as the ABX, might have caused substantial



pain and event risk in the short term, but in the longer term has been crucial in restoring confidence in the funding markets.

Scenarios for 2008 Going forward, our views for 2008 depend substantially on how the 4 factors evolve. Especially significant is the housing market and its effect on the economy. There are 3 possible scenarios. Our base scenario is one where real GDP grows at a modest, but below trend, pace of 2%, and housing prices fall by about 5% over the course of the year (adding up to 10% from the peak). The Fed continues to cut the fed funds rate to 3.75%, and ARM resets continue at a strong pace. This scenario would give plenty of room for the continuation of trends in deleveraging and capital replenishment for the coming year. The working out of the market crisis would involve the repricing of risk assets, but constrained by the relative scarcity of liquidity. The repricing of risk assets will occur as investors, primarily banks, seek to deleverage and reduce the balance sheet overhang by selling these assets, particularly in the mortgage sector, and recognize losses.

There will also be continued prospects of event risk early in the new year. We see two sources of event risk: Q4 earnings announcements that will force the laggards to recognize further losses on subprime portfolios, and rating agency downgrades that might force portfolio restructuring. As this process proceeds, we should see an improvement in transparency and market confidence, but at the same time, risk premia will widen given the amount of risk transfer that will have to take place. This sets the foundation for a recovery in the fixed income sector.

In this scenario, we would expect a mild bearish move to 4.50% on 10Y Treasuries, and curve steepening to 150 bp in 2/10Y Treasuries, as assets that had previously been financed through short-term funding such as ABCP or bank deposits instead are shifted to investors who fund through long-term debt or equity capital. This would put upward pressure on yield levels on the long end of the curve.

The next scenario is one with an especially bad outcome. The economy falls into recession, and housing prices fall steeply, approaching Depression-era levels. The credit crunch moves from the banks to consumers, causing a sharp drop in consumption, as both the capital and funding crises transition into the real economy. Global growth is also threatened with a slowdown or outright recession, with commodity prices falling and sovereign wealth investors becoming more conservative and restricting the equity investments into banks. In this case, we could see very low rate levels, with 3.50% on 10Y

Treasuries, 1% fed funds, and substantial curve steepening.

The third scenario is one where housing prices bottom out during 2008, perhaps falling 5% in the first half of the year, then rising 5% to end up unchanged for the year. GDP growth in this scenario returns to or moves above the long-term trend. In this scenario, inflation becomes a concern, with the Fed possibly starting to withdraw its recent rate cuts. However, this doesn’t mean a return to pre-crisis markets. The conundrum of low long-term rates does not return, nor does mortgage equity withdrawal revive. In this scenario, bearish steepening of the yield curve occurs on the back of higher inflation fears, with 10Y Treasury yields north of 5%.

Trade recommendations Pursuant to our base case, moderate growth scenario, we would implement a duration underweight. But across our scenarios, one consistent theme is the increasing steepness of the yield curve, either through rising risk premia, aggressive Fed rate cuts, or rising inflation premia. We thus favor 2/10Y curve steepeners.

We also favor buying out-of-the-money volatility, in both calls and puts, across the various scenarios. In our base case, the extension trade in mortgages, via the credit impact on slower refis, should be supportive for volatility. Lower rates in a recession, if prolonged, and accompanied by substantial mortgage origination in new coupons, could set up the market for another convexity hedging episode. Finally, a steepening sell-off and higher inflation in the third scenario could expose the market to substantial shocks in adjusting to the new market environment. Thus, we favor buying volatility on the tails of the distribution.

We also think that the present inversion of swap spreads curve has been primarily driven by the funding crisis and consequent LIBOR resets. The funding crisis, in our view, will eventually dissipate because of central bank actions and more transparency. The transparency will improve, either because of more disclosure of mark to market losses by financial institutions or more data on actual subprime defaults. Thus we target disinversion of the spread curve. With regard to spread level, we see 10Y swap spreads narrowing only modestly, to 60 bp, as narrower spreads in the short end and increased Treasury supply, are offset by poorly-performing mortgage markets, and wider risk premia among spread product.

Mustafa Chowdhury (1) 212 250-7540 Marcus Huie (1) 212 250-8356

14 Decem

ber 2007

Fixed Income W

eekly

Deutsche B

ank Securities Inc.

The flow-through effects of the credit crisis

ARM resets andmortgage supply

Fed tightening beyondneutral rate

Declining housingmarket and

expectations

Hidden leveragecreated thru SIVs

Markdown

Balance sheet anxietyand capital crisis

Transparency issueand lack of trust

Unwillingnessto lend

Funding crisis

Supply

Low NIM and no bidfor assets

Leveraged loan and CDO losses, markdowns

Strong needto borrow

More transparencyand disclosures

Fed cuts andTAF program

FHLB advances

Fed cuts

SWFs

Cheapequity




Cross Rates Strategy

2008 - Mortgage Extension & ARM Resets – Bear steepening Swap spread slope disinverting and bullish for vol

We expect prepayment related mortgage extension to provide a bear steepening pressure on rates, widening pressure on swap spreads with a steeper swap spread slope and structural support for vol in 2008.

Lending standards are likely to tighten and the housing market, worsen further. Mortgage extension due to lower prepays in 2008 could add about $450bn in 10Y equivalents with most of the increase focused on the long end of the curve. The 10Y swap spread is likely to widen by 4bp due to these effects

ARM resets are likely to add to the same exposures as well. We estimate $ 100bn in 10Y eqv. added mostly on the long end with the 10Y swap spread likely to be another 2-3 bp wider due to this issue.

Call risk should continue to trend lower, and extension risk higher. Structural preference for payer skew over receiver skew should thus continue.

The initial part of the year could see high spread volatility of mortgages, as views on the basis remain varied (valuation vs supply/demand technicals). Mortgage options should thus reflect higher vol.

2007 saw considerable turmoil in the mortgage markets that eventually led to a broader credit and liquidity crisis, forcing the Fed to cut rates to forestall slowing of the overall economy. Volatility came back after being initially range bound, with a structural bid re-emerging from mortgage players, due to ARM to fixed refinancing and mortgage extension. In 2008, we expect the constriction in mortgage credit and available products and its interplay with the housing market to further extend mortgages due to lower prepays, aid ARM to fixed refinancing, as well as influence the general economic volatility.

Duration and Vega: Turnover vs Refinancing effects

We have previously discussed the impact of slower prepayments on duration (partial and total) and vega of mortgages, and the resulting effects on rates and vol. We now examine the effects of turnover and refinancing separately.

If turnover slows without any change in refinancebility, then effectively you are extending the expiry of the call option held by the borrower due to longer duration of the mortgage, while the sensitivity of that option to rates remains about the same, increasing the vega. In the case of refinancing however, there are two competing pressures. If refinancing reduces, the sensitivity to rates reduces, but the expiry of the option increases as duration extends. As we show below, while the duration is affected by both factors, vega is a function of turnover, as the competing pressures in the case of refinancing net each other out.

Duration Exposure: Refi and Turnover both contribute Base Duration

-20% Refi & Turnover -20% Refi -20% Turnover

5 4.97 0.35 0.08 0.26 5.5 4.04 0.37 0.12 0.24

6 2.95 0.44 0.21 0.22 6.5 1.81 0.54 0.30 0.23

Changes

Source: Deutsche Bankg

Vega Exposure: Turnover dominates Refi Base Vega

-20% Refi & Turnover -20% Refi -20% Turnover

5 (23.1) (2.9) 0.6 (3.6) 5.5 (24.8) (2.8) 0.4 (3.3)

6 (24.3) (2.7) 0.2 (2.9) 6.5 (17.3) (2.5) (0.1) (2.3)

Changes


Cash-out refinancing is a function of home equity buildup, which is closely related to turnover as well as lending standards, which are currently influencing premium prepayments considerably.

Turnover continues to worsen: The Pending Home Sales figure was much worse than it looks

While the market focused on the stronger seasonally adjusted Pending Home Sales figure, in reality, the weakness of the report lay in the non seasonally adjusted figure, which increased 7%. In 2007, October had 22.5 business days, while September 19. Ignoring the seasonal decrease in turnover, day count in isolation should have caused an 18.5% increase in home sales. Last year, on a day count adjusted basis, the index fell about 5%, while in 2005 it was unchanged, adjusted for day count. Even after using a 5% decline based on 2006 numbers, the net shortfall in the index amounts to: 18.5-7-5 = 6.5%. YOY this figure has fallen 21%., daycount adjusted

and as does the Refi atmosphere



The Refi index has consistently overestimated the extent of premium coupon prepayments due to increasing ARM resets, multiple applications from borrowers and lower approval rate due to tighter lending standards. Moreover actual refi index prints continue to increase in a benign fashion. For a 14 bp rally in mortgage rates, the Dec 7th print of the Refi index showed only a 4.3% increase. Premium prepayment response to decreasing rates is thus quite muted.

2008 Outlook

In our view, lending standards should continue to tighten, despite better liquidity as delinquencies continue to rise and also due to political initiatives. Lenders also continue to scale back or discontinue their subprime origination business. The housing market is also likely to worsen. Moreover, the tightening in lending standards and constriction in mortgage products also impact turnover indirectly by considerably shrinking the buyer pool for a particular home.

During 2008, a 20% reduction in turnover and refinancing speeds from current levels (where we already dial the models -20%) is a reasonable base case scenario. In a rally, the disparity between expected duration (and vega) and actual is likely to widen substantially as prepays surprise on the lower side. Note that indices may not have fully adjusted to current lower prepayments and thus the index managed community could be longer duration.

Changes in duration (total and partial) and vega of

FRMs and ARMs OAD Vega

2Y 5Y 10Y 20Y5 0.54 (0.05) (0.03) 0.08 0.55 (2.1)

5.5 0.53 (0.04) 0.01 0.11 0.47 (2.6) 6 0.53 (0.01) 0.04 0.13 0.40 (3.2)

6.5 0.55 0.02 0.09 0.17 0.30 (3.2)

ARM 0.15 0.03 0.07 0.04 0.04 (0.7)

Partial Durations


Assuming a 60/40 split between fixed and ARMs, with $10 trillion mortgage debt outstanding, we expect the duration extension in 2008 due to prepayments to be of the order of $450 bn 10Y equivalents, if rates remain at current levels. A selloff could intensify the housing market downturn and further exacerbate the prepay related extension. As can be seen in the above figure, the extension is a considerable curve steepener. The extension is likely to influence swap spreads wider as well, with an additional 4 bps or so in 10Y swap spread, based on the relationship between the agency MBS DV01 and swap spreads. While the vega increase in absolute terms for ARMs is not that high, in percentage terms the vega exposure increases about 21% versus 11% for fixed

rates. From a servicer perspective, the effects are even larger as the duration of the IO becomes less negative and vega higher (more negative).

As we discuss in the section on ARM resets, these effects are compounded by the ARM to fixed refinancing.

If turnover and refinancing deviate to some extent, the effects can be determined from the discussion in the beginning of the article. One can also use the elbow shifts in models to change the refi incentive to obtain a dynamic extension picture in rallies.

Call risk should thus continue to be benign and extension risk significant in 2008. We continue to look for violent selloffs, due to convexity related flows in backups. In the first half of the year, as the market focus shifts away from immediate liquidity and P&L concerns, we could see full adjustment of MBS durations across the investor spectrum. Already, models have begun to release new versions, particularly due to the enormous variation in expected vs model prepays following the breach of the 5.5% threshold in the current coupon rates in early December. Buying of MBS in rallies to make up duration should continue to be muted, as prepays do not increase as much. Structural bid for low strike receivers is thus likely to be benign as well.

We also like the long vol optionality offered by PACs versus MBS as the vega becomes longer as prepays slow further. Investors can sell this volatility through swaption straddles or low strike receivers for example, to monetize some of this exposure. We favor curve steepeners and recommend bear steepeners as a support hedge. Other instruments such as VADM/Supports and Reference Remics are an excellent asset/liability management vehicle.

ARM Resets:

In the figures below, we show the history and schedule of ARM resets by quarter.

ARM resets by quarter

Resets with prepayment projections

0

50

100

150

200

250

300

350

400

450

07Q4 08Q4 09Q4 10Q4 11Q4 12Q4

$b

n

Prepayment

First Reset

Source: Deutsche Bank, eMBS, Loan Performance



ARMs as a percentage of total applications continue to reduce, both in $ and numerical terms. Moreover the White House proposal to aid subprime borrowers points to the increasing channeling into fixed rate mortgages.

ARMs as a percentage of total applications – back to

pre mortgage product expansion levels

0

5

10

15

20

25

30

35

40

Jan-

02

Jul-0

2

Jan-

03

Jul-0

3

Jan-

04

Jul-0

4

Jan-

05

Jul-0

5

Jan-

06

Jul-0

6

Jan-

07

Jul-0

7

Source: Bloomberg

Risk premium on ARMs continues to increase. The fixed to ARM spread has continued to shrink despite steepening of the yield curve and this spread could lag further steepening of the yield curve in 2008. Borrower demand for this product is also likely to be lower, due to increased aversion of related risks.

ARM to fixed – net duration and vega effects (dialing

20% lower prepays for both products) OAD Vega

2Y 5Y 10Y 20YARM 1.96 1.03 0.59 0.08 0.03 (0.03) Fixed (6%) 3.48 0.62 0.75 0.71 1.17 (0.27) Net 1.51 (0.41) 0.16 0.63 1.14 (0.24)

Partial Durations


Assuming a 30% conversion from ARM to fixed, the net duration added to the 30Y universe in 2008 is about $100 bn in 10Y equivalents, most of it on the long end of the curve. This has the effect of about 2-3 bp widening in 10Y swap spreads as well.

As can be seen in the figure, considerable vega is added through the transaction. A similar trend occurs in the IO space as well, relevant to the vol needs of servicers. Vega exposure of the total market may increase about 7-10% because of this effect.

We expect mortgages to provide a bear steepening pressure on rates, widening pressure on swap spreads with a steeper swap spread slope and structural support for vol in 2008.

Mortgage Options:

The hedge ratio implied by mortgage options continues to remain high as compared to that implied by the pass-throughs, indicating high MBS spread volatility.

Hedge ratio comparison between TBA and mortgage

options for the 6% cpn

20%

25%

30%

35%

40%

45%

50%

55%

60%

65%

70%

6/7 6/21 7/5 7/1

9 8/2 8/16

8/30

9/13

9/27

10/11

10/25 11

/811

/22 12/6

Option HR TBA HR


Moreover, views on the mortgage basis remain diverse as seen in the Mellon survey of money managers. Essentially the carry of mortgages is quite attractive but supply continues to overwhelm demand. Money managers are considerably overweight mortgages, and banks have been buying whole loans in favor of MBS.

25th vs 75th percentile money manager MBS

overweight

579

1113151719212325

May-03

Nov-03

May-04

Nov-04

May-05

Nov-05

May-06

Nov-06

May-07

Nov-07

25th vs 75th

Source: Mellon Survey

We thus recommend against selling mortgage vol vs swaption vol until these views are reconciled, which may happen by mid-2008.

Anish Lohokare (1) 212 250 2147



Treasuries

Increase in deficit and issuance expected 2008 is likely to be the end of the 3 year trend

toward narrower deficits, as reduced fiscal discipline, greater willingness to engage in fiscal stimulus, and the slower economy pressure the budget deficit wider.

We think the additional Treasury issuance will be primarily done through the Treasury bill market. Thus we are likely to see the bill share of total issuance rise.

As the Treasury repo market normalizes, we should see Treasury futures richen relative to the CTD, and on-the-run spreads start to narrow back.

The federal budget deficit is likely to exceed the estimates of the Congressional Budget Office due to higher-than-assumed cost of the operations in Iraq and Afghanistan, the AMT provision, and possibly higher non-discretionary spending and lower revenue. These point to increased Treasury issuance in 2008. Our forecast for the deficit for FY 2008 is $250 bn, which is likely to be skewed upwards as the economy weakens.

Recent improvement in federal deficit likely to reverse

in 2008

-500

-400

-300

-200

-100

0

100

200

300

400

500

97 99 01 03 05 07

-500

-400

-300

-200

-100

0

100

200

300

400

500

Deficit - LHS

Net marketable coupon issuance - RHS

Our 2008projection


The AMT fix adds about $50 bn in additional deficit for FY 2008. On the war funding front, the Administration’s revised request was $43 bn higher than the original estimate. Moreover, there is also a chance of additional funding requests on this front later in 2008. The early months of FY 2008 give some indication of this: Both the October and November 2007 outlays for defense, social security, Medicare/Medicaid, Social Security and net interest on public debt increased considerably over the previous year (adjusted for business days and

seasonality). The total spending was up 6% from Nov 06 to Nov 07, with net interest on public debt soaring 12.6% YOY. Overall, the Nov deficit widened slightly beyond economist expectations.

Also, fiscal discipline is under threat, given the political dynamics in Congress. The recent passage of the AMT provision in the US Senate, in absence of an offset under the “Pay-as-you-go” principle advocated by the Democratic leadership earlier this year, is perhaps an indication of the continued impasse between the opposing parties. Once this principle is abandoned, the likelihood of additional populist legislation increases, particularly due to the proximity of the US presidential elections.

The revenues side of the equation could also fall lower than the CBO estimates. While early FY 2008 data indicates healthy individual income tax receipts, the year-end bonus payments for individuals are likely to be on the weaker side. Moreover, overall individual tax receipts for the year could decline, particularly if the weaker economy exerts pressure on wages and/or employment. Employment taxes could decrease in proportion to income taxes, further exacerbating the deficit situation.

The CBO reported a drop in corporate income tax receipts for the first two months of FY 2008. While the first two months may not set the course for the entire year, in the face of weak retail sales, poor financial company earnings and also general weaker economic and credit conditions, we find it hard for corporate income taxes to be stellar, despite the weakening of the dollar. Corporate spreads have widened considerably and stocks have declined and served as an expectation of lower earnings, at least in the near future, until the Fed cuts take effect. This also bolsters the possibility of wage and employment pressures.

The net note and bond issuance for the first half of the year is negative and points to increased issuance sizes. Given the considerable increase in the debt burden, the Treasury is likely to increase issuance that minimizes the net interest cost and thus T-bill issuance should show a considerable increase, followed by modest increases in the 2Y and 5Y .auction sizes.



Net Note and Bond issuance expectations for Jan-Jun

2008 Treasury Notes and Bonds Gross Issuance Maturing Net Issuance Coupon Net Supply2Y 124 132 -8 10 -183Y 0 44 -44 4 -485Y 82 57 25 15 1010Y 46 35 11 17 -620Y 0 0 0 0 030Y 15 0 15 17 -2Callables 0 0 0 2 -2TIPS 29 16 13 6 7Notes/Bonds 296 284 12 71 -59

Source: Wrightson

Share of bills as pctage of total debt is likely to

increase

18%

20%

22%

24%

26%

28%

30%

01 02 03 04 05 06 07


The breakdown in the financing markets have contributed to a widening of on-the-run spreads. With the recent reduction in the financing crisis, the liquidity premium for on-the-runs could start to decline, compressing spline spreads and also lead to narrower quoted swap spreads. In addition, the cheapness of the back Treasury futures contracts should correct in the improving environment, leading a long futures position to outperform cash.

Spline spread of on-the-run 10Y Treasury

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

Apr-07 Jun-07 Aug-07 Oct-07 Dec-07


10Y futures CTD net basis

-4

-2

0

2

4

6

8

10

12

14

16

Jun-07 Jul-07 Aug-07 Sep-07 Oct-07 Nov-07 Dec-07

H8 Z8


Marcus Huie (1) 212 250-2147

Anish Lohokare (1) 212 250-8356



Derivatives

We see developments in 2008 as the conundrum in reverse with a sticky reaction to the Fed cuts.

Our base case scenario is a moderate economic slowdown with at least a partial resolution of liquidity problems in the first two quarters of the year.

We are structurally bullish on vol due to mortgage related factors and would look to take advantage of the liquidity induced dislocations in an RV context.

The conditions of forced capitalization which are behind the recent trust preferred issuance are likely to reduce the swapping activity and supply of vega to the market.

2008: The conundrum in reverse?

2007 saw a big comeback of volatility after several years of continued decline. In the second half of the year we saw the unraveling of the housing bubble and what could be a beginning of a log U-turn in rates and credit markets. The year started with one of the largest drops in implied volatilities with both indices hitting their 10Y lows after a supply shock at the end of 2006. This was followed by a relatively quiet period with occasional episodes of convexity hedging, but without major excitements pushing vol even lower in May until the first indications of sub prime crises showed up during the summer. The figure shows the recent history of the two indices.

Volatility indices reach 3-year highs

17-May

12-Sep

11-Dec

15-Feb

50

60

70

80

90

100

110

120

130

140

04 05 06 07

60

70

80

90

100

110

120DGX (left)

DVX (right)


As the housing market slowdown began to take its toll, we saw a gradual return of mortgage related demand, while high realized volatility provided support for gamma. Although liquidity problems never disappeared, the

immediate reaction and positioning which emerged after a long period of low volatility and complacent markets soon began to reflect concerns abut he long term impact of the housing market on the economy. As rates trended lower in a flight to quality mode, different trades kicked in at the same time across all sectors of the curve leading to fragmentation and breakdown of long-term relationships across different market and curve sectors which further impaired already fragile liquidity.

The big picture We believe that the first two quarters of the next year will revolve around the resolution of liquidity and the direction of the economy. Although the two issues had the same origin, each gained life of its own. We see the economic drivers as the long-term agenda and liquidity as transient. Our base case scenario is a moderate economic slowdown with gradual improvements of liquidity.

How we got here? The last tightening cycle was in many respects different than any previous episode in recent history. In order to match increasing future liabilities, investors took higher and higher risks across different products leading to compression of the risk premia together with its manifestations through curve flattening and spread tightening across all market sectors. With the entire economy riding the real estate boom, creative mortgage lending resulted in proliferation of instruments which pushed the effect of rate hikes further into the future. Although the Fed funds target rate continued to rise, the relevant rate levels (e.g. 10Y, swaps, mortgages, …) did not follow in the same manner. As a result, effectively the Fed had not tightened to the consumer despite 425bp of rate hikes. The 10Y rate, for example, remained near its levels in the middle of the cycle, Figure below. At the end of the tightening cycle, there was about 150-200 bp of rate cuts missing from the long end of the curve.



10-year swaps vs Fed funds target rate

3

4

5

6

7

8

9

93 94 95 96 97 98 99 00 01 02 03 04 05 06 07

0

1

2

3

4

5

6

7

8FFT

10Y


Similar effect propagated across different sectors of the interest rates market. Most notably, the expansion of the real estate market brought in buyers at the long end pushing spreads tighter and reducing the carry across the board as risk premia compressed. The Figure below shows the mortgage carry (CC mortgage vs. 5Y swaps rate) across different Fed cycles. The decline in the spread during 2004-2006 hikes was largely triggered by the low levels of Fed funds which catalyzed subsequent growth of the real estate market.

Mortgage carry across Fed cycles

50

100

150

200

250

300

99 00 01 02 03 04 05 06 07

0

1

2

3

4

5

6

7

Carry (Primary Mtg - 5Y Swap Rate)

FFTR


Despite continued hikes, mortgages remained in demand keeping the spreads and carry low. Although vol declined, it remained out of sight for the hedgers both due to structural as well as carry reasons.

What to expect in 2008? During the conundrum days, reaction of the market to the Fed remained sticky and delayed. We believe that the next year would be a replay of the conundrum in reverse with the same type of reaction function likely due to eroded trust and confidence. The ability of the Fed to

control rates relevant for the consumer by easing is questionable in these market conditions. As Treasury rates follow the Fed, the actual lending rates, e.g. mortgage, swaps etc., could remain largely unaffected by this maneuver with spreads wide due to the same stickiness as during the conundrum. Lack of transparency regarding the banks’ balance sheets would affect their ability to lend, while credit availability would continue to contract. With high cost of capital, the lending rates should remain high which, in turn, would make it difficult for people to borrow. Fed would cut rates substantially, but the 10Y rate, for example, could remain high. This is the high risk premium environment. As economy is deleveraging, the consumption could gradually slow down.

During that time, we would see periods of high realized volatility due to substantial repricing as the market digests new information. In that environment, there would be little consensus and wide divergence in terms of positioning. Generally, such markets are very volatile with substantial reaction to information shocks and large swings in rates due to repricing. Together with higher risk premia next year should be supportive for gamma. Even a favorable resolution of the liquidity problems in the economy with moderate slowdown is likely to lead to high volatility. An improvement in liquidity, if it happens, would be followed by a period of large swings in rates like those seen in November and December as the cumulative effect of monetary policy kicks all at once after a prolonged delay.

Risk scenarios We see the following two departures from the base case as the likely risk scenarios.

1) Sub prime fallout has a deeper impact on the economy which would enter into a recession with consumer credit crunch together with a further escalation of capital and funding crisis. Fed would continue to cut with the curve bull steepening while the long end remains sticky. As rates decline, there is a new production of low-coupon mortgages (5s and 4.5s), which would increase the convexity. The deterioration of the credit and liquidity would make it only a partial replay of the early years of the decade without a massive participation of the US consumers in the real estate growth. This environment would be highly bullish for vol both fundamentally and structurally, although we would not see a return of the golden days of convexity hedging.

2) On the other side of the spectrum, we have possibility of a less substantial impact on the economy, but continued rate cuts forced by liquidity issues, which could produce higher inflation as a side effect. As the curve prices out the Fed cuts, we might see initial flattening,



followed by the long-term steepening pressures. This type of market would be close to neutral on vol, although we do not expect vol to decline substantially due to accumulation of various risks in the market.

Structural demand for vol

We are long-term bullish on vol as the current structural changes in the market point out to return of demand. The housing market slowdown has changed the convexity exposure of mortgages considerably. As it became increasingly more difficult to refinance, gamma has been pushed towards lower rates, while optionality extended leading to more negative vega. The figure below shows the snapshot of convexity profile of the mortgage universe at the end of 2007.

Dec-11 snapshot of convexity profile of the mortgage

universe

-0.3

-0.25

-0.2

-0.15

-0.1

-0.05

0

-210

-180

-150

-120

-90

-60

-30 0 30 60 90 12

015

018

021

0

-3

-2.5

-2

-1.5

-1

-0.5

0

0.5

Vega OAC - RHS


We are currently about 60bp above the peak convexity. Although gamma has declined since the beginning of the year, vega increased as the optionality extended with the peak moving towards higher rates.

The main effect of the housing slowdown is coming from decline in turnover. This is going to cause a long-lasting effect on servicers and mortgage hedgers in general. The extension of the mortgage portfolios both in terms of vega as well as duration is making hedgers longer than initially anticipated. In the next year, this trend will continue and we expect to see a continued need to shed duration in a sell off which would make the corresponding episodes more volatile than we are used to seeing. On the other side, the call risk remains benign. In that environment we would prefer payers to receivers as rate protection instruments.

The transition from ARMs to fixed rate mortgages is also supportive of both steepeners and higher vol, as we move from less to more convex instruments. In terms of duration hedging, this transition would cause an increase in paying beyond 10Y and unwind of hedges at the short end (receiving in 5s and 2s). Keeping away from the liquidity contaminated short end, the best bet in this context are 5s/10s steepeners, both outright and conditional over 1Y or longer horizon.

As mortgages widened in the second half of the year, their carry increased. Generally, hedging decisions are made in relationship to the carry. The Figure below shows the mortgage carry (adjusted for the scale) against 1Y10Y vol. Their long term relationship is a result of an RV twists introduced by mortgage hedgers – in the hedging process they are constantly arbing between the curve and vol. For a given return they decide how much of carry they can give up in order to reduce the variations of their portfolio by hedging their convexity exposure. Whenever carry is high, potentially high hedging costs can be justified. This brings in the hedgers to the vol market and maintains a bid for vega. Conversely, when carry is low, hedging becomes too costly and demand for vol declines. Clearly, return of carry is supportive for vol as it is likely to intensify convexity hedging which had been reduced to a minimum during the conundrum days.

Comparing the mortgage carry with vol

60

70

80

90

100

110

120

130

140

150

160

02 03 04 05 06 07

Nominal Spd vs Swaps

1y10y


We expect servicers to be the main players in this process, while the rest of the hedging community might be less active in the fist half of 2008. Due to its focus on capital issues they might stay away from the MTM exposure of derivatives with more attention on accounting than actual hedging. So, while vol levels might remain high due to buildup of risk in the market, we might see somewhat lower vol of vol.



Supply

While demand for vol continues to grow, we had, in general a declining trend in vol supply both in term of callable as well as the trust preferred issuance. In the past, most of the callable structures were matched by the 10Y intermediate vol. As the log end of the curve flattened FHLB, which has been one of the main suppliers of vol in this sector, switched to shorter tenors and expiries with vol exposure corresponding to the upper left corner. As a consequence, vega dropped considerably since than. In the fist three quarters of 2007, we have the callable supply below 2005 and 2006 levels as we saw a large wave of redemptions with declining rates and the curve inversion at the short end. Due to declining optionality, there has been a dramatic drop in issuance of the same structures as the inverted curve makes optionality less valuable and the corresponding coupon less attractive for the investors. In 2008 we could see a continuation of this trend if the curve does not steepen materially at the short end, although persistence of these conditions and further steepening of the swaps curve at the long end might create an environment where issuing 10NC1-type structures would amount to a pickup of the vega supply. We believe that in this environment where investors are not yield-hungry, this transition might take some time to kick in.

Due to their longer maturity trust preferreds have had a much bigger impact on the vega supply. However, in the current market environment created by the sub-prime crisis, they are also likely to have less of an impact on the vol supply in the next year. Recent issuance of preferred securities has been driven by forced capitalization with pressure to raise tier 1 capital. Most of the banks maxed out on their tier 1 capital (limit to 15%). Currently, there is a room for about $5bn of new trust preferreds among the top 10 banks. Given their typical structure of 60NC5, their vega is matched roughly by 10Y20Y, around 4mn/$1bn) which would result in 10-20mn of potential vega supply, if swapped. As balance sheets grow and capital ratios improve, banks might issue more to repair their capital.

Because of the conditions under which they were issued, these securities have had high coupons and less favorable structure for the issuer than is generally seen in the times of opportunistic issuance. The embedded optionlity (call feature) in this case contains both a call on rate as well as a put on credit. Given this, it is less likely that the optionality of these issues would be swapped with the street, as they have a substantial value once the market conditions normalize, the credit ratings improve and especially if the Fed continues to cut rates. Because of this, we see this wave of issuance having less of an effect

on the vol market in terms of supply pressure at the long end of the surface.

Who will be the vol sellers? In the last year we have seen an increasing participation of the real money in the options market which came in mostly as vol sellers with different variants of short strangle strategies as a way to play the range bound markets, or improve carry on short mortgage positions. Although, they might continue their presence in this market, the change of the dynamics is likely to reduce their impact through these types of trades. However, they could come in as vol sellers in a different way. As the credit related losses are making investors look for yield in alternative products, as replacement for their high yield exposure convexity is arising as an alternative. If this trend picks up and various structured notes replace the conventional credit products, real money would emerge as vol sellers in this context. Although this is still a possibility, the effect of such a transition would not materialize in the near term.

Between 2004 and 2007 specs have emerged as the largest suppliers of vol in the market. As they grew in size and number, their impact overpowered the declining demand since they became core sellers of vol during the conundrum years. In the current environment it is difficult to see RV players coming in with similar strategies while realized volatility remains at such levels with continued prospects of large swings and structural breaks. It is likely that they will look for carry away from short gamma and in different sectors of the surface. Supply outlook might bring in vol sellers at the long end while intermediates remain well bid due to mortgage related demand and higher realized vol. RV players are likely to be a regulating factor in this sector. As the gap between the regions of the surface widens, they are likely to take advantage of the favorable vol rollup and sponsor the long dated sector.

Dislocations

As the first wave of sub-prime shocks hit the market, the defensive strategies and focus on the first order risk left behind a number of the dislocations across different market sectors, which persisted through the last quarter of the year awaiting the resolution of liquidity before the RV players come in and take advantage of these inconsistencies. Part of the reason for the delayed reaction of the market has been the compromised integrity of the curve triggered by the market segmentation. As a consequence of the breakdown of the long-term relationships between various market sectors, risk management became difficult increasing the friction and causing the dealers to further withdraw liquidity from



the market. The persistent spread widening at the short end has lead to an increased widening of the condor spreads along the swaps curve. Typically, 2s/5s/10s/30s condor spreads show high degree of stability as the number of legs allows an optimization procedure which accounts for the bulk of the risks contained in the curve with standard deviation of only a few bp. Currently, the condor spread has widened by more that 13 bp, a magnitude seen only a few times in recent history. As liquidity issues get resolved and the 2s/5s sector of the swaps curve undergoes an appropriate correction, the convergence of the condors should follow.

Disconnect between vol and curve. Although vol continued to move higher in November, it did not appear excessive to mortgages as their cheapening drove the mortgage spreads to swaps higher in a way which was commensurate with the bid for vol. Currently, vol appears in tune with mortgage carry, but is consistent with a steeper curve. The figure below shows the history of the 1Y5Y against the levels implied by the levels of rates and curve slope together with projections across different rates scenarios. Current levels of vol are consistent with 50bp steeper curve and higher rates.

Current vol levels are consistent with a steeper curve

112111

90 79

60

70

80

90

100

110

120

130

140

150

160

01 02 03 04 04 05 06 07 08

1Y5YTREND*50bp sell off & 25bp steepening50bp rally & 50bp flattening


The positioning in this context would be a short vol against the curve steepener which would most likely converge due to steepening of the curve as the pressures at the short end fade away.

As rates move in and out of the flight to quality mode, they create an environment of declining correlation across different sectors of the curve as well as across different curves. With the persistence of this mode and continued spread volatility we would look to take advantage of the inconsistencies across Swaps and Treasury vol markets, especially since we believe that relative lack of liquidity and related problems might keep a large fraction of RV players away from these trades. Similarly, we see a growing fundamental widening between the value of caps and swaptions in this context, but the market’s focus and degraded liquidity might prevent a favorable execution. We would look for attractive entry levels to the wedge trade, preferably at the short end of the curve.

We still see a disconnect between vol and carry in the context of conditional trades with vol differential between short and long tenors still wide relative to the carry of the curve. While the dip at the front end of the curve is keeping carry players away from steepeners, given our rates and vol outlook, we still see good value in 5s/10s conditional bear steepeners 1Y forward.

Aleksandar Kocic (1) (212) 250 0376



Agencies

The crucial OFHEO 30% capital surplus requirement is not expected to be relaxed in the first half of the year. This will be the main constraint for portfolio growth and debt supply for 2008. If the 30% requirement is lifted in the second half, increased supply could put pressure on spreads. We see no risk to the senior unsubordinated credit outlook.

A moderate relaxation of portfolio caps may occur by H208, in the range of 5-10%. Conforming loan limits are less likely to be changed in 2008.

There is scope for outperformance of callables relative to bullets in the base case scenario of unchanged to moderately higher rates for 2008, especially given the current high levels of implied volatility.

Subordinated Agency debt will remain vulnerable to headline risks, but the GSE’s are expected to adequately manage their capital requirements. We recommend adding subdebt in periods of heightened capital adequacy concerns.

FHLB advance growth is expected to remain in the $30bn per month range for the next one or two quarters. The $200bn increase in FHLB discount notes seen in Q407 may not be termed out if liquidity and credit conditions begin to improve in the first half of 2008.

Credit portfolio growth instead of retained portfolio growth for 2008

When year-to-date declines of $2bn and $1.5bn in capital surpluses were reported by Freddie Mac and Fannie Mae, respectively, in their November Q3 10Q reports, the market became focused on whether the GSE’s were at risk of falling below capital adequacy, what measures could be taken to restore capital, and whether OFHEO would soon drop the 30% additional capital requirement. OFHEO introduced the 30% buffers in January 2004 for Freddie Mac and in May 2006 for Fannie Mae. Before the credit events began this summer, there had been a general expectation that the additional capital requirements would be dropped when the GSE’s updated their internal accounting and resolved material weaknesses, which included timely quarterly financial reporting. Although Freddie expects to be timely by the end of 2007 and Fannie is currently timely, the probability of dropping the 30% buffer in the first half of 2008 has been reduced by the worsening housing market outlook and the enormous impact it has had on credit markets.

Although OFHEO has indicated that it will enter into more discussions regarding the 30% rule after the Q407 financials are released in February, it is likely to remain very conservative in its approach to fostering safety and soundness at the GSE’s. If the 30% rule is lifted, we would expect it to happen in the second half.

Credit portfolio growth will be the focus for 2008

Change in credit portfolios: 3-mo rolling avg

-

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

Dec

-06

Jan-

07

Feb-

07

Mar

-07

Apr

-07

May

-07

Jun-

07

Jul-0

7

Aug

-07

Sep

-07

Oct

-07

$

FNMA

FHLMC


Both Agencies successfully recapitalized via preferred stock transactions of $6bn to $7bn late in November. But because retained portfolio growth is so capital intensive (3.25% capital required), the growth path for 2008 will be the guarantee portfolios (0.585% capital required). Growth in the GSE’s guarantee business is more in line with the core purpose of the Enterprises which is the packaging of mortgage loans into passthrough securities backed by the credit of the Agencies, which are in turn implicitly backed by the credit of the US Government. This is the mechanism by which investors channel funds to borrowers yet face the uniform credit of the GSE’s, and it is considered the core purpose of the GSE’s.

Fannie and Freddie will increase their guarantee fees for 2008 to reflect the higher level of credit risk. Fannie averaged fees of 22.8bp and Freddie averaged 18bp in Q307. Initial indications are for a fee increase of around 6bp. Although given the high rates of the recent preferred stock issues (8.25%-8.375%) even with the higher guarantee fees it will be challenging to achieve desired levels of profitability in 2008. Year-to-date, the GSE’s have grown the credit portfolios by $488bn, or 13% annualized, Given the ARM resets expected for 2008, we expect about a 25% increase in the net supply of conforming loans available for GSE’s to securitize. This would allow portfolio growth closer to $580bn in 2008. We do not expect the Agencies to significantly reduce the size of



their retained portfolios but this will be an important option for them in their capital management process in 2008. A decrease of $100bn in the retained portfolio frees up $3.25bn of core capital. This amount of capital could support an increase of $550bn in the credit portfolio. Because of the limited increase, or possibly decrease, in the retained portfolios for 2008, the net supply of Agency debt is expected to remain unchanged or decline modestly. As a result, the focus on credit portfolio growth in 2008 will be a positive for Agency debt spreads. The GSE’s will tactically replenish retained portfolio runoff when valuations of mortgage-backed securities are relatively favorable versus GSE funding levels. We believe that the GSE’s will manage their capital constraints adequately throughout the year and would look to add subdebt exposure in periods of heightened concern.

Caps and conforming loan limits unlikely to experience significant modifications in 2008

The portfolio caps were established by OFHEO to limit growth in the retained portfolios of mortgage-backed securities, which consist of whole loans, Agency MBS and private label MBS. OFHEO considers these portfolios more risky than the credit portfolios because of the significant interest rate and implied volatility risks that must be actively managed. The initial caps set in 2006 were 2% annualized growth for Freddie Mac, and 0% growth for Fannie Mae. But in response to the initial phases of the credit crisis, in mid-September 2007 OFHEO adjusted the caps upward, allowing both portfolios to grow to as large as $742bn by the end of 2007. The caps were effectively increased to 5.5% for Freddie Mac and 2% for Fannie Mae. For 2008, the current allowable rate of increase is 2% for both, or a $757bn maximum (measured as outstanding average unpaid principal balance in the month). Yet despite the moderate increase in caps in 2007, portfolio growth was hindered by capital constraints. Both portfolios declined in September. In October, Freddie’s portfolio again declined, (to $704bn) while Fannie’s increased $8.5bn to $732bn due to a single transaction involving the purchase of whole loans. An increase in portfolios to the maximum allowable size in 2008 would require capital of about $1.7bn for Freddie and $1bn for Fannie. If caps were raised to 5% above the current year-end $742 maximum, capital required would be $3bn for Freddie and $2bn for Fannie. This may be too large of a capital drain. As a result, although it is possible that OFHEO raises the caps in 2008, we believe that 1) the rate would not exceed a 5% maximum growth limit, and 2) the portfolios would not grow more than 2% because of the capital constraints. If OFHEO were to drop the 30% additional

capital requirements, however, then caps of 5-10% would make more sense.

The conforming loan limits are a more difficult restriction to modify because they are codified in law and would require an act of Congress to change. It is difficult to justify the need to increase conforming loan limits given the expected large supply of conforming loans in 2008. Furthermore, increasing conforming loan limits would introduce additional risk that stretches the GSE’s beyond their core business strengths. In light of the heightened need for safety and soundness in the current market environment, together with the political focus on presidential elections for 2008, we think that conforming limits are unlikely to be modified in 2008.

2008 supply limited at FNMA and FHLMC, while FHLB continues funding advance growth with discount notes

Debt growth at Fannie and Freddie will ultimately be a function of housing market conditions.

Flat or declining portfolios should lead to declining

debt outstanding at Freddie and Fannie in 2008

700

720

740

760

780

800

820

840

Mar

-05

Jul-0

5

Nov

-05

Mar

-06

Jul-0

6

Nov

-06

Mar

-07

Jul-0

7

Nov

-07

$bn

640

660

680

700

720

740

$bn

FHLMC debtoutstanding

FHLMC RetainedPortfolio (right)


If the housing market remains as weak as expected in 2008 (another 5% decline on average), net debt supply could be close to zero. In the event that housing is worse and defaults higher than expected, capital will be threatened and the portfolios will likely shrink. A historical precedent is Fannie’s reduction of its retained portfolio by $177bn for capital purposes in 2005. Debt outstanding declined $189bn from $955bn to $766bn that year. The best case housing scenario could lead to a combined Fannie/Freddie net supply of $100bn on the year, most likely in the second half. Callable issuance versus bullets will largely be a function of the attractiveness of par callable coupon levels versus bullet yields. This depends on implied volatility levels and curve slope (forwards



versus spot). A better than expected housing market should lead to lower implied volatilities but a less inverted front-end, which have offsetting effects on callable coupons. In a worsening economic environment with high implied vols, the demand for callables could potentially be strong as long as the front end of the curve doesn’t invert too much with Fed easing expectations. Our base case scenario is one in which volatility remains high, and the front end prices moderate eases – an environment which is mildly supportive for callable issuance. Among the outstanding Fannie and Freddie callables that can be called in 2008, most have their first call dates in the first quarter. In the event of a large rally early next year, approximately $120bn could be called in Q1.

2008 call dates for FNMA and FHLMC outstanding

callables are heavily weighted in Q1

Outstanding callable notionals ($bn)

-

20

40

60

80

100

120

140

Q407 Q108 Q208 Q308 Q408 Q109 Q209 Q309 Q409


Issuance at FHLB in 2008 depends on the growth in advances to its 8100 member institutions. Advances growth was spectacular in the fourth quarter as FHLB became a major source of funding for financial institutions during the onset of the credit crisis. Although the base case outlook for 2008 will be for improving credit conditions, advances should continue to grow at least for the first quarter or two, depending on how the credit situation evolves. The first large jump in advances (+110bn) occurred in August when ABCP outstanding fell

the most (-205bn). We expect advances to continue partially filling in the funding holes left by ABCP. The ABCP market has averaged a decline in outstanding of about $50bn per month since the first large drop in August. FHLB advances growth has averaged $30bn per month in the same period. We expect the $30bn per month rate to persist for the next quarter, if not the first half of 2008. This advance funding has been achieved almost exclusively via discount notes. Since July, total debt at FHLB is up $207bn while discount notes are up $197 and bonds are up $10bn. Discount notes have the advantage that they can be expired whenever normal credit conditions return. But if it becomes clear that credit conditions will not improve materially in 2008, a portion of the discount notes would likely be rolled into one-year or longer maturities. This would probably have comparatively little effect on spreads in the front end given the overall flight-to-quality conditions that would prevail in this environment.

FHLB advance growth has been keeping pace with the

decline in ABCP outstanding

400

500

600

700

800

900

1000

1100

1200D

ec-0

3

Apr

-04

Aug

-04

Dec

-04

Apr

-05

Aug

-05

Dec

-05

Apr

-06

Aug

-06

Dec

-06

Apr

-07

Aug

-07

Dec

-07

billi

ons

ABCPOutstanding

FHLBAdvances


Ralph Axel (1) 212-250-7104



BMA Swaps

Ratios widened massively in the second half of the year as the BMA market absorbed several rounds of sub-prime shocks following relatively calm first two quarters.

We see a downgrade of major guarantors as unlikely although the credit rating issue may persist for a while. As a major risk scenario for the muni market in 2008, should such a downgrade occur, TOB dealers may have to sell assets and unwind BMA hedge.

The initial signs of reestablishing directionality of the ratios were interrupted by the credit crisis. We see continuation of this trend in the first quarter of 2008 and discuss a possibility of the restoration as a function of relative carry between the two curves.

The long-term municipal bond issuance in 2008 could pick up further to $450 bn while liquidity for the short-term securities may dry out temporarily in early 2008 due to soft issuance and strong demand from tax-exempt money market funds.

BMA market in 2007

The BMA swaps experienced two dramatically different regimes in 2007. In the first half of the year, ratios remained stable with RV players continuing to take advantage of the carry trade on the ratios curve. In the second half the market saw a substantial re-pricing of credit premia of the muni products. The first episode of the re-pricing is the widening credit spreads in July, and the BMA market was hit by the herding behavior of BMA hedging which drove ratios wider. Although the ratios recovered quickly at end of August, the market got caught again by the monoline insurers’ credit rating stories in early November. Since then, the possible downgrade of AAA guarantors has been the dominant challenging risk factor causing another wave of widening at the long-end. Market conditions were gradually improving after the Thanksgiving, although the long-end ratios remain still much higher relative to their pre-crisis level. The front-end ratios also saw massive widening as dealers were clearing their floater inventories at the year end. As a consequence the 2s/10s ratios slope almost inverted, which are under quick correction right now.

BMA ratios widened due to credit premia re-pricing

66

68

70

72

74

76

78

Dec-06 Feb-07 Apr-07 Jun-07 Aug-07 Oct-07 Dec-07

2Y

10Y

30Y


2008 outlook

Pressure on guarantors is likely to persist for awhile The threat of monolines downgrade is still evolving and is likely to persist in the next few months. In our view, it is unlikely for major guarantors to get downgraded because AAA rating is crucial for their business model and the options for keeping their credit ratings, although costly, seem substantial. The fact that CIFG and MBIA successfully raised new capital demonstrated that cash is available from private equity firms and other types of investors. Besides, insurers may also seek to add reinsurance to their relatively low risk business lines, or as a last resort, they could even sell their lucrative municipal bond guarantee business to a healthier insurance company to separate that business from sub prime risks. Given that, we still think the market sentiment would prevent the back-end ratios from tightening further in a significant way and would like to keep the monolines downgrade as the major risk scenario for BMA market in 2008.

Directionality In the first half of 2007, ratios curve flattened while swap curve steepened. This could be an initial sign of return of directionality between BMA and swaps markets, although the directionality was broken down in the second half of 2007. Historically the ratios curve flattens when the swaps curve steepens, and vice versa, as shown in the chart below. Such negative correlations exist because the Muni curve tends to be less reactive than the swaps in either direction. Substantial deviations from this pattern had been observed since April 2005. This was catalyzed by an increasing interest of the carry players. They were



driven to the Muni market as the foreign buying continued to flatten the yield curve, while the absence of foreign participation in the Muni market kept the BMA curve steep. Given the likely steepening swap curves in 2008, will the directionality be reestablished, especially if the credit issues get resolved?

The historical directionality between ratios and rates

reappeared in 1st half of 2007, but broke down again.

y = -0.0426x + 8.2202

R2 = 0.7185

-10

-8

-6

-4

-2

0

2

4

6

8

10

-10 40 90 140 190 240 290swap 2s/10s slope

BM

A ra

tio 2

s/10

s sl

ope

1998-2005

1st half of 2007

2nd half of 2007


We believe when directionality will be reestablished in a large part depends on the way RV players migrate back to the swap curve. As shown in the figure below, RV players came into BMA market attracted by superior carry in early 2005. Carry in ratios curve has been trending down since the proliferation of RV players in BMA market, while carry on swap curve has been improving in 2007. The diminishing difference has made BMA markets less attractive than it was in 2005 and 2006. Yet the carry in the swap curve currently is still way off from the level of pre-2005, and seems not providing enough incentive for carry players. Moreover, although we maintain a steepening bias on the swap curve, a reversal of “conundrum” may appear in this steepening process in the sense that the swap curve may fail to steepen as much as expected while fed keeps cutting rates. In such scenarios, a sticky flatter curve may delay a return of the RV players, and hence postpone the reestablishment of directionality.

RV players migrated to ratios curve since early 2005

for superior carry. Are they ready to return?

-1

1

3

5

7

9

11

13

15

17

Jan-03

Jul-03

Jan-04

Jul-04

Jan-05

Jul-05

Jan-06

Jul-06

Jan-07

Jul-07

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4carry - swap curve 10s/30s (lhs)carry - ratio curve 10s/30s (rhs)

Proliferation of RV palyers in BMA market


Ample issuance, yet possible liquidity dry out in early 2008 The long-term municipal bond Issuance in 2007 has peaked at $420 bn, and it may pick up even further in 2008 as the low rate environment after Fed cuts provides more incentive for the issuers. We estimate that 2008 will see around $450 bn supply, based on the historical issuance and 5Y5Y rates.

Issuance follows 5Y5Y very well historically.

0

5

10

15

20

25

30

35

40

45

50

99 00 01 02 03 04 05 06 07

Long-Term Issuance ($bn)

Issuance forecasted by 5Y5Y

Source: Deutsche Bank, Bond Buyer

However, the primary market will most probably be quiet in the first quarter due to seasonal effects and, more importantly, cheapening of the municipal bonds and ongoing credit concerns. The issuance pattern in 2007 already showed that issuers withheld new borrowing during credit crunch, as shown by the severely reduced issuance in July, August, November, and December in 2007 in the chart below.



Issuance severely reduced in months of cheapening

munis in 2007. The bars show the difference between

the actual issuance and the projected issuance

amount according to seasonality.

-20

-15

-10

-5

0

5

10

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

$bn

Source: Deutsche Bank, Bond Buyer

If a prolonged credit crunch keeps issuance at a slow pace in 2008, the liquidity of the short-term securities may dry out as demand from the tax-exempt money market funds is usually strong when muni cheapens, as shown by the quick growth of the asset size of MMF in the second half of 2007. Such liquidity dry out may put some bear steepening pressure on the front-end. A prolonged credit crunch may also create a crowded primary market after the situation crystallizes, and the concentrated issuance could shift the steepening pressure to the back-end.

Asset size of tax-exempt money market funds grew

fast in second half of 2007 due to cheapening muni.

250

300

350

400

450

500

Jan-03

Jul-03

Jan-04

Jul-04

Jan-05

Jul-05

Jan-06

Jul-06

Jan-07

Jul-07

Source: Bloomberg

Risk scenarios

The credit downgrade of guarantors presents one of the major risks for the BMA market in 2008. Although we think the downgrade to a major insurer is less likely, it could cause a serious liquidity crunch in the muni market via Tender Option Bond (TOB) programs should such a downgrade occur. TOB sponsors buy fixed rate, long term municipal bonds and finance them to create short term tax exempt floating rate securities. The major portions of such floaters are bought by money market funds, which require the underlying bonds rated AA or better. Many lower rating muni bonds become TOB eligible after credit enhancement using credit line of banks or being insured by an AAA wrapper. Should monolines get downgraded, money market funds may have to tender the tainted floaters back to the TOB remarketing agents. In such a scenario, remarketing would involve applying new credit enhancement to the underlying bonds. Or dealers may be forced to sell such lower-rated assets, which could be accompanied by massive unwinding of the BMA hedging position, pushing ratios at the long-end lower. Meanwhile, due to the reduced supply of short-term instruments from TOB, the BMA index could reset lower.

Lei Chen (1) 212 250 9830

Aleksandar Kocic (1) 212 250 2131



Euroland

Overview

Growth in the Eurozone will also be affected by the deleveraging we are seeing in financial markets as well as the tightening in credit conditions

Negative wealth effects from housing may also manifest themselves for some European countries

Unlike the US and the UK, Euroland is also having to cope with an extremely strong currency which is helping to contain price pressures on the one hand, but also squeezing exporters

Unlike in the US and the UK, however, the front-end is not pricing any chance of a rate cut from the ECB, thanks to continued hawkish rhetoric

We believe this offers an extremely attractive opportunity to enter long positions in EONIA, which will start to perform as headline inflation recedes and ECB talk softens

We also see considerable value in 2Y-10Y steepeners which should benefit from a softening economic outlook in Euroland, as well as a likely repricing of the US markets

We enter a USD-EUR 5Yx5Y widening position, as the most attractive way of expressing the view that the risks to global growth are disproportionately reflected in US pricing

Credit to be rationed in Euroland, too

European banks are likely to be as affected as US banks by an unwinding of off-balance sheet vehicles such as SIVs and ABCP conduits, not least since many European names seem to have been even more heavily involved in securitizing and refinancing US mortgage assets than their US counterparts. The on-going discussions about the future of IKB and Sachsen LB, as well as the second round of write-downs at UBS serve as timely reminders of the global nature of the current liquidity crunch.

Consequently European banks are tightening credit standards almost as aggressively as their US counterparts, since the refinancing of former off-balance sheet vehicles requires significant balance sheet commitment from these financial institutions too, to the detriment of existing lending activities.

Credit standards are expected to tighten further

-20

-10

0

10

20

30

40

50

Apr-03 Jan-04 Oct-04 Jul-05 Apr-06 Jan-07 Oct-07

House purchases

Corporate

Consumer credit and other lending

Easi

ng /

tight

enin

g of

cre

dit

cond

ition

s


We have argued in Bond Market Strategy that the appearing cracks in the US housing market have been instrumental in kicking off the unwind of the excess leverage built up since 2004. On many measures, the European housing markets look as overvalued as the US market. While financial engineering and increasing household leverage was key in pushing housing valuations to the extreme in the US, lending standards were not relaxed to the same extent in the Eurozone, However the fundamental concerns about a normalization of house prices and its effects on the consumer are clearly mirrored to some extent also in Europe.

European housing may be as overvalued as in the US

-20

-15

-10

-5

0

5

10

15

20

25

30

35

71 74 77 80 83 86 89 92 95 98 01 04 07

Euroland

US

% deviation from trend




ECB needs to do more to combat liquidity strains

Even though Wednesday’s announcement from the Federal Reserve, the ECB and the BoE amongst others will likely improve conditions in USD and GBP money markets, it effectively only brings US and UK central bank tools to European standards. The Term Auction Facility resembles the regular term auctions held by the ECB, while the widened collateral pools now accepted by the Fed and the BoE mirror the ECB’s existing definitions. This represents in our view a significant enough departure from current policy to improve the situation in USD and GBP money markets, however, we do not expect any such mechanical relief in the Eurozone.

Euribor bases are still far from normal

0

10

20

30

40

50

60

70

80

90

100

Mar-07 May-07 Jul-07 Sep-07 Nov-07

3M Eonia - Euribor spot basis3M Eonia - Euribor Mar basis


Making USD funds available to European institutions through a swap line with the Federal Reserve against ECB eligible collateral does not alter the fact that European banks need to fund US assets rather than ECB eligible assets. We therefore expect the basis between interbank and overnight rates in EUR to remain wide and to improve only slowly going into next-year, in line with current forward pricing. This implies continued funding difficulties for all Euribor based borrowers, including financials, homeowners and corporates.

Currency reduces chance of de-coupling…

Given the still very hawkish rhetoric by the ECB, as well as the more aggressive accommodation on display in the US and the UK, the currency remains at extremely elevated levels. Not only does this put significant pressure on exporters, it considerably increases the risk that the Eurozone fails to decouple from the US.

The currency raises the risk that US weakness spills-

over into the Eurozone

115

120

125

130

135

140

Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07 Jan-08

EURTWI

calendar year averages


… and mitigates inflationary concerns

The combination of weakening economic indicators and a strong currency make us relatively sanguine about the inflationary outlook. We do believe that we are likely to receive confirmation that the current high headline numbers are transitory and that base effects will result in a sharp improvement in inflation figures early next year. Also, the weakening in growth that we expect to take hold suggests that companies will have very little opportunity of passing on increased costs of production, be they through higher input prices, or rising wage bills. Second round effects should therefore remain limited.

Inflation is expected to fall back sharply

0.8

1.3

1.8

2.3

2.8

3.3

Jan-01 Jan-03 Jan-05 Jan-07

HICP

co re (to tal excl energy, fo o d, alc, to b)

% y/y

fo recast

base effect fro m German VA T


Forward-looking indicators point to trouble

Not surprisingly, given the tightening in credit and financial market conditions, as well as the strength in the currency, the forward looking components of the main confidence



indicators, especially for the service and financial sectors are pointing to a weakening in the economic outlook.

Forward looking indicators point to weakness

50

52

54

56

58

60

62

May

-05Ju

l-05

Sep-05

Nov-05

Jan-0

6

Mar-

06

May

-06Ju

l-06

Sep-06

Nov-06

Jan-0

7

Mar-

07

May

-07Ju

l-07

Sep-07

Nov-07

90

92

94

96

98

100

102

104

106

108

PMI CompositePMI ServicesPMI ManufacturingIFO - Expectations


Hawkish ECB supports front-end

The market is now pricing in the chance of ECB rate hikes, and no downside risk to the Eurozone economy is reflected in current market pricing at all. We think that a long position in the EUR front-end is one of the most compelling trades for 2008. This is especially true when comparing the current EUR pricing with that in USD and GBP. If those relative rate paths were to be realized the currency is likely to come under further upward pressure, in itself raising the probability of the ECB cutting rates.

Hawkish front-end pricing does not reflect

distribution of risks, in our view

ECB

-25%

0%

25%

50%

Jan Feb Mar Apr May Jun Jul Aug


It is important to remember that the Eurozone is the only market experiencing the higher Libor resets, the credit tightening and deleveraging in combination with a strengthening currency. This need for rate cuts can also be backed up by a simple arithmetic around monetary conditions in Euroland. Our economists estimate that 2% appreciation in the EUR trade weighted index accounts for

about 25bp of tightening. Thus, the 7% appreciation of the EUR TWI in 2007 (5.5% if we compare 2007 vs. 2006 average) accounts for 60-90 bp of tightening. Currently, the average Euribor-Eonia bases spread for the March08 to Dec 08 contracts averages about 30bp, which means that Euribors are pricing another 25bp of tightening even beyond the year end effect. Thus at current market pricing monetary conditions are arguably 85bp above the (generous) neutral rate of 4% (2% real growth and 2% inflation) which happens to be the current ECB refi rate. Thus unless the liquidity situation improves faster than what is priced in the forward bases and the EUR depreciates somewhat, the combined impact the credit rationing and the EUR appreciation should start to be reflected in the macroeconomic data. Waiting for the economic data to confirm the ECB’s working assumption of growth remaining close to trend is therefore the equivalent of judging by your rear view mirror whether the road you have just come down on was indeed straight. We believe that once inflation figures start moderating, and key wage negotiations are behind us, we are likely to see a noteworthy change in ECB rhetoric.

Long-end at risk from US correlation

So, although we are bullish European fixed income, in particular relative to current market pricing, the long-end of the curve obviously remains exposed to the likely bearish dynamic of the US market. We therefore see little value in the long-end of the EUR curve on an outright basis, due to the correlation risk and instead would instigate a long position in EUR 5Yx5Y against a short position in USD 5Yx5Y (see Bond Market Strategy section for more details).

Enter EUR-USD 5Yx5Y widening trades

EUR-USD 5Yx5Y spd (bp)

-25

0

25

50

75

100

125

150

175

Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08




ALM impulses uncertain

The excess flatness of the long-end of the curve has corrected with respect to the short-end in EUR. In our view this reinforces the directionality of 10Y-30Y and makes positioning on the curve guess-work. Solvency ratios in the Netherlands have continued to improve to such an extent, that pension fund hedging can remain extremely opportunistic. Sharp falls in general yield levels and a severe underperformance of equity markets may change that picture, but for the time being we remain neutral on the long-end.

Solvency ratios

100%

110%

120%

130%

140%

150%

160%

2004 2005 2006 2007

ABP

PGGM


The fact that equity valuations are not expensive (as measured by the equity risk premia – see Bond Market Strategy for details) also suggests that it will be very difficult to predict appetite for further ALM hedging.

10Y swap spreads fair, 2Y swap spreads driven by the basis

Although we are less optimistic than the market, or indeed governments on budget deficits for next year, net issuance will continue to fall, thanks to very large redemptions. This should provide some support for swap spreads, even if total issuance is likely to come in slightly higher than currently expected.

Tightening of credit standards

-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07

DEM 10Y ASW

Long term forecasts


Relative to current budget deficit expectations 10Y swap spreads are fair. Should growth weaken more materially than we expect, than swap spreads could come under tightening pressure.

Front-end swap spreads on the other hand will remain completely driven by the basis. Rolling into the new year, should mechanically tighten front-end swap spreads somewhat by removing the turn of the year, however, we remain cautious and see the risk that the Euribor bases do not tighten as quickly as currently priced in by the forwards.

Ralf Preusser (44) 20 7545 2469

Francis Yared (44) 20 7545 4017



EUR Derivatives: 2008 Outlook The second half of 2007 has turned-out to be the

period when the excess leverage in the financial system finally began unwinding, and risk premia in general and volatility in particular started rising

For 2008, we expect:

risk premia and volatility to remain elevated

the directionality of volatility with rates to persist

CMS issuance to increase but the supply of volatility through range accruals to decrease

Limited ALM driven demand for long tenor Vega

We present a series of trades that are attractive in that context

What to expect in 2008?

The second half of 2007 has turned-out to be the period when the excess leverage in the financial system finally began unwinding, and risk premia in general and volatility in particular started rising. European volatility followed the trend with Gamma and Vega reverting back to levels more consistent with historical averages. Volatility bottomed out in April and bounced in May as US mortgage market convexity sell-off and the repricing of Fed rate cut expectations drove the bear steepening of the curve. Since then, the ongoing credit and liquidity crisis and the reduction of financial leverage in the system supported volatility. 2007: Volatility finally bounced back

40

45

50

55

60

65

70

75

No v-04 No v-05 No v-06 No v-07

DGXEUR IndexDVXEUR Index


In the mortgage-lead bear steepening, Gamma on the belly of the curve led the spike in volatility as the EUR front-end was still anchored by a predictable ECB and the long end sold-off in sympathy with the USD curve. Then, as the credit and liquidity crisis unfolded, volatility moved

back to the front end of the curve as the ECB’s (and other central banks’) monetary policy path got repriced. As we highlighted back in June, central banks benefited from an exceptional macroeconomic environment during the last tightening cycle. The cycle started from an extremely low level of real rates while, with the benefit of globalization, inflation levels remained subdued. Thus central banks had the opportunity to gradually normalize interest rates and the luxury to be exceptionally transparent in their forward looking statements. This very benign macro environment enabled central banks to be more predictable, which reduced volatility in the front end of the curve and risk premia generally. However, now that inflation is close to, or above central banks’ comfort zones and that the trade-off between growth and inflation is becoming more acute, central banks are bound to be more agnostic about the direction of interest rates and will therefore be less predictable. This will support elevated volatility in the front end of the curve. Given the strong performance of the upper left corner (short expiry/short tenor) of the volatility surface, we are entering 2008 with a volatility surface which is strongly inverted along both the tenor and expiry axis. Volatility surface is strongly inverted along both the

tenor and expiry axis

-15

-10

-5

0

5

10

15

20

Dec-06 A pr-07 A ug-07 Dec-07-4

-2

0

2

4

6

83m2y-3m10y3m5y-6m5y


The other salient feature of the recent crisis has been the reestablishment of a strong negative correlation between short term rates and short dated volatility. Given the current environment, we expect this directionality of volatility with interest rates to persist at least until we see a normalisation of the liquidity situation and the EURIBOR-EONIA bases.



Volatility is expected to remain negatively correlated

with rates (6M rolling correlation)

-1.2

-0.7

-0.2

0.3

0.8

Dec-02 Dec-03 Dec-04 Dec-05 Dec-06 Dec-07

3M 2Y 3M 5Y

3M 10Y 3M 30Y


From a flow perspective, 2007 started with extremely strong volatility supply through range accruals and callables, which drove volatility well below historical ranges. The effect of the range accruals can be seen in the cheapness of cap-floor volatility relative to swaptions and the cheapness of bottom left corner (long expiry/short tenor). Given the more uncertain macro outlook, we expect the supply of volatility through range accruals to remain low in 2008, although we could see the usual seasonal pattern of volatility selling in the early months of the year as risk appetite returns. Later in the spring, the unwind of some range accruals and some ALM related activity provided support to Vega. Finally, following the crisis, financial institutions have increasingly been resorting to CMS issuance to satisfy their funding needs. We expect the issuance of CMS to continue as financial institutions seek permanent funding solutions for the SIV and longer maturity ABCP conduit assets that are back on their balance sheets. CMS Issuance expected to continue in 2008

0

500

1000

1500

2000

2500

3000

3500

4000

4500

Feb-02 Jun-03 Oct-04 Feb-06 Jun-07

CM S (P lain)EUR mln


Finally, ALM activity has been relatively subdued with volatility buying broadly in line with seasonal patterns. The impact of ALM flows is observable in the richness of low strike receivers. 5y30y Payer-Receiver skew

-3.00

-2.00

-1.00

0.00

1.00

2.00

3.00

4.00

Jun-05 Dec-05 Jun-06 Dec-06 Jun-07 Dec-07


Given the good performance of equity markets over the last few years, the funding ratios for major ALM players have considerably improved. For instance the funding ratio of ABP (a Dutch pension fund) has spiked up and is now close to 150%. Funding ratio for ABP

1.00

1.10

1.20

1.30

1.40

1.50

1.60

04Q1 04Q3 05Q1 05Q3 06Q1 06Q3 07H1


Given the high funding ratios, we expect ALM players to remain more focused on equities and alternative investments such as commodities. The hedging both in volatility and in delta space should remain opportunistic and at this point, we do not expect large ALM driven vega demand in 2008. Actually, in rates space the focus is more likely to be on inflation linked bonds as the high funding ratios are triggering the indexation of liabilities to inflation. Of course, should the equity market crash and yields rally the hedging needs may change. However, given the still high equity risk premia, we see the risk of an outright crash of equity markets remote, even though the outlook for corporate earnings may be challenging.



High equity risk premium limits the risks of a crash

-3

-2

-1

0

1

2

3

4

5

6

92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07

EUR Equity Risk Premium

EUR ERP average = 1.9%

%

Stock market bubble


The trades for 2008

Long EONIA vs. sell calls on Euribor As highlighted in Bond Market Strategy and Euroland Strategy, we find long positions in the front end of the EUR curve attractive from a risk reward perspective. Currently the EONIA curve is pricing 40% chance of a hike in Europe which is in stark contrast with the additional 100bp and 50bp priced in the front end of the US and UK curve respectively. While Euroland’s economy is likely to be less impacted by the credit rationing than the US or UK economies, its growth outlook will nonetheless suffer both directly (tighter lending criteria) and indirectly (rising exchange rate) from the credit crisis.

We see asymmetric risks around our central scenario of one rate cut by June of next year. Given the downside risk to growth and the still present event risk, it is possible for the ECB to cut rates more aggressively than we expect. On the other hand, given the regime shift in the repricing of risk and the reduction in credit availability due to balance sheet constraints, the liquidity situation is unlikely to normalize quickly enough for the ECB to hike rates more than once in the worse case scenario. Therefore, we seek to construct trades that offer protection against one rate hike, but remain profitable if the ECB cuts rates more aggressively than expected.

An attractive way of expressing our view is to .go long the June 3M EONIA, and sell OTM calls on the corresponding 3M Euribor struck at 95.75 (4.25%) for a premium of 8 cents. The trade performs in a rally, but also offers protection for small sell-offs through the premium received on the short call position. The table below shows the performance of this trade for different EONIA and basis scenarios.

Scenario analysis (P&L of the trade in bp)


We see that this trade performs well if EONIA rallies and the basis remains elevated. We do however give up some upside if the basis normalises in a rally, to protect against a sell-off in EONIA, with a breakeven EONIA rate of 4.23%.

Long Front End: Payer fly on 1Y2Y swaption In our central scenario, the ECB would cut rate to 3.75%. Assuming that the basis does not fully normalise and a 50bp spread between 2Y swap rates and the terminal ECB rate, we expect the 2Y swap rate to be around 4.25% in one year’s time. Although this is the most likely scenario we believe the risks are skewed towards lower rates. Hence, we look to create a portfolio which has a positive payoff around a rate of 4.25%, remains profitable for big rallies in rates and offers protection in case the ECB remains on hold or hikes once if the basis normalises.

We express this view by buying payers with 4% strike on 1Y2Y and selling 4.25% and 4.5% strike payers for an 8bp take out. The payoff of the trade including the initial take out is given in the chart below. The breakeven for the trade is 4.83% which is above the maximum level in 2Y swap rates achieved during this tightening cycle.

Payoff Profile

-70

-60

-50

-40

-30

-20

-10

0

10

20

30

40

3.60% 4.10% 4.60% 5.10%

Instant P &LDecayed P &L



Long Forward Vol Given our positive volatility bias and the extreme inversion in the volatility surface we like to take exposure to forward volatility. Indeed, our models suggest that spot Gamma is expensive relative to the curve even after adjusting for the impact of the basis. However, forward volatility 3M or 6M out is closer to fair value.

DGX forward and yield curve slope

y = 0.2344x + 56.54R2 = 0.6417

y = 0.178x + 51.292R2 = 0.438

40

50

60

70

80

90

100

-20 0 20 40 60 80 100

DGX Dec DGX M ar

DGX Jun DGX Sep


The trade could be implemented either by entering a DGXEUR March IMM contract (at 60.5bp at the time of writing), or by entering a calendar spread on a specific tenor.

Given the current shape of the Gamma surface, we find the 6M5Y -1Y5Y calendar spread attractive. The trade would be structured so that it is Gamma neutral and with strangles rather than straddles to ensure a more stable Gamma profile over time (see FIW 23 November 2007).

6M forward 6M5Y is attractive

45

55

65

75

85

95

105

Dec-99 Dec-00 Dec-01 Dec-02 Dec-03 Dec-04 Dec-05 Dec-06 Dec-07

6m5y vo l6m fwd 6m5y vo l


5y2y-5y30y Bull steepener Combining our macro views with our understanding of the flows, we find entering a 5Y2Y-5Y30Y bull steepener attractive. The trade fits our core view for a steepening of the EUR curve. It also benefits from the cheapness of the 5Y2Y point (on the back of range accruals) and the richness of the 5Y30Y point (on the back of ALM flows). As mentioned above, we do not expect strong ALM flows in 2008 whilst we expect the supply of vol through range accruals to recede. Also by striking both options 100bp out, we take advantage of the richness of the 5Y30Y skew relative to the 5Y2Y skew. The trade details are summarised in the following table.

Trade details EUR 5Y2Y Rcr EUR 5Y30Y Rcr Net

Notional 810 -100

ATMS 4.59% 4.86%

ATMF 4.72% 4.93% Strike 3.72% 3.93%

Delta -28,541 26,031 -2,510

Gamma 309 -346 -37 Vega 1bp 97,052 -91,670 5,382

1y Carry -130,366 338,204 207,838

Take Out -2,532,220 1,996,838 -535,381


Aditya Challa (44) (020) 7547 5966

Alessandro Cipollini (44) (020) 7547 4458

Gopi Suvanam (44) (020) 7547 5966

Francis Yared (44) (020) 7545 4017



Eurozone Issuance Outlook: 2008

Eurozone gross issuance is expected to increase in 2008, while net issuance falls. This discrepancy is largely due to larger redemption payments in 2008 compared to 2007.

We are neutral on countries like Ireland, Spain and Belgium, which will remain susceptible to negative sentiment, but expect spreads for Austria, Finland and Netherlands to tighten further.

We provide the outlook for government bond issuance in the Eurozone for 2008. Gross issuance is likely to increase by approximately EUR 36 bln in 2008, predominantly due to larger redemptions. However, net issuance is still set to improve by about EUR 28bln. Full details of respective issuance programs have yet to be announced, and are expected over the coming weeks. As such these are preliminary estimates based on budget deficit projections and redemption payments.

Redemptions amount to EUR 511 bln in 2008, up EUR 63 bln from 2007. These are highest in Germany, France and Italy, where France in particular sees the largest increase from the previous year. Gross issuance should increase in Italy, France, and Greece but remain roughly stable elsewhere. With a few exceptions (Italy, Greece, Portugal and Ireland), net issuance should decrease in 2008, with total net issuance at EUR 78bln.

In the past, debt management agencies have typically front-loaded issuance. Market conditions are currently more volatile than in previous years, and the timing of debt sales will be important for issuers next year. Having said that, it is likely that we still see large issuance in Q1 – last year, approx 14% of the total year’s issuance was done in Jan, of that most was long end supply. Q1 saw about a third of the year’s issuance.

The past couple of years have seen bond issuance scaled back over the second half of the year as favorable economic performance led to higher-than-expected tax takes. Looking forward, the growth uncertainty makes it difficult for us to imagine that this will continue, and we would not expect positive fiscal surprises.

Gross Issuance 2007 2008F

Germany 143.0 143.0

France 103.7 123.0

Italy 171.6 187.0

Spain 22.4 17.0

Netherlands 18.2 18.0

Belgium 25.6 27.0

Austria 17.6 12.0

Finland 5.0 4.7

Portugal 8.8 13.5

Greece 32.0 40.0

Ireland 6.0 5.0

Total 553.8 590.2Source: Deutsche Bank

Net Issuance 2007 2008F

Germany 16.5 3.6

France 31.9 20.5

Italy 33.4 38.4

Spain -4.8 -9.9

Netherlands 3.5 -3.3

Belgium 3.6 -2.4

Austria 6.4 3.9

Finland 0.0 -2.0

Portugal 2.2 4.8

Greece 13.7 19.5

Ireland -0.1 5.0

Total 106.3 78.1Source: Deutsche Bank

Breakdown by Maturity 2Y-3Y 5Y 10Y 15Y 30Y IL CCT/

CTZFRN Total

gross

2007 104.2 110.5 144.5 27.9 46.2 43.2 44.7 4.0 525

2008R 127.1 120.8 154.2 38.1 49.1 49.4 48.6 3.8 591Source: Deutsche Bank

Redemptions

0

20

40

60

80

100

120

140

160

2008 Redemptio ns2007 Redemptio ns

Source: Deutsche Bank, various government agencies



Germany

Germany ( EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 56 32 39 0 10 6 143

2008R 60.775 30.03 34.32 0 10.01 7.865 143Source: Deutsche Bank

We expect issuance to remain at last year’s levels, despite the commitment to budget improvement. This is due to redemptions increasing by EUR 12.9 bln to EUR 139.4 bln. It is most likely that the maturity splits remain roughly the same. We would also expect continued support for the linker program with the possibility of a new 30Y IL.

France

France (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y+ IL Total

gross

2007 17.8

25.8

27.3

10.3

5.8 16.7 103.7

2008F 27.675 30.75 24.6 12.3 7.38 20.295 123Source: Deutsche Bank

In France, issuance should increase, again due to significant redemptions, which rise by EUR 50bln to EUR 102.5 bln. We would expect potentially more issuance in the front end, with new benchmarks in the 2Y, 5Y and 10Y sector. In the long end, the 15Y and 30Y benchmarks will probably be tapped, as will the ultralong. Given that 2009 is another big year for redemptions, the AFT may conduct buybacks to reduce this. Linker issuance should continue, as is the case for taps of current bonds.

Italy

Italy (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL CC

T ICTZ Total

gross

2007 30.5 28.5 30.7 7.5 11.0 16.4 20.3

23.5 168

2008F 30.9 32.7 37.4 9.4 11.2 16.8 22.4

26.2 187


In Italy, both gross and net issuance is likely to increase. We expect new benchmarks for the 3Y and 5Y sector. In the long end, we see potentially a new 15y, while the current 10Y and 30Y benchmarks continue to be tapped. We anticipate the linker programme will continue to be supported with further taps of current issues and possibly a new issue.

Austria

Austria ( EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 0 2.8 7.8 1.3 5.8 0 18

2008F 0 2.1 5.7 1.0 4.3 0 13Source: Deutsche Bank

In Austria, we estimate the Treasury will issue EUR 12 bln in 2008, an improvement from this year. We expect a new 10yr RAGB benchmark, most likely in Q1, issued by syndication.

Belgium

Belgium (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y FRN Total

gross

2007 0 7.7 10.8 2.4 0 4 25

2008F 0 8.4 11.9 2.7 0 3.8 27Source: Deutsche Bank

The political deadlock in the country since June has meant that a 2008 budget has not been proposed yet. However, it has been reported that legislation will be enacted soon, which will enable spending to continue but will cap it to 2007 levels. The finance minister has announced its financing requirement, which entails EUR 27 bln in OLO issuance. We expect at least two new benchmarks, most likely issued via syndication. The buyback program is also likely to continue

Finland

Finland (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 0 5 0 0 0 5

2008F 0 0 4.7 0 0 0 4.7Source: Deutsche Bank

Gross issuance should remain broadly unchanged from 2007. As last year a 5Y benchmark was opened, a new 10Y benchmark is possible.

Greece

Greece (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 1.7 5.8 9.8 6.2 4.0 4.5 32




We see Greek issuance at around EUR 40 bln in 2008. In terms of maturities, we expect new benchmarks in the 5Y and 10Y sector, while the 30Y and 15Y sector current issues will likely be tapped.

Holland

Holland (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 3.0 1.9 9.5 0.0 3.9 0.0 18


In the Netherlands we see gross issuance remaining steady at EUR 18bln as redemptions increase by EUR 7 bln. We might see a new 10Y benchmark and further taps in other sectors.

Ireland

Ireland (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 0 0 6 0 0 6

2008F 0 0 0 5 0 0 5Source: Deutsche Bank

While Ireland has no significant redemptions in 2008, it is likely to post a deficit. As a new 10Y benchmark was already issued in 2007, we expect a new benchmark, possibly in the 5Y sector, or even a longer issue.

Portugal

Portugal (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 0.9 1.0 5.9 0.0 1.0 0.0 9


Portuguese issuance is likely to increase by EUR 4.7 bln in 2008. Redemptions increase by EUR 2 bln. We expect at least a new 10y, and perhaps another new line.

Spain

Spain (EUR bln) 2Y-3Y 5Y 10Y 15Y 30Y IL Total

gross

2007 1.4 8.6 5.9 0.0 6.5 0.0 22


In Spain we expect EUR 17 bln in issuance, a decrease from 2007. We believe a new 3Y and 10Y benchmark is possible, as the current 5Y continues to be tapped. In the long end, no new issuance is likely as the current benchmarks continue to be tapped.

Comparative Deficit-to-GDP Estimates DB DB Govt

2007e 2008F 2008F

Germany 0.0 -0.1 -0.5

France -2.5 -2.4 -2.3

Italy -2.4 -2.4 -2.2

Spain 1.7 0.9 1.2

Netherlands -0.2 0.6 0.5

Belgium 0.3 0.0 n/a

Austria -1.5 -1.3 -1.1

Finland 0.8 1.1 1.1

Portugal -3.0 -2.7 -2.4

Greece -3.0 -2.0 -1.6

Ireland 0.5 -1.0 -0.2Source: Deutsche Bank

Growth Forecasts DB Consensus Govt

2008F 2008F 2008F

Germany 1.9 1.9 2.0

France 1.8 1.8 2.3

Italy 1.4 1.3 1.5

Spain 2.8 2.6 3.3

Netherlands 2.3 2.3 2.3

Belgium 1.8 2.0 n/a

Austria 2.4 2.4 2.4

Finland 2.7 3.0 3.0

Portugal 1.5 1.9 2.2

Greece 3.4 3.3 4.0

Ireland 4.0 3.7 3.5Source: Deutsche Bank

On the whole, our economists are less optimistic than the respective governments in deficit projections, as revenues are unlikely to continue improving, while budgets remain largely neutral. Although net issuance looks set to improve over 2008 at this point, things can easily shift over the course of the year. In addition, specific local risks may emerge, or in some cases worsen. Spain, Ireland and Belgium will probably continue to feel the brunt of worsening sentiment. We may start to see further differentiation between the core group, and almost certainly more susceptibility to “headline” risk.



Estimate of housing overhang as % of GDP Germany -6.5

Spain 24.6

France -8.6

Ireland 44.3

Italy -0.9

Euro Area 1.5

UK 4.9

US 7.1

Japan 10Source: CEPS working document No 276 Bubbles I Real Estate, GEPS Oct 2007

The housing sector will likely remain a source. A GEP publication (8 Oct 2007) and a working paper by the CEPS that estimates the amount of housing overhang suggests that both Spain and Ireland will remain on radar screens.

CDS spreads appear to have become one way of playing this, not just via shorting the bonds, as repo markets in the periphery are less developed. We expect this to continue.

10Y ASW Spread to Germany

0

5

10

15

20

25

M ar-07

A pr-07

M ay-07

Jun-07

Jul-07

A ug-07

Sep-07

Oct-07

No v-07

Dec-07

B elgiumirelandSpain


10Y ASW Spread to Germany:

0

2

4

6

8

10

12

14

M ar-07

A pr-07

M ay-07

Jun-07

Jul-07

A ug-07

Sep-07

Oct-07

No v-07

Dec-07

A ustriaFinlandNetherlands


Peripheral spreads ( to Germany) have recovered from the highs since the beginning of the month. We are neutral on countries like Spain, Belgium and Ireland, as it is likely that negative newsflow will continue to adversely affect sentiment. Meanwhile, we are positive on Austria, Finland and the Netherlands, looking for a further 5bps of tightening as their housing markets present little potential for negative headline risk, and given their strong fiscal positions.

Soniya Sadeesh +44 207 547 3091



Covered Bond Outlook 2008

The strong spread widening of EUR Jumbo covered bonds, particularly in US, UK and Spanish covered shows the increasing differentiation of investors according to issuer specific (cover pool credit quality, issue structure, bank credit quality) and market specific topics (housing market risk, country risk).

Given that housing market concerns are unlikely to disappear quickly, we see very little chance of returning to where we came from in terms of spreads, i.e. we expect that investors will continue to demand very different spreads for different covered bonds.

In 2007, the average spread difference between mortgage and public covered bond increased considerably. As housing market concerns in some countries may get worse before they get better, we do not expect this spread difference to reverse.

In our view, the willingness of issuers around the world (Europe, USA, Canada, Australia; Japan) to issue EUR Jumbo covered bonds might well amount to EUR 230 bn in 2008. Depending on the average issuance volume, this would mean almost one new EUR Jumbo covered bond issue per business day. Given the monthly supply in 2007, one can easily see how severe financial market turmoil reduced monthly covered bond supply in August/September and November/December. Hence, to allow issuers to bring the intended issues to the EUR Jumbo covered bond market, we believe a no-crisis scenario in the financial market environment is needed.

There are already 13 Jumbo covered bonds deals in the pipeline. Hence, liquidity will be crucial, in our view. With interbank market making not yet fully back in place, liquidity in the interbank market is still limited. In our view, interbank market making is crucial to providing continuous liquidity in size to rates investors.

Some spread recovery in case of 2Y swap spread tightening likely

Spread divergence* continued – covered bond market

got split even further

-12

-2

8

18

28

38

48

Jan-

07

Feb-

07

Mar

-07

Apr

-07

May

-07

Jun-

07

Jul-0

7

Aug

-07

Sep

-07

Oct

-07

Nov

-07

France CoveredGerman Mortgage PfandbriefeGerman Public PfandbriefeSpanish CoveredOther CoveredIreland CoveredUK Covered

*indices have different durations Source: iBoxx

The strong spread widening of EUR Jumbo covered bonds, particularly in US, UK and Spanish covered shows the increasing differentiation of investors according to issuer specific (cover pool credit quality, issue structure, bank credit quality) and market specific topics (housing market risk, country risk). However, in line with their status as bonds typically backed by a dual claim against a bank and a cover pool, covered bonds strongly decoupled from general bank credit spreads, which trade around an 8-year high.

Covered bonds showed stability versus senior and

subordinated bank debt

-15-55

152535455565758595

105115125135145

Oct

-06

Nov

-06

Dec

-06

Jan-

07

Feb-

07

Mar

-07

Apr

-07

May

-07

Jun-

07

Jul-0

7

Aug

-07

Sep

-07

Oct

-07

Nov

-07

Dec

-07

Corporate Financial

CoveredCorporate Financial: Banks : Senior

Corporate Financial: Banks : Subordinated

Source: IBoxx



Demand for covered bonds from real money accounts The secondary market in the second week of December showed better buying interest from customers across the curve. We also saw further buying interest from central banks especially in German, Nordic and French covered bonds. All in all, bonds traded on slightly tighter vs. swaps.

Spreads may tighten in line with 2Y swap spread tightening Cédulas spreads versus swaps had not been correlated to 2Y swaps since 2003 (R square of almost 0). However, since July this year, Cédulas spreads versus swaps have been correlated with 2Y swaps (R square of almost 0.9). Hence, beside pure credit concerns, the liquidity shortage is clearly weighing on spreads of covered bonds. Hence, in line with the expectation of some normalisation of swap spreads next year and in the absence of negative credit events, spreads of Spanish and UK covered bonds should tighten. However, given that housing market concerns are unlikely to disappear quickly, we see very little chance of returning to where we came from in terms of spreads, i.e. we expect that investors will continue to demand very different spreads for different covered bonds. Moreover, event risk regarding bank credit quality (also impacted by the ongoing deterioration in housing markets) remains.

Cédulas spreads versus swaps highly correlated with

2Y swap spread (since July)

-70

-60

-50

-40

-30

-20

-10

0

10

20

30

Jul-0

2

Nov

-02

Mar

-03

Jul-0

3

Nov

-03

Mar

-04

Jul-0

4

Nov

-04

Mar

-05

Jul-0

5

Nov

-05

Mar

-06

Jul-0

6

Nov

-06

Mar

-07

Jul-0

7

Nov

-07

iBoxx Euro: Collateralized: Covered: Spain Covered(ASW in bp)2Y Smooth GER (Par ASW in bp)


Investor demand is critical

In contrast to widespread expectations, 2007 will end with less EUR Jumbo covered bond supply. Around EUR 160 bn of EUR Jumbo covered bonds have been issued. In 2006, EUR 182 bn of Jumbo covered bonds were issued. Due to current funding problems in senior bank bonds and RMBS, covered bonds might be the first choice for most banks in 2008. With three-month Euribor showing its nineteenth consecutive rise on Tuesday (fixing at 4.927%), hence 13.5 bp above the previous post-

summer high and 35 bp above its post-summer low, pressure on bank liquidity remains high.

Investors still in the driving seat With 13 new EUR Jumbo covered bonds already in the pipeline for 2008, it appears that investors will likely be price makers rather than price takers in 2008. This should also be true for the covered bonds that have remained most stable so far. E.g. 5Y public Pfandbriefe of BHH trade at ms – 4 bp in the secondary market and 5Y public Pfandbriefe of 10Y DG Hyp trade at ms – 3 bp in the secondary market. In the case of new issues, even the issuers of public Pfandbriefe may have difficulties funding sub-libor currently.

EUR Jumbo covered bond issuance in 2007 lower

than in 2006

0

20

40

60

80

100

120

140

160

180

200

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

AT CA FI FR GE IR IT LX

NL NO PT SP SW UK US


In our view, the willingness of issuers around the world (Europe, USA, Canada, Australia; Japan) to issue EUR Jumbo covered bonds might well amount to EUR 230 bn in 2008. Depending on the average issuance volume, this would mean almost one new EUR Jumbo covered bond issue per business day. Given the monthly supply in 2007, one can easily see how severe financial market turmoil reduced monthly covered bond supply in August/September and November/December. Hence, to allow issuers to bring the intended issues to the EUR Jumbo covered bond market, we believe a no-crisis scenario in the financial market environment is needed.



2007 EUR Jumbo volume lower in August/September

and November/December

0

5

10

15

20

25

30

35

40

Jan Feb Mar Apr May Jun July Aug Sep Oct Nov Dec

2006

2007


This can also be seen by the quarterly EUR Jumbo covered bond issuance distribution. Whereas Q2 2007 was the strongest Q2 ever in terms of supply, Q3 fell short of Q32006. Despite financial market turmoil, Q3 and Q4 2007 supply slightly exceeded Q3 and Q4 2005 supply, respectively, showing that there is a strong willingness or even need on the part of banks to issue EUR Jumbo covered bonds as soon as there is an opportunity.

Q3 and Q4 2008 with lower EUR Jumbo covered bond

supply

0

10

20

30

40

50

60

70

1999

2000

2001

2002

2003

2004

2005

2006

2007

Q1 Q2 Q3 Q4


The banks’ full-year results will not yet be out in the opening days of January. Hence, in our view, investors will continue to be cautious. There has not been even one long-term Jumbo covered bond issue since August. As already mentioned, we believe investors continue to be in the driving seat. In 2008, circa EUR 92 bn of EUR Jumbo covered bonds become due. Redemptions of Spanish Cédulas in 2008 amount to less than EUR 5 bn. In this respect, by issuing long dated in recent years, Spanish issuers (at least) do not face the pressure to refund high

redemption volumes. The risk to our cautious view on investor demand is that most bad news is out already and that hunting for yield by real money in 2008 could allow issuers to get all their funding needs done. However, even in such an scenario, it seems unlikely that we return to the pre-crisis environment where covered bonds of whatever structure could be issued without material pick-up to established covered bonds of established issuers.

Around EUR 92 bn of EUR Jumbo covered bonds

become due in 2008

0

20

40

60

80

100

120

140

2008

2010

2012

2014

2016

2018

2020

2022

2024

2026

Austria FinlandFrance GermanyIreland ItalyLuxembourg NetherlandsNorway PortugalSpain SwedenUnited Kingdom USACanada


Low redemptions of EUR Jumbo covered bonds in

January and February 2008

0

2

4

6

8

10

12

14

16

18

01-07

03-07

05-07

07-07

09-07

11-07

01-08

03-08

05-08

07-08

09-08

11-08

0

1

2

3

4

5

6

7

8

9Volume in EUR bn (LS)

No. of maturing issues (RS)


Investor demand the key question Particularly for countries currently under increased observation regarding credit quality, like US, UK, Spain and Ireland, we believe investor demand will be the key question. In some covered bond issues since September, bank buyers accounted for over 50%. Even though demand from banks for covered bonds since September has been reasonably strong, we continue to see the need for issuers to diversify away from banks to get their increased funding needs done. That said, in our view, due to a lack of safe investment alternatives and the ECB



eligibility of covered bonds, which by the way provide a carry of around 50 bp currently, banks will continue to be strong buyers of covered bonds. Covered bond investors have a tendency to buy domestic covered bonds, e.g. while German demand for Jumbo Pfandbriefe is around 50%, the bid for foreign covered bonds is only around 20% on average. Spanish investors take on average around 5% of Spanish Cédulas but do not buy foreign covered bonds. Also in the Nordics and Benelux, there is typically a higher demand for domestic covered bonds.

Banks continue to dominate as investors in Jumbo

covered bonds


RMBS investors may start buying covered bonds in the future. But given the pre-crisis investor distribution of ABS/RMBS (see appendix), high non-bank investor demand from former RMBS investors seem unlikely, particularly for longer-dated covered bonds.

Market making important for investor demand There are already 13 Jumbo covered bonds deals in the pipeline. Hence, liquidity will be crucial, in our view. With interbank market making not yet fully back in place (according to the minimum standards recommended by the German Association of Pfandbriefbanks), liquidity in the interbank market is still limited. In our opinion, interbank market making is crucial to providing continuous liquidity in size to rates investors (even though the EUR Jumbo covered bond market is only a fraction in terms of outstanding volume compared to EUR government debt market and with EUR 1.6 bn has far lower average volume). Without decent liquidity in size, some investors may no longer see Jumbo covered bonds as a government surrogate or rates product. Hence, we believe some kind of preservation measure has to be taken to restore market making fully. Currently, interbank market making is still reduced to EUR 5 m and bid-offer spreads are still tripled.

Adjustments in market making may be necessary

Given the ongoing differentiation of covered bonds by investors, there are many proposals and discussions about how to change market making going forward, i.e. how to preserve high liquidity: One possible avenue might be to differentiate market making according to spread ranges. If one assumes that covered bonds with greater credit risk trade wider to swaps than covered bonds with lower credit risk (and hence have differences in volatility), different bid-ask spreads might be justified. As the spread volatility is usually higher for longer-term covered bonds, one would have to adjust volatilities according to maturity bracket (unlike in the following chart). The chart shows spread volatilities for covered bonds trading in different spread ranges. The sudden drop of the volatility for covered bonds trading between 25 and 50 bp versus swaps was mainly the result of Northern Rock and WAMU covered bonds leaving even this spread range.

Volatility different for different spread ranges

0

5

10

15

20

2528

-Aug

-07

05-S

ep-0

7

12-S

ep-0

7

19-S

ep-0

7

26-S

ep-0

7

03-O

ct-0

7

10-O

ct-0

7

17-O

ct-0

7

24-O

ct-0

7

07-N

ov-0

7

14-N

ov-0

7

21-N

ov-0

7

28-N

ov-0

7

05-D

ec-0

7

(- x to 5 bp)5 bp to 15 bp15 bp to 25 bp25 bp to 50 bp


The following chart shows the year-to-date mean of 30-day volatility plotted according to spread range and duration bucket for around 300 EUR Jumbo covered bonds. The results basically show that for any given duration, volatility is higher in a higher spread

Banks40%

Central bank11%

Fund Manager27%

Insurance10%

Pension Fund6%

Other6%



Higher volatility for EUR Jumbo covered bonds

trading at a higher spread (for any given duration)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

2 to 3 3 to 4 4 to 5 5 to 6 6 to 7 7 to 8 8 to 9 9 to 10 10+

Upto 5

5 to 15

15 to 25

25 to 50


Due to different spread volatilities, covered bonds with the tightest spreads might deserve to have the tightest bid-ask spreads. There are various pros and cons to this idea: one counter-argument is that liquidity will dry up in case of a spread widening -- exactly at the time it is needed most. Moreover, the different spread ranges and the bid-offer spreads for each respective spread range in relation to the ‘normal’ market making commitment is certainly up for discussion. However, in our view, some measure has to be taken to improve interbank market making. Given the differences in volatilities of different covered bond sectors, a return of all covered bonds to the ‘normal’ market standards early January seems difficult. Without a well functioning market making, the structure of the Jumbo covered bond market may change going forward. Generally, no matter which measure is taken and which form of market making commitment is the result of the current discussion, we see no way to avoid frictions in market making in times of severe crisis. This is not restricted to covered bonds but is also true for sovereign bonds.

Trading platforms enhance transparency but may make providing liquidity more costly for banks In our view, well-functioning market making is crucial to get rates investors investing in covered bonds. Hence, in a case where the covered bond community is not able to re-establish well-functioning market making, part of the investor base may reduce its exposure to covered bonds, particularly in upcoming new issues. Providing liquidity to the market is a costly service. The suggestion of shifting all trading to trading platforms may enhance transparency, but may make providing liquidity even more costly for market makers. Hence, the market making mechanisms should, in contrast to the current minimum requirements, provide sanctioning mechanisms for banks acting as free riders. In cases of covered bonds trading at wide levels, credit investors may take a closer look at investing in

covered bonds, as has already happened in the case of covered bonds by Northern Rock or Washington Mutual.

Public Covered Bonds remain well supported

AyT – still No 1 issuer in terms of outstanding volume With around EUR 50 bn in terms of outstanding volume, AyT remains the No 1 issuer of EUR Jumbo covered bonds, followed by French CFF with around EUR 45 bn.

AyT remains the No 1 issuer of EUR Jumbo covered

bonds

0 10 20 30 40 50 60

AYTCED

CFF

EURHYP

HYPESS

BBVASM

CRH

DEXMA

DEPFA

CAIXAB

SANTAN

Sum of Outstanding Volume (bn)Source: Deutsche Bank

The difficulties in the primary market of EUR Jumbo covered bonds fed through into the 2007 year’s top ranking of issuers. CFF was the No 1 issuer with almost EUR 15 bn of EUR Jumbo covered bonds issuance, followed by BNP with almost EUR 9 bn and DEXMA with around EUR 8 bn. Hence, French issuers dominated the primary market, interestingly, without severely hurting spreads.

CFF – No 1 issuer in 2007 EUR Jumbo covered bonds

0 3 6 9 12 15

CFF

BNPPCB

DEXMA

EURHYP

HBOS

LBBW

BAC

NWIDE

CAJAMM

SANTAN

Sum of Outstanding Volume (bn)




Public Pfandbriefe continue to dominate the public covered bond market Public Pfandbriefe continue to dominate the EUR Jumbo public covered bond market. As around 40% of the collateral of public Pfandbriefe still consists of state-guaranteed public-sector bank debt, their share looks set to shrink in the next years. In 2008, we should see EUR 65 bn of redemption of German Pfandbriefe, accounting for the highest share of the total of EUR 92 bn. Due to the high negative net supply, spreads of public Pfandbriefe continue to be well supported.

Public Pfandbriefe still dominate the public sector

covered bond market with a market share of 72%

PP HYP54%

PO9%

PP LB18%

PACS8%

PL1%

CDEP3%CT

5%

ATFB2%


Short cuts public sector covered bonds

PP HYP Public Pfandbriefe of former Mortgage Banks

PP LB Public Pfandbriefe of Landesbanks

PO Public Sector Obligations Foncières

PACS Public Sector Asset Covered Securities

CT Cédulas Territoriales

CDEP Public Sector Covered Bonds of CDP

PL Public Sector Lettres de Gage

ATFB Austrian Fundierte Bankschuldverschreibung Source: Deutsche Bank

Even though non-German Pfandbriefe make up some of the decline, the outstanding volume of the whole public covered bond market continues to decline and is likely to do so also in 2008.

Outstanding volume of public covered bonds (EUR

bn)

0

50

100

150

200

250

300

1999 2000 2001 2002 2003 2004 2005 2006 2007

Outstanding Public (bn)


With the decline in outstanding volume and the increasing volume in mortgage backed covered bonds, the share of public covered bonds continues to decline. Hence, besides the typically strong quality of public collateral, the lack of public covered bonds in general is expected to support spreads of all public covered bonds.

Share of public covered bonds decreasing

72.1% 69.7%66.3%

60.5%

53.8%48.7%

42.7%

35.8%30.8%

0%

10%

20%

30%

40%

50%

60%

70%

80%

1999 2000 2001 2002 2003 2004 2005 2006 2007

Share of Outstanding


High redemptions in public covered bond indicate

declining volume

0

10

20

30

40

50

60

70

80

1999

2001

2003

2005

2007

2009

2011

2013

2015

2017

2019

Gross Issuance

Redemptions




In 2007, the average spread difference between mortgage and public covered bond increased sharply. As housing market concerns in some countries may get worse before they get better, we do not expect this spread difference to reverse. Another reason why we believe a reversal will be difficult is that ongoing new issuance of mortgage covered bonds should prevent spreads from tightening significantly. On the other hand, as described above, the negative net supply should support spreads of public covered bonds. The risk to our view is a strong improvement in the situation on the money market and a softening of housing market concerns. In this scenario, mortgage covered bonds could outperform. However, even in this case, we believe a return to previous levels is unlikely.

Spread difference between public an d mortgage

collateral increased in 2007

0

2

4

6

8

10

12

14

16

18

Jan-

07

Feb-

07

Mar

-07

Apr

-07

May

-07

Jun-

07

Jul-0

7

Aug

-07

Sep

-07

Oct

-07

Nov

-07

Dec

-07

Duration All Mortgage - AllPublic

Spread (All Mortgage - AllPublic)


Although the spread difference between mortgage and public Pfandbriefe is still comparably low (mainly due to limited housing market concerns and the negative net supply of German Pfandbriefe in general, in our view), the spread also increased strongly. Given that inaugural issuers in 2008 should be mainly in the mortgage sector and negative net supply is mainly in public sector collateral, we do not expect this to change significantly.

Mortgage Pfandbriefe have to pay a higher pick-up

versus public Pfandbriefe

0.00

1.00

2.00

3.00

4.00

5.00

6.00

Jan-

07

Feb-

07

Mar

-07

Apr

-07

May

-07

Jun-

07

Jul-0

7

Aug

-07

Sep

-07

Oct

-07

Nov

-07

Dec

-07

Spread (Mortgage Pfandbriefe-Public Pfandbriefe)

Duration (Mortgage Pfandbriefeover Public Pfandbriefe)


The Spanish Cédulas market experienced a significant divergence of spreads between Cédulas Territoriales (CT) and Cédulas Hipotecarias (CH). As housing market concerns should continue in 2008, we not expect this to change significantly. Moreover, the enhanced protection provided by the amendment of the legal framework for Cédulas increases the focus of investors on the collateral.

Structured covered bonds continue to increase in importance

With outstanding volume of almost EUR 190 bn, structured covered bonds have seen sharply increased outstanding volumes. Since 2001 the annual share of structured covered bonds in the gross supply of Jumbos has risen steadily to reach over 32% in 2006. At the end of 2007 it is around 35%. With that, the share of outstanding volume amounts to 22%. Even though in the current environment, structured covered bond are not well bid, we continue to see potential in structured covered bonds.



Structured covered bonds: sharp increase in

outstanding

0

20

40

60

80

100

120

140

160

180

200

1999 2000 2001 2002 2003 2004 2005 2006 2007

Outstanding SCB (bn)


Moreover, with the introduction of the legal framework for covered bonds in the UK, a substantial share of structured covered bonds will have to be categorized as legal framework covered bonds in the future.

Structured covered bonds have a share of 22%

9.8%

13.3

%

18.2

%

22.2

%

0%

5%

10%

15%

20%

25%

1999 2000 2001 2002 2003 2004 2005 2006 2007

Share of Outstanding


Spread of structured covered bonds widened sharply As expected, due to the ongoing skepticism about structured products in general and doubts regarding insolvency remoteness, structured covered bonds widened sharply. The difference is somewhat exaggerated as a big share of legal framework based covered bonds, German Pfandbriefe, have an average duration of around 3 years only. However, the chart nicely shows the trend.

Structured covered bonds widened sharplly versus

swaps (average of all outstanding EUR Jumbo

covered bonds)

-20

-10

0

10

20

30

40

50

Jul-0

6

Sep

-06

Oct

-06

Dec

-06

Jan-

07

Mar

-07

Apr

-07

May

-07

Jul-0

7

Aug

-07

Oct

-07

Nov

-07

StructuredCovered Bonds

Legal FrameworkBonds

SCB-Legal Spread


If we exclude Northern Rock and Washington Mutual covered bonds, which both got hit hard because of deterioration in senior credit quality (though for different reasons), the spread widening of structured covered bonds is less severe. Moreover, it is generally difficult to isolate the impact of the difference of structured versus legal framework based covered bonds. Around EUR 60 bn of outstanding UK covered bonds count as structured covered bonds at the moment. In March 2008, a specific legal framework for UK covered bonds is supposed to come into force. However, mainly due to concerns about the UK housing market, it seems unlikely that the introduction of the specific legal framework will have a groundbreaking impact on spreads. Due to higher security for investors, it should be positive; however, we believe easing housing markets concerns are needed to see a notable spread tightening in UK covered bonds.

Structured covered bonds widened less sharply if

Northern Rock and Washington Mutual are excluded

-10

-5

0

5

10

15

20

25

30

Jul-0

6

Sep-

06

Oct

-06

Dec

-06

Jan-

07

Mar

-07

Apr

-07

May

-07

Jul-0

7

Aug

-07

Oct

-07

Nov

-07

Structured Covered Bonds ex( NRBS ,WM)

Legal Framework Bonds

SCB - Legal Spread ex (NRBS, WM)




Non bank covered bond issuers possible in 2008 It is possible that non-bank borrowers start issuing covered bonds. Veolia Environment showed interest in issuing Obligations Foncières backed by payments it receives for its services under contract with local authorities. In the case of Veolia, the specific topic is that Veolia would issue via a Societe Credit Foncier (SCF), i.e. a banking subsidiary. Even though the issue of Veolia has not yet appeared in the market, other non-bank issuers, even directly issued out of corporates, might also try to tap the covered bond market.

BNP – the fourth-largest issuer of structured covered

bonds

0 10 20 30 40 50 60

AYTCED

HBOS

CEDTDA

IMCEDI

BNPPCB

NRBS

NWIDE

BRADBI

AAB

WMTop Ten Issuers


Expected issuance volume of EUR Jumbo covered bonds

In the following, we give some very rough indication what can be expected in terms of EUR Jumbo covered bonds issuance volume. However, we highlight that the whole covered bonds market amounts to around circa EUR 2000 bn, whereas the EUR Jumbo covered bonds amounts to circa EUR 860 bn. Assuming an average maturity of 5 years, redemptions would on average amount to EUR 400 bn (as a lot of covered bonds have been issued in the recent past, it might be lower than the average in 2008). Assuming a growth of EUR 140 bn (similar to the past few years), total gross issuance of covered bonds in all currencies and all types (Jumbo, non-Jumbo, bearer format, registered format, public, private placements) may well amount to EUR 540 bn. In our view, it will become increasingly important to have a view on the whole covered bond market. As funding will likely remain a major topic for banks in 2008, banks may focus even less than in the past on the covered bond type (Jumbo versus non-Jumbo), legal form (bearer versus registered ), type of placement (public versus private) or currencies. As we see it, the important topic will be how funding can be done.

Hence, the isolated view on the EUR Jumbo covered bond market may be less useful than in the past

However, as we believe the demands of rates investors will continue to be extremely important for issuers to get Jumbo issues executed, we focus the following discussion on the EUR Jumbo covered bond market.

A new record in terms of new issuers of EUR Jumbo covered bonds is expected for 2008.

The continuous demand from real money accounts for covered bond in the secondary market in recent days, even though not in decent size, makes us feel comfortable that the primary market for covered bonds will be the first to re-open. But as full-year results of banks will not be published in the first days of January, a new issuance wave is unlikely. However, almost EUR 40 bn of Jumbo covered bonds (26 benchmarks) were issued since September. With around 20 inaugural issuers to be expected in 2008 (more than in any recent year), new issuers should continue to be the main driving force of covered bond market growth, even though with less than EUR 30 bn, this driving force lost some of its power in 2007 compared to 2006.

Number of new covered bond issuers declined in 2007

0

2

4

6

8

10

12

14

16

2000

2001

2002

2003

2004

2005

2006

2007

Number of New Issuers


Whereas 15 new issuers tapped the market in 2006, only 12 new issuers showed up in 2007 (including CRH, which issued in benchmark format for the first time, and Achmea, which introduced a new covered bond program). As almost every big international bank (and also a lot of regional banks) intend to establish a EUR Jumbo covered bond program, we expect a new record number of new issuers tapping the market in 2008.



New issuers brought around EUR 30 bn of new EUR

Jumbo covered bonds

0

5

10

15

20

25

30

35

2000 2001 2002 2003 2004 2005 2006 2007


Due to decline in issuance volume in 2007, share of

new issuers’ volume even increased slightly

0%

5%

10%

15%

20%

25%

2000 2001 2002 2003 2004 2005 2006 2007


Germany – high Pfandbrief issuance volume expected In 2007, EUR 33.5 bn of EUR Jumbo Pfandbriefe have been issued. The outstanding volume of EUR Jumbo public Pfandbriefe amounts to circa EUR 180 bn, the outstanding volume of EUR Jumbo mortgage Pfandbriefe to circa EUR 64 bn. Due to high redemptions, the outstanding volume of German Pfandbriefe declined for the fourth year in a row in 2007. The same should hold true in 2008.

In the mortgage sector, the discussed increase of the LTV for residential mortgage loans to be eligible for Pfandbriefe, to 80% from 60% currently, could lead to higher issuance. However, we do not expect the amendment to come into force earlier than at the end of H1 2008, if at all. Hence, it may not have a significant impact on new issuance of Pfandbriefe. We expect EUR Jumbo Pfandbrief issuance amount to around EUR 40 bn (EUR 18 bn mortgage Pfandbriefe, EUR 22 bn public Pfandbriefe). We expect that new issuers like Deutsche Postbank will increase the issuance volume of mortgage Pfandbriefe. The pooling model of Landesbanks and

savings banks may also finally appear on the Jumbo Pfandbrief market. Given redemptions of around EUR 65 bn (circa EUR 13 bn mortgage, circa EUR 52 bn public), we see a continuous decline in the outstanding volume of EUR Jumbo Pfandbriefe. As non-Jumbo Pfandbriefe continued to work well during the crisis, issuers may easily shift between non-Jumbo and Jumbo funding and vice versa, depending on market conditions. HSH Nordbank is to issue its inaugural ship Pfandbriefe. We do not expect another bank to issue ship Pfandbriefe in Jumbo size. Deutsche Schiffsbank and NordLB will most likely stick to the non-Jumbo market.

Public Pfandbriefe continue to dominate the

(declining) EUR Jumbo Pfandbrief market

0

50

100

150

200

250

300

350

400

Jan-

95

Jan-

96

Jan-

97

Jan-

98

Jan-

99

Jan-

00

Jan-

01

Jan-

02

Jan-

03

Jan-

04

Jan-

05

Jan-

06

Jan-

07

0.00

0.40

0.80

1.20

1.60

2.00

2.40

2.80

3.20

3.60Total volume of outstanding Jumbo MortgagePfandbriefe in EUR bnTotal volume of outstanding Jumbo PublicPfandbriefe in EUR bnAverage size of Jumbo Pfandbriefe in EUR bn (RS)


German housing market not a major driver of Pfandbrief issuance With circa EUR 1.2 tn of total mortgage loans outstanding at the end of Q3 2007, the German mortgage market is the second-largest in Europe. UK ranks No. 1. With around 47% of the housing stock, rental housing dominates the German market.

German house prices continued to fall

85

90

95

100

105

110

115

120

Jan-

03

May

-03

Sep

-03

Jan-

04

May

-04

Sep

-04

Jan-

05

May

-05

Sep

-05

Jan-

06

May

-06

Sep

-06

Jan-

07

May

-07

Sep

-07

Existing Single Apartments

Existing Homes

Source: Hypoport



Moreover, subsidies for owner-occupiers were abolished as of 1 January 2006. Hence, despite a low owner-occupation rate of only 42%, new mortgage lending should not be a driving factor of mortgage Pfandbrief issuance. Also, for new issuers, the driving factor should not be sharp growth in new lending, but existing collateral or at least an existing good market position.

France – like in 2007 – strong new issuance expected Due to high lending volume, issuance of French covered bonds is likely to remain high. Uncertainty remains regarding issuers of French structured covered bonds, which may still be in the process of optimizing their funding plans, taking into account the new funding tool that has been extensively used by BNP so far. If market conditions improve, French structured covered bond issuance is likely to be very strong, amounting to up to EUR 25 bn. On the other hand, issuance of Obligations Foncières may amount to EUR 48 bn (in all currencies). Hence, depending on the market conditions in different currencies, EUR Jumbo Obligations Foncières issuance may amount to EUR 40 bn.

Secondary market spreads not a good indication for primary spreads Secondary market spreads will likely continue not to provide a good indication for primary spreads. E.g with 10Y Multi-Cedulas spreads at around ms + 32 bp and a new issue pick-up of 5-6 bp recently, the primary market spread is perhaps not where issuers want to tap the market. Moreover, the secondary market spreads are more indications than anything else. If a big block were to be offered, spreads would likely widen significantly.

EUR 30-38 bn of Jumbo Cédulas expected With 68 banks issuing Cédulas, the Spanish covered bond market is (together with Germany) the deepest in terms of participating banks. AyT still dominates the market with a market share of 21%.

With a market share of 21%, AyT still dominates the

Cédulas market in terms of outstanding volume

BBVASM15%

CAIXAC1%

CAJAMM10%

CAIXAB11%

AYTCED21%

BANEST6%

PASTOR1%

BANCLE2%

SANTCF0.5%

IMCEDI6%

PITCH0.5%

CEDGBP2%

CEDTDA9%

BPESP2%

BANSAB3%

SANTAN11%

CAGALI1%


Whereas EUR 55 bn of Cedulas were issued in 2005, EUR 65 bn (including EUR 4 bn Cédulas Territoriales) in 2006 and EUR 36 bn in 2007, we expect EUR 30-38 bn of Cedulas issuance in 2008. The lower expectation is not only based on concerns about access to the primary market, but on reduced lending volumes and lower volumes that are eligible for Cédulas funding due to the increased OC requirement of 25% and the reduced LTV for non-residential mortgage loans from 70% to 60% (reducing the maximum issuance volume of Cédulas). In January 2005, EUR 34 bn of Jumbo covered bonds were issued, EUR 17 bn of which were Cedulas. These times are gone, in our view. Moreover, we believe the issuance of long-dated Cedulas is quite unlikely in the next months. However, we do not expect the Cédulas market to remain shut, particularly not for short-dated Cédulas of broadly diversified banking groups. As the Bank of Spain and the Spanish Confederation of Savings Banks warned that the level of central bank borrowing by Spanish banks is not sustainable long term, Spanish banks will have to come to the capital market, providing investors a good position regarding spread negotiations. Even though in other banking systems the current amount of ECB lending may also not be sustainable for the long term, the Spanish economy is dependent on external funding, making the situation much more pressing compared to e.g. Germany, where Pfandbrief issuers typically find ways to use their own Pfandbrief issues for ECB lending.



Growth of Spanish Cédulas flattened in recent

months

0

50

100

150

200

250

300

01-99

08-99

03-00

10-00

05-01

12-01

07-02

02-03

09-03

04-04

11-04

06-05

01-06

08-06

03-07

10-07

0.0

0.4

0.8

1.2

1.6

2.0

2.4

2.8

3.2Total volume of outstanding JumboCédulas Territoriales in bn EUR (LS)Total volume of outstanding Jumbo MultiCédulas in bn EUR (LS)Total volume of outstanding JumboCédulas Hipotecarias in bn EUR (LS)Average size of Jumbo Spanish CoveredBonds in bn EUR (RS)


Domestic bid likely to remain low The domestic bid might increase somewhat given that Cédulas of other issuers will be eligible as substitute assets up to 5 percent. But given that the assets in Spanish insurance companies only amount to EUR 206 bn and in pensions are only EUR 83 bn, the domestic bid in Spain should remain limited, As a comparison, French life insurance companies already had AuM of EUR 800 bn in 2003. German life insurance companies currently have around EUR 650 bn of assets under management. German pension funds administer around EUR 350 bn. Hence, in our view, the Spanish bid for Cédulas will remain in the single-digit percentage range. And the legal amendment that allows substitute assets up to 5%, which also allows Cédulas, should bring no significant change. First, bank bonds are typically substitution assets; also in other legal frameworks, like German Pfandbriefe. However, it is unlikely that Cédulas issuers make extensive use of Cédulas from other banks in their cover pool. Even if this were done, the outstanding volume of around EUR 275 bn of Cédulas could lead to EUR 13 bn maximum demand. This may bring some small relief but not change the structural domestic demand deficit. This may also be a consequence of the fact that Spanish households have a much higher share of their wealth in property leading to the ownership ratio.

Impact of the strength of the legal framework overestimated Even though we think a strong legal framework for covered bonds is crucial, we suspect that the impact of the legal framework on spreads is currently overestimated. E.g. the legal framework for Pfandbriefe is, in our view, very strong. However, Fitch does not gives German Pfandbriefe the best discontinuity factors. In our view, the main reasons for Pfandbriefe trading so tight are their negative net supply and the strong domestic bid. The outstanding volume of German Pfandbriefe decreased for

the fourth year in a row. With the amendment of the legal framework, we believe Cédulas should be considered as founded on a sound legal basis. However, Cédulas trade as wide as a lot of structured covered bonds based on contractual law only. Due to the OC resulting from a high mandatory OC and an even higher OC taking into account the preferential claim on the whole mortgage book, the need to have substitute assets to ease liquidity concerns is not so pressing as in cases of covered bond with less OC.

UK covered bond issuance with big question mark The outstanding volume of EUR Jumbo covered bonds amounts to EUR 61 bn. With a market share of 42%, HBOS dominates the market.

HBOS still dominates the UK EUR Jumbo covered

bond market

NWIDE15%

HBOS42%

NRBS18%

BRADBI12%

ABBEY6%

YBS5%

HSBC2%


Although most individual issuers’ bank credit quality still looks sound, market access may require considerable spread concessions due to housing market concerns. In some cases, market access in general may be very difficult. UK issuers may try to increase their non-Jumbo issuance -- and in other currencies. With no UK covered bonds coming due in 2008, there is at least no refinancing pressure from the covered bonds side.



No UK covered bonds coming due in 2008 and 2009

Maturing

0

1

2

3

4

5

6

7

8

9

2004

2006

2008

2010

2012

2014

2016

2018

2020


So far, there have been no taps of UK covered bonds. GBP investors typically do not buy covered bonds. HBOS’ social housing backed covered bonds in GBP are currently the only exception in size. Given that the RMBS market continues to be very difficult, UK issuers might take an even harder look at covered bonds.

S&P recently compared different European countries’ dependence on RMBS for mortgage refinancing.

UK highly dependent on RMBS funding

0

5

10

15

20

25

30

UKIR

E GR SP NE ITBEL

GER FR

% o

f gro

ss le

ndin

g

Source: S&P

Not only is UK the country with the highest dependence on RMBS funding of mortgage loans, but this dependence has also increased strongly since 2002. The outstanding volume of UK RMBS amounts to over EUR 300 bn (as a comparison, the total outstanding volume of German mortgage Pfandbriefe as of 30 Sept 2007 amounted to EUR 212 bn). S&P also said it expects renewed regulatory focus on the adequacy of the current liquidity regime, and potentially higher liquidity requirements for higher-growth wholesale funded lenders in particular. In our view, such an increase in regulation would be positive from a credit quality perspective and hence support covered bonds, which, first and foremost are bank bonds.

UK’s dependence on RMBS funding increased steadily

0

5

10

15

20

25

30

35

2000 2001 2002 2003 2004 2005 2006 H1 07

% o

f gr

oss

lend

ing

Source: S&P

Given the difficult conditions in the primary market for RMBS, UK issuers will likely take an even closer look at covered bond (UK building societies are not allowed to issue RMBS; hence, they have not relied on RMBS in the past anyway).

Given the continuous negative news flow regarding the UK housing market, the primary market for UK covered bonds will remain difficult, in our view. However, as full recourse bank bonds should continue to better bid than segregated securitized mortgage portfolios, we do not expect the UK covered bond market to remain completely shut. UK issuers paying the right price should be able to find buyers. Given the ongoing uncertainly, we expect UK issuers to issue around EUR 20 bn.

Outstanding volume of UK covered bonds increased

continuously

0

10

20

30

40

50

60

70

2000 2001 2002 2003 2004 2005 2006 2007




Monthly issuance never exceeded EUR 6 bn

0

1

2

3

4

5

6

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

2003 2004 2005 2006 2007


Ireland – only mortgage ACS expected The outstanding volume of EUR Jumbo ACS amounts to EUR 37 bn. As WestLB covered bonds bank (quite successfully) focuses on non-Jumbo issues and Depfa seems to focus on USD issues in Jumbo format, the Irish market may see only mortgage-backed ACS, residential and commercial (issued by Anglo Irish) in 2008. Allied Irish bank successfully tapped its 2017 issue by EUR 175 m recently. Generally, Irish issuers might increase their share of taps, non-Jumbo and foreign currency issues. In EUR Jumbo Format, EUR 7 bn seem likely, depending on market sentiment.

Depfa still dominates the EUR Jumbo ACS market

DEPFA56%

BKIR16%

AIB19%

WESTLBCB9%


US – difficulties to access EUR covered bonds In our view, US issuers might well be willing to issue EUR 6-8 bn in covered bonds each. Given the continuing uncertainties in the US housing market, we doubt that US issuers have access to the EUR covered bond market. A successful USD issue could inject trust in the market and open EUR funding opportunities. A significant pick-up to other EUR covered bonds would likely be required. If default statistics do not deteriorate further in Q1 and Q2, investors might gain some new confidence.

Canada – very different than US AAA-rated EUR-denominated Canadian covered bonds should continue to benefit from a general shortage of AAA EUR Canadian exposure. Due to years of budget surplus, the government bond market is almost non-existent. We expect every major Canadian bank to set up a covered bond program. Some issuers like Royal Bank of Canada will likely even tap the market two or three times a year, partly also in USD. With over 30%, the first issue of Royal Bank of Canada received strong demand from Germany too. Fundamentally, the Canadian mortgages benefit from some specifics like the fact that there is no lending with a LTV higher than 80% (if so, a mortgage insurance is required for the part above 80%), interest rates are not tax deductible (other than in the US) and banks are allowed to charge prepayment penalties (preventing constant refinancing, as is common in the US). All in all, even though Canada might not be able to fully withstand an economic downturn in the US, we believe fundamentals of Canadian covered bonds look sound. We expect Bank of Montreal and Bank of Nova Scotia to tap the EUR Jumbo covered bond market in Q1 2008. Total 2008 issuance volume might amount to EUR 8 bn.

EUR 14 bn from Portugal and EUR 10 bn from Italy Portuguese covered bonds benefit from a strong legal framework and limited long-term supply. The whole mortgage market amounts to only around EUR 90 bn. Hence, despite a very low domestic bid, we expect Portuguese issuers to have no problems accessing the market. With Banco Espirito Santo, and Santander Totta further issuers should follow. Caixa Geral and BCP may tap the market, perhaps twice a year. Also public sector covered bonds might be issued out of Portugal. With Unicredit, Banca Opi (subsidiary of Intesa Sanpaolo SpA), BPU, BPM and Banca Carige expressing interest in issuing covered bonds, Italian covered bonds may well amount to EUR 10 bn at the end of 2008. In addition, CDP might again tap the market (under the current program or under the new law). Housing market concerns are as limited as credit quality concerns about Italian banks. However, the wide spread levels of BTPS may burden new issue spreads.

Nordic covered bonds – continued stability expected With DnB NOR, Sparebank 1, Terra and BN Bank four Norwegian issuers will have tapped the market, with issuance volume expected to amount to around EUR 7 bn. As acknowledged in a new S&P report, Swedish and Danish banks typically rely heavily on wholesale funding. However, the domestic mortgage/covered bond market strongly supports funding and liquidity. Nordea, Stadshypotek and SBAB plan to continue issuing in EUR. Besides SEB, Swedbank is expected to join the EUR Jumbo covered bond market in 2008. As SEB issued



Pfandbriefe in the past via its subsidiary SEB AG (typically trading very tight), the new issue via SEB AB should bring an interesting comparison between Nordic covered bond and Pfandbriefe against the backdrop of the question about the adequate spread difference.

New issue volume could amount to EUR 14 bn. Like Norwegian banks, we expect that Swedish banks will continue to look at CHF funding. With EUR 2 bn of expected issuance, OP Mortgage Bank will likely be the only EUR Jumbo issuer out of Finland. We believe Aktia is likely to continue to issue non-Jumbos. With Danske and Nykredit the most probable issuersk, in our view, EUR 4 bn might be possible out of Denmark. According to our understanding, Danske will have two pools -- one Danish domestic and one international with Norwegian, Swedish and Irish mortgages. In case of Nykredit, it seems unclear yet that they will tap the market in 2008. Due to stable banks and less housing market concerns, we continue to see Nordic covered bonds as fundamentally very safe. S&P commented that Swedish and Danish banks have a high wholesale funding dependency. In line with S&P, due

to strong domestic mortgage/covered bond markets, we remain confident on these markets.

New markets Being mainly a non-Jumbo market in recent years, Lettres de Gage might experience a revival in 2008 by issuance through Dexia. The total Luxembourgian covered bond market amounts to around EUR 34 bn (end-2006: EUR 28 bn), whereas the outstanding volume of Jumbo Lettres de Gage amount to only EUR 2.25 bn currently. Given the lack of public covered bonds supply, we are confident that if Dexia emerges as a Jumbo Lettres de Gage issuer, given the strong legal framework and the strong issuer credit quality, investor demand should be substantial, certainly depending on the spread. Japan, Australia, Greece, Turkey and Croatia are also potential new issuers. In our view, Japan, Australia and Greece are most likely to tap the market in 2008.

Bernd Volk (49) 69 910 31967



UK

Overview

We believe the UK economy will be impacted severely by the global deleveraging that we are seeing in financial markets

The combination of significant levels of household debt, coupled with large mortgage resets, and tightening credit standards will make life difficult for consumers

Rising foreclosures, increasing refinancing costs for smaller volumes and tighter margins on the other hand will put pressure on banks and mortgage originators

We see the risks skewed firmly to the downside compared to current market pricing. We retain a long position in the front-end of the UK curve, as well as 2Y-10Y steepeners

For the long-end we see the risk of a re-appearance of the LDI bid, as long-end valuations have normalized somewhat and pension fund activity has been slow in Q4

Consumers feel the heat

The leverage of the UK consumer is not only significantly higher than that of its mainland counterpart, but has also been increasing much more rapidly. Since 1999, the ratio of consumer debt to GDP has increased by 36 pp in the UK to over 100%, compared to only 13 pp (to 54%) in the Eurozone. This obviously makes the UK economy much more vulnerable to a change in the availability and cost of credit, in particular given the clear overvaluation in the UK housing market (see “A sharper slowdown”, UK Housing Watch, 4 December 2007). We see two main risks:

a tightening of credit standards, which will make the rolling of existing obligations more difficult

an increase in margins/rates as existing obligations are refinanced

We therefore believe we are likely to see more tangible evidence of the de-leveraging of financial markets and consumers in the UK, than in the Eurozone.

Leverage of the consumer poses risks to the economy

Household credit % of GDP

50%

60%

70%

80%

90%

100%

110%

1993 1995 1998 2001 2003 2006 2009


The evidence from the mortgage market suggests that we will be seeing a significant amount of loans resetting over the coming quarters of 2008, similar to the mortgage resets experienced in the US over the second half of 2007.

Loan resets to pick-up in Q4 2007

0%

2%

4%

6%

8%

10%

12%

Q 1 Q 2 Q 3 Q 4 Q 1 Q 2 Q 3 Q 4 Q 1 Q 2 Q 3 Q 4

2007 2008 2009


We also see these resets taking place at significantly higher interest rates, with even the prime sector experiencing rate shocks in the order of 20-30% (see chart below). Our colleagues in RMBS research have also collated some evidence of the non-price components of the lending decision with significant evidence of tighter borrowing criteria (see “Spotlight”, European Securitisation Monthly, December 2007). In the near-prime sector, LTVs have been lowered on average by 5



pp, while in the riskier segment of the non-conforming market, LTVs have been lowered by as much as 10 pp.

Resulting in large payment shocks

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

01 02 03 04 05 06 07

75% LTV Refi to cheapestmortgage

75% LTV Refi to 2yr fixedmortgage


With the availability of credit coming under pressure we would expect to see a further increase in foreclosures and repossessions, the early signs of which are clearly visible in the data.

Repossessions are picking up and may rise further

0

25

50

75

100

125

150

175

200

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

Claims

Orders


With the effects of the credit crunch so visible not just in the non-conforming mortgage market, but also the prime mortgage market, one would expect the tightening of credit in the unsecured lending market to be even more severe. The evidence we have suggests that certainly on a price basis we are seeing a more significant adjustment within this space. The Bank of England data suggests that for unsecured lending, rates have increased as much as 100 bp since the summer, while for mortgage lending, the rate increases have been closer to 25 bp. The buy-to-let market (16% of outstanding mortgages) in particular raises concerns: with average rental yields below mortgage rates this market has clear speculative elements; the dominance of specialized lenders in this sector highlights the significantly higher stress in funding.

Lenders under pressure, too

Lenders are also coming under pressure, not only because of the increase in delinquencies highlighted above, but also, because of the very sharp increase in funding costs across the entire liability spectrum.

Refinancing costs are rising not just in money markets

-40

-20

20

40

60

80

100

120

140

160

180

Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07

Weighted Prime RMBS Spread

Weighted Non-conf RMBS Spread

UK Covered Bond Spread

GBP Libor vs Base


Lenders are effectively caught between a rock and a hard place. Not passing the increase in funding costs onto borrowers leads to significant reduction in margins and erosion of profitability. Even banks with relatively diverse funding options such as HBOS report new mortgage lending spreads of around 5bps, with lending done on the assumption of a normalisation of base rate / libor at some point helping margins recover somewhat. However, as long as liquidity conditions remain as challenged as at present, substantial pressure on margins and earnings forecasts are inevitable.

The BoE’s announcement on Wednesday (providing 3M liquidity against a wide pool of collateral through a variable rate tender with the base rate as the minimum rate) should alleviate some of the funding issues at UK institutions. However, even assuming the BBR/LIBOR gap does normalise, the overall remaining increase in the cost of bank funding - both in savings and in wholesale markets - and an increased demand for balance sheet strength and greater liquidity by bank managers and investors means that credit will cost borrowers more and will be harder to source in future. This is expected to lead to further declines in leverage asset prices (esp. residential and commercial property), higher bank bad debts and lower economic growth.

Risks to the downside

We believe that the risks to consensus growth forecasts lie firmly to the downside. Since we are at unchartered levels in terms of the build-up in credit since 1999, the impact of the reversal of this trend will be very difficult to



gauge. What leading indicators are available are pointing to a significant cooling of economic activity since this summer.

PMIs point to downside risks for growth

40

45

50

55

60

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

-0.75

-0.50

-0.25

0.00

0.25

Weighted PMI index (lhs)

Change in repo rate, % (rhs)

BoEeasingrange

SoSource: Deutsche Bank

So while the market is indeed pricing in a fairly aggressive easing cycle from the Bank of England, we believe that the risks remain skewed to the BoE acting faster and cutting deeper than what is currently priced in. We therefore maintain our longs in the front-end of the UK curve.

Even relative to current market pricing we see upside

in long front-end positions in GBP

4.25

4.50

4.75

5.00

5.25

5.50

5.75

6.00

Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07 Jan-08 Jul-08

Base rate pricing (assuming 6 bp SONIA spd)

Base rate


We see two main risks to our view: First, the easing in monetary conditions that we believe is necessary to buffet the impact of tightening credit conditions could at least in part take place through a further depreciation in GBP. Second, a significant pick up in inflation through either a weaker currency or through passing through of the price pressures visible in the leading indicators, which would make it very difficult for the BoE to outcut current forward pricing.

We would therefore also maintain steepeners as a core position within our portfolio, which we believe should also perform on concerns that the BoE is acting too aggressively. The risk to this position is a return of the LDI bid in the long-end of the curve, which would also expose the 2Y-10Y sector of the curve to further flattening risk.

A review of the long-end

One of our main positions on the UK curve in 2007 has been our long-end normalization trade. We argued a year ago, that the main driver of the further inversion of the long-end of the UK curve over the course of 2006 had been the consistent underestimation of the Bank of England’s rate hikes. Stripping out the directionality of 10Y-30Y with respect to 2Y rates would therefore yield a much cleaner way of trading the long end of the curve. This measure of the excess flatness of the long-end has almost normalized over the course of this year, as the combination of corporate and government long-end and Linker supply has filled the demand from pension funds for inflation protection and duration.

Excess flatness of 10Y-30Y has corrected

-0.40

-0.30

-0.20

-0.10

-

0.10

0.20

0.30

0.40

00 01 02 03 04 05 06 07

GBP 10-30 slope regressed on 2Y


With hindsight, we had recommended taking profits on this trade somewhat early (see FIW 5 October 2007), as the LDI bid we had been expecting in November failed to materialize. Against the backdrop of reduced corporate supply and the risk of reduced Gilt issuance in 2008/09 relative to current expectations a re-emergence of pent-up LDI demand could renew pressures on the long-end of the UK curve.



But is now at risk from further de-risking of pension

funds

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

-

0.05

0.10

Jun-

05

Aug

-05

Oct

-05

Dec

-05

Feb-

06

Apr

-06

Jun-

06

Aug

-06

Oct

-06

Dec

-06

Feb-

07

Apr

-07

Jun-

07

Aug

-07

Oct

-07

Dec

-07

5000

5500

6000

6500

700010-30 residual

FTSE-100


With current deficits on aggregate having been closed we believe that the LDI bid we had been expecting for the end of the year in line with normal seasonality may instead come in early next year, as the de-risking incentives remain very powerful.

Given current surplus, the de-risking incentives are

powerful

-120,000

-100,000

-80,000

-60,000

-40,000

-20,000

0

20,000

02 03 04 05 06 07

FRS17 Surplus - FTSE 350 Co's (£m)

Source: Watson Wyatt

The changes to the PPF levy enacted and proposed may also provide further incentives to accelerate the de-risking of pension funds. The PPF levy is now charged up to a higher level of solvency, and the PPF has not ruled out returning to a risk based levy which would add significant pressure to the long-end.

Typical Q3 seasonal flattening pressure did not

materialize – risk for Q1?

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

0.25

feb mar apr may jun jul aug sep oct nov dec

%


Another key risk going into next year is corporate supply. Q4 linker and long-end issuance was at its lowest level since 2005, and the outlook for next year remains highly uncertain. PFI issuance is dependent on the support of monolines, and given the strains they find themselves under due to their involvement in the subprime market, this raises question marks over their ability to print significant new business in this sector. Also, issuance in USD and EUR remains easier for UK corporates than in GBP, given the relative market sizes.

Corporate issuance needs to recover for long-end

valuations to remain stable

0

1

2

3

4

5

6

7

8

9

10

Q105

Q205

Q305

Q405

Q106

Q206

Q306

Q406

Q107

Q207

Q307

Q407

Non IL (>15)

Index Linked

£b non-gilt sterling long dated issuance


Gilt issuance for the remainder of the fiscal year, in particular in the long-end of the curve is somewhat higher than in Q4 (but similar to Q1 07), with GBP 10.7 bn of long-dated and linker issuance expected compared to GBP 8.5 bn in Q4.



Gilt supply in Q1 reasonably high

Issuance Q1'08

0.00

0.50

1.00

1.50

2.00

2.50

3.00

S M L Inflation Linked

Jan-08

Feb-08

Mar-08

£ bn


However, the bigger risk is the fact that Gilt issuance is expected to come down reasonably sharply from 2007/08 to 2008/09, from GBP 58.4 bn to GBP 50 bn (see “UK Review & Outlook: The risks to Gilt issuance”, Focus Europe, 26 October 2007). There are also further downside risks to this forecast, since our economists see a reasonable chance that the current fiscal year is over-funded.

But 2008/09 issuance is lowest for three years

0

5

10

15

20

25

30

S M L I/L

2006/07

2007/08

2008/09

£ bn


Ralf Preusser (44) 20 7545 2469

Francis Yared (44) 20 7545 4017

GBP Derivatives

The second half of 2007 has turned-out to be the period when the excess leverage in the financial system finally began unwinding and volatility rebounded from the lows

GBP Gamma outperformed EUR Gamma since late August as it became apparent that the inadequate response of the authorities to the liquidity crisis, the Northern Rock debacle and the vulnerability of the UK economy were increasingly likely to result in rate cuts

The degree of inversion of the Gamma curve makes long forward volatility trades attractive.

The low level of GBP vega, the seasonal patterns in LOBO’s issuance and the likely lower availability of bank’s funding for local authorities makes long dated GBP vega attractive

Back in July, the UK Treasury announced that it was envisaging measures to promote the use of long term fixed rate mortgages in the UK. Assuming that prepayment risk will be borne by the lender, if the issuance of fixed rate mortgages does take off, it will generate additional vega demand which will be positive for Vega.

Given the recent market development and the interest rate reset risks faced by consumer, it is possible that fixed rate mortgages will now appear more attractive to borrowers. However, on the supply side, the appetite for launching new mortgage products which involve additional prepayment risk taken by the lender is in all probability low.

Thus we do not consider the development of the long term fixed rate mortgage market as being imminent although the most recent market development probably highlighted the benefit for the economy to have such a product on offer

2007- The return of volatility

The second half of 2007 has turned-out to be the period when excess leverage in the financial system finally began unwinding and volatility rebounded from the lows. Gamma in GBP (in line with other markets) bounced back from the lows achieved in March 2007 first on the back of the bear steepening of the curve and subsequently as a result of the liquidity and credit crisis.



GBP Volatility in 2007

45

50

55

60

65

70

75

80

Jan-07 M ar-07 M ay-07 Jul-07 Sep-07 No v-07

Gamma(3M 10Y)Vega (5Y5Y)


GBP Gamma outperformed EUR Gamma since late August as it became apparent that the inadequate response of the authorities to the liquidity crisis, the Northern Rock debacle and the vulnerability of the UK economy were increasingly likely to result in rate cuts. The steep inversion of the Gamma curve which has reached extreme levels is also giving rise to some interesting relative value opportunities which we explore below.

GBP Gamma outperformed EUR Gamma as the crisis

unfolded

45.00

55.00

65.00

75.00

85.00

95.00

105.00

Mar

-03

Jul-0

3

Nov

-03

Mar

-04

Jul-0

4

Nov

-04

Mar

-05

Jul-0

5

Nov

-05

Mar

-06

Jul-0

6

Nov

-06

Mar

-07

Jul-0

7

Nov

-07

EUR 3M10Y

GBP 3M10Y


GBP vega on the other hand continues to be mostly driven by LOBO activity. The unwind of some LOBO transactions and their corresponding hedges in Q2, when rates backed up, was positive for Vega. But increased LOBO supply in Q3 generated downward pressures on Vega.

GBP Volatility in 2007

45

50

55

60

65

70

75

80

Jan-07 M ar-07 M ay-07 Jul-07 Sep-07 No v-07

Gamma(3M 10Y)Vega (5Y5Y)


.

Outlook for 2008

Go long forward volatility As we mentioned above, Gamma has generally been well bid and short expiries in particular have outperformed the rest of the grid. As a result, the volatility surface is highly inverted.

GBP Volatility Surface

2y 5y 7y 10y 20y 30y

1m 94.9 82.6 78.1 74.3 66.6 64.4

3m 90.0 77.6 73.3 69.7 63.8 61.8

6m 87.3 75.3 71.2 68.2 62.0 59.1

1y 81.5 70.3 67.2 63.3 58.2 55.5

2y 73.0 65.7 63.0 60.1 55.7 52.1

5y 62.4 59.7 58.0 56.1 51.0 47.7

7y 58.7 57.1 56.0 55.2 49.8 45.8

10y 56.7 54.9 54.4 53.2 47.7 44.0

20y 49.3 49.1 48.3 47.1 42.9 39.3 Source: Deutsche Bank

Mean reversion in rates and in volatility will generally result in a volatility surface which will be downward sloping when volatility is high and upward sloping when volatility is low. However, a strongly inverted volatility surface (such as the one in place at the moment) can offer opportunities to be long forward volatility. We believe that such a strategy would be attractive as we expect the current volatile environment to persist

Hence, we like to go long forward volatility in GBP. We like to express this in terms of a gamma neutral calendar spread as Forward Volatility Agreements (FVA) are not liquid in GBP. Our favoured point to express this view is 6M fwd 6M 10Y Volatility (as discussed in FIW 23 Nov 2007).



6m forward-6m10y Vol is attractive

40

50

60

70

80

90

100

No v-00 No v-01 No v-02 No v-03 No v-04 No v-05 No v-06 No v-07

6m10y vo l6m fwd 6m10y vo l

6m fwd 6m10y


Go long Vega While forming our expectations for LOBO issuance in 2008, we consider the relative share of local authority funding from government vs. banks and the seasonal nature of this borrowing. Over the past two years, the share of financing from the government has gone up relative to bank financing. Local authorities have therefore funded more at a fixed rate from the government, relative to the financing from banks. Moreover there are some seasonal patterns in LOBO issuance: issuance in Q1 tends to be lower than in Q4.

More systematic LOBO issuance in Q4 relative to Q1

-200

-100

0

100

200

300

400

500

600

700

Q1 Q2 Q3 Q4

2003 2004 2005 2006 2007£mn


We look to find a similar pattern in long dated volatility. To remove the effect of trends affecting the whole volatility surface, we look at the residual of 15Y15Y volatility (proxy for long dated vega) regressed on 5Y5Y volatility (proxy for the first principal component of the volatility surface). We find that the residual tends to decrease in Q4 and increase in Q1. The one exception has been Q1 2004, when there had been unusually high issuance of LOBO.

Seasonal residuals of 15Yx15Y regressed on 5Yx5Y Q4 Q1

2007 0.32 -0.57 2006 -1.04 2.86 2005 -0.47 0.93 2004 -4.08 -2.48 2003 3.77 2.81 2002 -0.23 0.48 2001 -4.4 3.8 2000 -2.85 5.96 Source: Deutsche Bank

Given the current liquidity crisis, the rising cost of borrowing for UK banks and the pressures on their balance sheet, we would expect lower LOBO issuance next year as borrowing from government becomes more attractive. We therefore like to enter 2008 long GBP vega.

Development of long term fixed rate mortgage in the UK Back in July, the UK Treasury announced that it was envisaging measures to promote the use of long term fixed rate mortgages in the UK. The introduction of long term fixed rate mortgages would involve the hedging of the early prepayment risk as it is unlikely that consumers would be willing to pay repayment charges at the time of prepayment. Such prepayment risk would generate a natural bid for low strike receivers, as prepayments are more likely when rates are low. The DMO considered supplying low strike receiver swaptions to the market to cover such potential demand. However, we do not believe that it will be in the DMO’s interest to sell low strike receivers. The DMO is indeed more likely to face increased funding needs when the economy is doing poorly which is likely to be when rates are low. Thus, the DMO will be faced with additional liabilities resulting from the written options exactly at the time when pressures on public finance will be the highest. Thus, we would expect that if the issuance of fixed rate mortgages does take off, it would generate additional vega demand which may not be matched by the DMO issuance (if any) of swaptions. This would be a positive for the vol market.

Given the recent market development and the interest rate reset risks faced by consumer, it is possible that fixed rate mortgages will now appear more attractive to borrowers. However, on the supply side, the appetite for launching new mortgage products which involve additional prepayment risk taken by the lender is probably low in current market environment. Thus we do not consider the development of the long term fixed rate mortgage market as being imminent although the most recent market development probably highlighted the benefit for the economy to have such a product on offer.

Aditya Challa (44) (020) 7547 5966

Alessandro Cipollini (44) (020) 7547 4458

Gopi Suvanam (44) (020) 7547 5966

Francis Yared (44) (020) 7545 4017



EMEA Three key global factors will drive rates returns in

EMEA in 2008: high non-core inflation, the global growth slowdown and the continued effects of the credit crunch

These will be superimposed on the idiosyncratic factors of each market: the expectations and pricing of supply and demand of government debt, monetary policy and inflation expectations

We are bearish on the outlook for Poland and Hungary and bullish on Turkey and Israel

Over the course of 2008 we expect most rates curves in EMEA to steepen with the ZAR curve likely to disinvert by 100bp

We recommend exposure using real rates to take market exposure in Turkey and Israel over Q1 before switching to nominal assets for H2

Position for Q1 rate cuts using HUF 2Y-10Y steepeners; receive the PLN 3X6 FRA at 6.1%

Inflation is cheap in Poland: buy the Aug-16 CPI-linker and pay PLN 10Y

We maintain our short-term recommendation to pay ZAR 5Y rates. We also keep our Euro-entry risk-reward position in Slovakia through paying SKK 2Y against EUR 2Y

EM 2008: ten in a row?

In 2007, emerging markets produced their ninth straight year of positive returns to both external (USD-denominated) and local debt markets. Recovery after the disaster of 1997/98 was slowed by the bursting of the equity market bubble 2000-02, but subsequent years have seen strong returns and vigorous growth of emerging markets as an asset class. Investor inflows have reflected an on-going asset allocation into EM and become increasingly focused on local currency as opposed to external debt markets as yields have diminished in the latter and opportunities and access have grown in the former. International investor portfolio flows have boosted EM currencies, compressed bond yields and allowed governments to increase local market issuance and extend their debt maturity profile.

2008 will prove to be an interesting year for emerging markets as growth across G10 economies is set to slow significantly on the back of the credit crunch. In previous cycles, a downturn in G10 would have affected emerging markets through a combination of lower commodity

prices and lower demand for exports leading to a rapid worsening of the external accounts. The virtuous cycle of current account surplus and net FDI is replaced by a vicious cycle of portfolio and debt-creating flows and higher local interest rates. Whilst there should be some deterioration of the EM balance of payments next year, the fundamental improvement seen over past few years has been impressive and the average credit rating of EM indices is at all time highs. Furthermore, it is hard to discount the possibility of a continuation of the asset allocation into EM that has been so supportive whilst global credit markets were falling apart in 2007.

Most probably a robust response from EM is largely in the price but even so, it is difficult to take a too bearish outlook when the central global scenario is for weakness in H1 followed by rebound in H2. With this in mind, we adopt a relatively market-neutral approach to the 2008 outlook with a focus on both discrepancies between the local pricing of global stories and any idiosyncratic themes.

EMEA 2008: three key factors

EMEA is probably the most diverse of the three principal regions of EM (the other two being LatAm and Asia) and one of the best examples of this diversification is the monetary policy cycle. With Turkey and Hungary cutting rates, Poland, Czech Republic, Israel and South Africa hiking and the Slovaks in the waiting room for Euro-entry, 2008 provides the opportunity for a strong divergence between the returns of different rates markets.

In our view there are three key factors which will mostly determine returns in EMEA local markets over the course of the next twelve months. All three of these are global factors and the differential response of the EMEA markets will be determined by a combination of idiosyncratic fundamental factors and current market pricing.

Non-core inflation and the central bank response. Surging food and energy prices have pushed headline CPI readings much higher over recent months. Given that most central banks tend to target headline rather than core CPI, in some cases inflation now exceeds top of the target band. In many cases policy makers have alluded to the temporary or non-core origins of the inflationary divergence and therefore are looking through the CPI spike to lower levels over the next 12-18 months.



Core market driven growth slowdown and the central bank and fiscal response. The dilemma for EMEA central banks will be how to react to a slowdown in growth whilst inflation remains high. DB expects the US to slowdown considerably in H1 07 before rebounding in H2. If this is the case, then it will add further caution to any policy response to higher-than-expected inflation levels and in some cases lead to a “wait and see” approach to determine the knock-on effects on the local economy. However, any rebound in US growth should mean that the medium-term impact on EM economies is minimal.

On the fiscal side, there are risks that after several years of fiscal out-performance on the back of strong GDP growth, 2008 proves the first reversal. Given the strong position of most countries in the region this is not a big directional risk but will make a difference at the margin.

The tightness of global credit markets and their impact on EMEA in 2008. Current account deficits are widespread across the region, especially in the new EU member states. Whilst FDI has historically been the principal source of financing (and major cause) of the current account deficit, the growth of FX-denominated borrowing by households has become increasingly important. This has led to a reduction of the effectiveness of the policy rate in monetary policy and a deterioration of external fundamentals. Whilst this is principally a macroeconomic issue in markets such as the Baltic States and Bulgaria, which are too small and illiquid to be dealt with here, Hungary finances some 50-75% of its current account deficit through consumer borrowing in CHF and even JPY. Continued tightness of liquidity across G10 market and raised funding costs will reduce both the attractiveness and availability of this financing. Furthermore, the build-up of external debt increases both balance-sheet and cash flow vulnerabilities and raises the risk of currency devaluation and disruption of local equity and bond markets. Rates markets are especially vulnerable as any currency weakness will be countered by a tightening of monetary policy.

The second effect of the tightness in credit markets is the departure of money market rates from policy rates and policy rate expectations. This is most keenly shown in the USD, EUR and GBP markets, but as we have noted before, there has been a substantial widening of the spread between interbank and the policy rate in Poland and to a lesser degree in Czech Republic. Whilst this is likely being exaggerated by year-end funding concerns, the principal cause is a lack of ready funding for bank balance sheets, a factor that is unlikely to disappear as the calendar moves over to 2008.

These three factors are our central themes for 2008 and should lead to four main investment conclusions:

1) Buy CPI-linked bonds. These are available in four markets in the EMEA region and are conveniently linked to headline rather than core CPI so they will receive the full benefit of the high CPI. Furthermore, the lack of central bank action should keep real rates steady.

2) Receive rates in markets where implied tightening is too aggressive; and, of course, pay rates where too much easing is expected. Lower growth will mean less aggressive central banks. In 2008, the key switch in this investment strategy will be between the CPI-linked bonds mentioned above and nominal rates to take full advantage of any slowdown.

3) Curve steepeners. Given that we expect growth to slow only temporarily in core markets, any implied monetary loosening by EMEA central banks is also likely to be temporary. Along with G10 rates markets we expect curves will steepen as the longer-term growth outlook appears solid and inflationary risks remain.

4) Pay rates in countries which have a heavy reliance upon debt financing of current account deficits. There is only downside for receiver positions in these markets as currency weakness and tight local liquidity are both bad for rates.

Finally, there are two clear risk scenarios which need to be considered when evaluating any investment decision in 2008:

What happens if global growth falls by much more than expected? The idea of a US recession in 2008 is moving into the mainstream of market thinking. A recession in the US and corresponding drawdown of US demand for emerging market exports would not be a positive for EM. External accounts would deteriorate as current accounts moved from surplus to deficit and capital accounts suffered outflows as equity markets were re-rated; currencies would likely weaken. Given that most markets in this region have current account deficits, they might benefit from any falls in commodity prices, but surely this would be overwhelmed by outflows on the capital account.

The degree of weakening will likely be governed by the extent of portfolio outflows from the capital account. These would have a greater effect on the typical “high beta” countries like Turkey and South Africa, but central Europe would not be immune. Central banks would be



torn between lowering rates to stimulate growth and raising rates to combat the inflationary effects of a weaker currency. The overall outcome for rates market is therefore unclear although it is likely that Turkey, South Africa and Hungary will underperform the rest.

What happens if inflation does not come down? Clearly, one of our major assumptions is that the higher levels of non-core inflation seen globally are a temporary phenomenon and that once food and energy prices normalise, or at least stop increasing, then CPI will eventually fall. Currency targets aside, all central banks in the region have a single mandate – controlling inflation – and therefore it is imperative for our second strategy that there are few second round effects. With most labour markets in the region at their tightest levels for several years, a lack of second round effects is not a given. Indeed, in this way, South Africa could be ahead of the curve compared to the rest of this region as second-round effects are thoroughly entrenched and the SARB are fighting to restore credibility.

South Africa therefore forms a useful model and one which gives a uniformly bad outlook for rates in a higher inflation scenario. The CPI-linker recommendation will benefit from higher CPI but should also lose-out to higher real rates. Furthermore, curve steepener positions will quickly reverse into bear-flatteners. We seek to combat this risk through paying rates in countries where pricing is too dovish to hedge the positions in our favoured longs.

EMEA 2008: supply and demand

In general, fiscal financing plans across the region are little-changed relative to those in 2007. In most cases, government fiscal balances have benefited from the economic cycle and given that most growth projections are little changed from 2007, fiscal balances are projected to be largely in-line. The one key change in the region is Hungary where the benefits of the September 2006 fiscal package will be palpable: net issuance is set to fall about a third from the peak in 2007 and the debt management agency, AKK, have announced that auction frequency will fall from 2 weeks to 3 weeks.

The chart shows the level of monthly gross issuance expected for 2008 against that in the past two years. Changes in net issuance are largely masked by redemptions, which largely account for the expected surge in gross issuance in Poland. In Israel, the increase in gross issuance relates to our uncertainty over the level of non-market issuance. The Israeli Ministry of Finance has reduced issuance of non-marketable bonds over recent years but these bonds, issued to local pension and

insurance companies form an important minority share of domestic financing.

For the demand side, we can construct a model for government debt demand by splitting the investment community into three broad groups: local real money asset managers, local banks and non-residents. For this analysis we forecast demand from asset managers and local banks and assume that non-resident demand will be the residual.

Monthly average gross domestic debt issuance

0

1

2

3

4

5

6

7

CzechRep

Hungary P o land Slo vakia So uthA frica

Turkey Israel

200620072008

USD bn

Notes: budgeted gross issuance in 2007 on a monthly basis. Source: Deutsche Bank, Czech MoF, AKK, Poland MoF, ARDAL, SA Treasury, Turkey Treasury, Israel MoF

Local asset managers Institutional investors in the local market can be divided into three main groups: pension funds, mutual funds and insurance companies. We give a comprehensive overview of pension funds in the region and in the rest of EM in “The rise of pension funds in emerging markets”, published 13th July 2007. Most pension funds are defined contribution schemes which are essentially a growing pool of cash which is invested in local assets under a series of asset allocation rules. In the case of Poland, Hungary and Slovakia, these state-run schemes are growing rapidly, whilst the more mature systems in Israel and South Africa are more stable. Turkey and Czech Republic lack a large pension fund industry and therefore their importance as investors is smaller. South Africa aside, asset allocations tend to be quite conservative with large holdings of cash and fixed income. In Israel, the majority of most pension schemes remains invested in non-marketable “ear-marked” government bonds, although the restriction in supply of this debt is slowly moving the asset allocation into market instruments. Flows into pension schemes are predictable as they are reliant upon medium- to long-term variables such as demographics and wage levels.

Relative to pensions, mutual funds have a more aggressive asset allocation, especially in Poland, Slovakia



and South Africa. Flows and the marginal asset allocation of these flows are less predictable and tend to be correlated with market performance. Insurance company reserves represent the underlying capital of the firm.

Apart from in Hungary, legislation is stable and we have assumed no major changes in the overall asset allocation. In Hungary, the state funded pension system is being revised to force pension funds to adopt a greater risk-taking strategy. This should reduce the overall demand for government debt, but the change is being implemented gradually, beginning at the start of 2008, and should have little impact on flows next year.

Local banks Our expected net demand from local banks is also indicated in the table. Unlike the flows from local asset managers, bank demand for government debt does not grow uniformly and depends on a series of factors involving both their ability to expand their balance sheets and their willingness to buy bonds. This means that some years will see large allocations to government debt and other years will see very little and possibly even debt liquidations.

Currency composition of the deposit base USD bn

end-05 end-06 now value of 1%PLN 16% 15% 14% 2.2

HUF 16% 21% 20% 0.7

CZK 12% 11% 10% 1.3

SKK 22% 22% 26% 0.6

ZAR 3% 3% 2% 2.4

TRY 35% 38% 35% 2.4

ILS 35% 37% 38% 2.2

Notes: proportion of deposit base in FX at end-2005, end-2006 and currently. Also shown is the USD value of a 1% shift in composition. Source: Deutsche Bank, NBP, NBH, CNB, NBS, SARB, CBT, BoI

Deposit growth is key as it regulates the size of a bank’s balance sheet and therefore the ability to expand the asset base. Our focus is on local currency deposits because it is the funds from these which can be invested in local government debt. We assume organic deposit growth in 2008 at a rate of projected nominal GDP growth, but aside from this, the key source of local currency deposits is reverse currency substitution, or reduction in FX deposit levels. The table above shows the level of dollarisation in history and at present and also the USD bn value of a 1% change in the ratio. The two countries which stand out are Hungary, where the ratio has not changed since the 2006 crisis and Turkey, where any lira weakness will likely be met by USD selling. The increase in FX deposits in Slovakia is probably in preparation for Euro-entry in 2009. We have assumed that that proportion of FX deposits falls by 2% in Hungary and Turkey in 2008.

We have not assumed any reduction in dollarisation in Israel, although this could be an important factor for boosting shekel liquidity in 2008.

Time for some retrenchment of bank balance sheets? The past few years have seen strong growth for banks in the region as low interest rates coupled with financial innovation have allowed credit growth to take off in both local currency and FX. The FX loan story is interesting from the capital account perspective but does not concern us here, but the rapid growth of local currency credit is important because banks will actively substitute balance sheet space away from government debt for more lucrative consumer and corporate credit lines. This is shown to some extent in the chart below which shows the current YoY growth of local currency deposits, loans and government debt. This shows that in Poland, Czech Republic and Turkey, growth of credit has outstripped deposit growth by a factor of 2 in some cases. Growth of government debt holdings has also been variable but is generally lower than credit growth except in Hungary. We believe that Hungary is the exception due to the slowdown in growth and the surge in domestic government debt issuance over the past 12-18 months which has effectively crowded out all other forint lending activities.

EMEA bank b/s growth: time for some retrenchment?

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

30%

35%

P LN HUF CZK SKK TRY ZA R ILS

Depo sitsCreditGo vt debt

Notes: YoY growth of local currency deposits, loans and government debt on bank balance sheets. Source: Deutsche Bank, NBP, NBH, CNB, NBS, SARB, CBT, BoI

Strong credit growth is a sign of healthy economic growth and rising confidence. Furthermore, levels of credit in EMEA economies are generally low relative to those in G10 and therefore banks are not over-exposed to a deterioration in the credit quality of the underlying loans. Where they are exposed is to the funding of this loan growth. The chart below shows the level of balance sheet extension through the ratio of local currency credit to local currency deposits. By far the most mature market in the region is South Africa, where the ratio is just above 100%, but it has recently been joined by Poland, Slovakia and Turkey. Note that despite the strong growth in Czech



Republic, loans remain a much smaller fraction of their balance sheet.

EMEA: ratio of local ccy credit to local ccy deposits

60%

65%

70%

75%

80%

85%

90%

95%

100%

105%

110%

P LN HUF CZK SKK TRY ZA R ILS

Oct-06Oct-07

Notes: ratio of the amount of credit extended by banks in local currency to the amount of local currency deposits in the banking sector at Oct-06 and Oct-07. Source: Deutsche Bank, NBP, NBH, CNB, NBS, CBT, SARB, BoI

What does this mean? Set against an external context of easy access to credit and liquid money market, we can expect further credit growth next year. However, with major money markets seized-up since summer, it is likely that access to financing is becoming difficult at a local level too. We have noted that zloty liquidity has tightened significantly since the summer and has worsened into year-end, a clear sign that credit growth in Poland was a step too far. Furthermore, we are concerned that similar problems could occur in Turkey and Slovakia. These conclusions would indicate that balance sheet expansion and the ability to take on risk in 2008 is going to be rather constrained in Poland and likely Turkey and Slovakia too.

Funding rates and carry The final piece of the puzzle for banks is the attractiveness of government debt. As banks are carry investors, this can be assessed using the spread between the yield government debt against the cost of the funding base, for which we use the average marginal deposit rate. This is shown in the chart. Rising funding costs in CZK and PLN are making bonds less attractive, but bonds are becoming more interesting in HUF, ILS and SKK. Interestingly, the margin has always been by far the highest in Czech Republic, possibly explaining why banks are lagging in terms of provision of credit.

Putting this all together, there is a strong case for bond investment for banks next year if they can get hold of the necessary cash. In Hungary, funding costs have fallen and FX sales by residents should allow banks to buy bonds. Furthermore, Israeli and Czech banks still appear in a good position to add more debt to their portfolios. The only question is whether credit provision takes priority, which

is likely given that our growth expectations are only for a slight slowdown in both countries.

Spread between funding cost and bond yields

0

50

100

150

200

250

300

350

No v-06 M ar-07 Jun-07 Sep-07 Jan-08-60

-40

-20

0

20

40

60P LNCZKILS

SKKHUF (rhs)

Notes: spread between yield on 5Y government debt and the average marginal deposit rate. Source: Deutsche Bank, NBP, CNB, BoI, NBS, NBH

In Turkey, balance sheets are rather extended and funding costs in TRY remain high; a massive increase in debt holdings is therefore unlikely unless there are very large sales of FX by locals. The same is the case in Poland, Slovakia and South Africa, although the problem is most acute in Poland. It is difficult to see Polish banks buying much of anything in 2008.

The overall picture for supply and demand The table below sets out our expectations for net demand for government debt from local institutions and banks and shows non-residents with the residual.

Supply and demand in EMEA (USD bn) USD bn PLN HUF CZK SKK ZAR TRY ILS

net supply 15.8 4.8 3.1 0.8 -0.7 3.8 2.3

pensions 7.2 0.9 0.8 0.8 0.2 1.1 1.8mutual fds 2.7 0.8 0.2 0.4 0.7 0.2 2.3insurance 0.9 0.6 0.3 0.1 0.0 0.0 0.3

banks 0.0 0.6 1.0 0.0 0.0 0.8 0.8

non-resid 5.1 2.0 0.8 -0.5 -1.7 1.8 -2.92007 0.6 1.7 0.9 -0.5 0.7 0.0 -0.72006 1.8 1.9 1.0 0.0 4.4 6.6 0.2

Notes: amounts in USD bn. Net supply is net issuance of government debt and other figures are projected demand from local pension schemes, mutual funds, insurance companies and banks. The residual is assumed to be bought by non-residents. Non-resident demand in 2007 and 2006 shown in bottom two rows. Source: Deutsche Bank, Poland MoF, AKK, Czech MoF, ARDAL, SA Treasury, Turkey Treasury, Israel MoF

The supply-demand situation appears as strong as ever in South Africa and like in 2007, debt for government debt will be very tight in both Slovakia and Israel. Israel has the potential to be the best performer as many local institutions are in the process of switching out of non-market bonds into marketable bonds. In Central Europe,



Czech Republic looks as bad as ever due to the lack of institutional demand, but bank demand improves the situation. The countries to be concerned about are Poland, Hungary and Turkey. All three have had a negative funding gap in past years but non-resident demand has been strong enough to cover the gap. Next year could be tough with all three countries requiring inflows in excess of those seen in 2007. For Poland the situation is the worst; a lack of bank demand means that non-residents need to buy more than USD 5bn of zloty bonds. The last time that demand was this high was in the 2004.

We see several risks to our core scenario for supply and demand in 2008.

The turning of the fiscal tide With 2008 likely to see a slowdown in global growth after several years of strong expansion, we believe that there is a significant risk that for the first year in a while, fiscal results are going to undershoot their targets.

We believe that the risk of a fiscal overshoot is lowest in Poland where the new government will likely fix a tighter deficit target from that currently in the budget and assumed in our projections. Aside from this, we highlight South Africa as being particularly vulnerable given that the strong fiscal story is fully priced-in.

Ratio of cash reserves to financing requirement

0

1

2

3

4

5

6

P LN HUF CZK SKK RON TRY ZA R ILS

cash/financingcash/budget

Notes: ratio of government cash deposits to the total financing requirement (including debt redemptions) and the budget deficit. Source: Deutsche Bank, NBP, NBH, CNB, NBS, NBR, CBT, SARB, BoI, Poland MoF, AKK, Czech MoF, ARDAL, SA Treasury, Turkey Treasury, Israel MoF, Romania MoF

Stocks of cash in the government account The strong fiscal performance of the past few years has another legacy; large reserves of cash in the government’s deposit accounts. The ratios of the Treasury cash reserve to the 2008 financing requirement and the 2008 budget deficit are shown in the chart below. The high level of these ratios illustrate just how comfortable the situation is for Czech Republic, Slovakia, Romania and South Africa and explains why both Slovakia and Romania can cancel the majority of their debt auctions – they

simply do not need the cash. Given adverse market conditions we would expect these countries to reduce issuance or halt it altogether, providing some support to the market.

On the other side, the countries with a severe lack of cash are Turkey, Hungary, Israel and Poland. Turkey is unsurprising given the high level of short-term debt maturities, and this appears to be the case with Israel too. Indeed, given the likelihood of roll-over of short-term debt, the more pertinent ratio is that of the budget deficit to Treasury cash, in black. The fact that the government does not have enough cash to cover half the budget deficit underlines the precarious state of the Hungarian government finances and the requirement for continuous funding. Running a projection to monitor developments in the cash stock and assuming full financing of redemptions, Hungary would default sometime in Q2 08. Poland is also surprisingly vulnerable. Market volatility in these countries would be exacerbated by continued debt issuance; a further deterioration of conditions would lead to the use of the Eurobond markets to raise extra capital, like Hungary in September 2006.

Non-government debt issuance Russia aside, there is only one market in the region where local currency non-government debt issuance is economically significant: South Africa. Corporate debt issuance has exploded over the past few years and currently rivals government debt issuance in terms of volume. However, the duration impact is generally much lower and it seems that many local investors still prefer the security of SAGBs over corporate debt. The risk in 2008 comes from an increase in parastatal issuance by companies such as Eskom to finance infrastructure projects. Over the past few years, parastatal net issuance has increased from ZAR 3bn in 2006 to ZAR 17bn in 2007 and is budgeted to be ZAR 29bn in 2008, greater than the level of gross government debt issuance which we estimate to be ZAR 25bn. Furthermore, the inversion of the curve and demand for duration will mean that this debt will be raised at maturities of at least 10 to 30 years, increasing the impact on the market. Even taking these numbers into account, the supply-demand situation in South Africa is still robust, but represents an increase in domestic net issuance over last year to ZAR 13bn (USD 2bn) from ZAR 5bn in 2007. Net issuance of government debt and parastatal paper in South Africa will be at its highest level since 2005.

EMEA 2008: monetary policy pricing

EMEA contains countries at all stages in the monetary policy cycle. Given that it is widely expected that global growth will slowdown in 2008, but it is uncertain just how



deep the slowdown will be, our economists have a baseline and a “risk” scenario for global rates. The lower growth environment would likely lead to increase in risk aversion and weaker emerging market currencies over the course of the year. In the higher-beta countries such as Hungary and Turkey, the inflationary effect of the currency sell-off would probably lead to monetary tightening relative to our baseline scenario. Being at an advanced stage in its hiking cycle, South Africa is the exception – a sharp slowdown would weaken the rand, but we believe that growth would become the SARB’s overriding concern and our projected cuts would be accelerated. In the less volatile countries, such as Poland, Czech Republic and Israel, lower US and Eurozone rates will mean a more dovish monetary policy. We have Slovakia, expected to join the Eurozone in January 2009, following any ECB rate moves.

EMEA: 2008 monetary policy pricing

baseline risk pricingCZK 0 -25 50HUF -75 -50 -40ILS 50 0 90PLN 75 25 90SKK -75 -100 15ZAR -100 -150 -70TRY -225 -75 -225

Notes: change in policy rate implied under the DB baseline scenario, risk scenario and current market pricing calculated from the local money market curve or in the case of Turkey, from the FX-implied curve. Source: Deutsche Bank

The table highlights that there is value in rates in both scenarios in all countries apart from Turkey. We would like to highlight Poland as especially over-priced given that more than 75bp of the 90bp in the price are expected at the next 3 NBP meetings. Hungary has a similar level of short-term value: even though inflation is currently high, the NBH are still sensitive to the level of the forint. If EUR/HUF maintains the 250-255 level it is likely that there will be at least 25bp of rate cuts over the course of Q1 – these are not in the price. Another country which stands out is Israel, which as a consequence of the prevalence of USD investments in the country and the ease of switching between USD and ILS, tends to be highly sensitive to USD rates. Further Fed cuts and falls in USD rates will be beneficial for both the shekel and ILS rates.

For the moment we are less interested in the “value” present in South Africa, Czech Republic and Slovakia because inflationary risks should maintain current market pricing for some time.

EMEA 2008: inflation breakevens

Our final source of analysis for 2008 is on the inflation-linked debt markets in the region. Inflation-linked bonds are available in Israel, Poland, South Africa and Turkey, although only Israel and South Africa have more than one bond. Liquidity is generally lower than for the nominal bonds but interest has increased recently due to the higher-than-expected inflation readings in EM. Real rate swaps are also available in South Africa and Poland. The table below details the current pricing of these bonds. We have calculated the real yield, inflation-breakeven and also the average inflation expected to the maturity of the bond. The average inflation is calculated using the DB forecasts for 2007-09 and then the central bank inflation target. The “irp” is the inflation risk premium, or difference between priced and expected inflation.

EMEA: inflation-linked bond pricing real yield b/e infl infl expect irp

Israel Jun-10 3.09% 2.02% 2.32% -30Israel May-36 3.59% 2.58% 2.03% 55Poland Aug-16 2.74% 3.06% 2.98% 8ZAR 5Y 3.59% 5.98% 4.78% 120ZAR 10Y 3.15% 5.68% 4.64% 104Turkey 8.96% 6.78% 5.39% 139

Notes: real yield and break-even inflation calculated using current market prices. Inflation expectations are calculated using the DB forecast and the central bank target rate. Source: Deutsche Bank

Turkey, as ever, stands out due to the high level of real yields. However, as the analysis shows, the CPI-linked Feb-12 bond is expensive to nominal bonds as the breakeven inflation level is almost 1.4% higher than the expected inflation rate over the life of the bond. Given that we expect disinflation to return later in 2007, it is likely that the inflation breakeven will also fall later this year. Having said this, the current annualised carry on this bond is a formidable 33%, much higher than any other asset in EM. Given the high sensitivity of Turkey to any further food and energy price inflation, and the possibility of falls in the real yield with CBT rate cuts, we maintain a positive view on this asset.

The inflation risk premium is also high in South Africa. Inflation is also expected to fall considerably over the course of 2008, and even in the presence of second round effects, we believe ZAR inflation is expensive.

Inflation appears much cheaper in Poland and Israel. In Poland, the break-even inflation level has risen above 3% for the first time since 2004, but the DB inflation projection shows that inflation projections will provide almost the same return as with nominal bonds. Furthermore, the high inflation readings mean that the annualised carry on this bond will maintain levels of at least 10% until at least end-Jan.



The market which has been least affected by the recent rise in global CPI is Israel. Here, monetary policy expectations have inched higher, but short-term inflation breakevens remain close to the 2% inflation target and the break-even curve remains normal. Whilst shekel strength has been keeping inflation low for some time, we believe that the risks of at least some effects from higher inflation are non-trivial and therefore there is value in buying short-term inflation risk.

EMEA 2008: views for 2008

The arguments above centre on three idiosyncratic factors: the expectations and pricing of supply and demand of government debt, monetary policy and inflation expectations. To formulate our views we must set these against the global background described at the start of this article and the risks to the baseline DB outlook. These global factors include the current spike in inflation, slowdown in the US and much tighter credit conditions.

The other pertinent factor relates to the steepening of the USD and EUR curves expected by our G10 strategy team. Purely from a relative valuation perspective, this view will force most of the central European curves steeper, but the combination of lower liquidity, risk aversion and inflationary uncertainty inherent in the G10 call will also have their effects on EMEA markets. We therefore make curve steepeners the baseline case for markets across the region. Another steepening factor relates to the excess of supply over demand in Poland and Hungary (and South Africa, including parastatals). The only curves that we believe will retain their shape are Turkey, where rates are too high to be affected by moves in EUR and USD, and Israel, where the supply-demand dynamic remains strong.

EMEA views for 2008: a summary

Direction Curve ASW InflationPoland Short Steeper Wider WiderHungary Short Steeper NeutralCzech Rep Neutral Steeper NeutralSlovakia Neutral Steeper NeutralS Africa Neutral Steeper Wider TighterTurkey Long Neutral Wider TighterIsrael Long Neutral Tighter Wider

Notes: summary of DB views for 2008 for fixed income direction, curve steepness/flatness, asset swap spreads and inflation breakevens. Source: Deutsche Bank

The monetary policy view argues for a relatively bullish outlook for EMEA fixed income, especially given that monetary policy expectations are more hawkish than the DB view in all countries apart from Turkey. However, at present the key problem for monetary policy is the current high level of non-core inflation. Unless a central bank believes that rates are already too high – such as Hungary

and Turkey – it will be difficult for the market to price-out tightening until we have real evidence that an economic slowdown is underway.

With Hungary, we have longer-term doubts about the continuation of rates cuts beyond Q1 and due to our concerns over the viability of FX-lending in the current credit environment we believe that rates will not perform this year. The key problem for rates markets in central and eastern Europe is the FX-passthrough, which makes rates very sensitive to FX performance; essentially receiver positions in rates have the risk-reward of a currency view but with negative carry. Given the global environment and the current pricing of HUF rates, we take a negative view on Hungary in 2008.

We therefore keep a more neutral view on monetary policy across most countries and use inflation-linked bonds to express bullish views on real rates, pick-up the high current coupon and also offer some protection against a further surge in inflation. The key countries here are Turkey and Israel. None of the factors we have considered highlight Turkey as a particularly strong rates play and therefore for the moment we are relatively cautious and exploiting the high inflation through the CPI-linker. However, even at 14%, policy rates are high and after a period of stasis in H1 we expect that the market will begin to price-in a renewed easing in 2009.

Israel is probably the strongest pick in the region. Spreads to USD are wide and the curve is still priced for hikes that will likely not occur. The supply-demand dynamic is strong for 2008 and could be enhanced either through de-dollarisation or an acceleration in the switch away from non-marketable debt by local institutional investors. At present, inflation pricing is low and the inflation break-even curve is normal so the risks are skewed to a short-term inflation scare, but over the course of next year, rates should perform well either outright or as a spread to USD.

Poland is where we have the most comprehensive bearish view; there is fundamental value in Polish rates, both short-term and long term but 2008 will be a difficult year. Short-term expectations of rate hikes seem overdone, but the inflation outlook remains poor. Furthermore, Poland is likely to see a continuation of the banking sector stress that has raised WIBOR rates over the past few months and with banks unable to increase their POLGB allocations, issuance of debt is going to put steepening pressure on the curve and cause bonds to widen further relative to swaps. We would not be surprised if the PLN curve disinverted in 2008.



At present the inflation situation in South Africa is serious and there remain substantial upside risks for Q1. However, the SARB’s focus is moving over to the growth outlook and any signs of a slowdown should be enough to prevent further hikes. This should mean rate cuts in H2 08 but there remains the strong likelihood of zero cuts if growth remains strong or a lot more than currently priced if growth collapses. Either way we expect the curve to steepen. Long-term ZAR forwards remain too low and the spread to USD too tight. Also, the surge in parastatal issuance should remove the tightness of bonds and the continual flattening pressure on the curve. Finally, the political outlook is decidedly messy. At present we are not sure whether South Africa will bull- or bear-steepen but we expect 2Y-10Y to be a full 100bp steeper by end-08.

2008 is a big year in Slovakia as the decision will be taken in H1 whether to let the country join the Eurozone in January 2009. A “yes” would glue SKK rates to EUR and relative volatility will fall to zero; a “no” would push SKK rates much higher as we expect that such a decision would be accompanied by crown weakness. Currently Slovakia adheres to all the Maastricht criteria but there is still a possibility that the country could be rejected on the expectation that Slovakia can’t meet the criteria on a sustainable basis. The year is therefore likely to be quiet for SKK rates but the event risk of strong volatility remains.

Despite Czech Republic inflation likely reaching 6% in Q1, it is difficult to be too bearish on the outlook. The crown is strong and likely to exert a dis-inflationary influence next year. Furthermore, the supply-demand dynamic remains poor but is not as weak as in 2007. We expect the curve to steepen, but likely in-line with EUR.

Our top trades for Q1

With a shorter time horizon we have more flexibility to express ideas with an eye on market timing. These are largely in-line with our 2008 views given above, but with a bias to the shorter term developments expected in Q1.

1. Buy inflation in Poland, Israel and Turkey. As outlined above, the current spike in inflation has pushed the carry on inflation-linked bonds to very high levels; 33% in Turkey and 10% in Poland. Furthermore, although the Turkey Feb-12 is not cheap relative to nominal bonds, the Israel Jun-10 and Poland Aug-16 are cheap.

In addition to the inflation carry, we expect real yield compression in Turkey and Israel over the course of Q1. We target a real yield of 8.5% on the Feb-12 in Turkey and 2.5% on the ILCPI Jun-10 by end-Q1. In Poland we have a bearish view on rates and therefore do not expect real yield compression; we recommend buying the Aug-16 in combination with a payer in PLN 10Y IRS. We target a break-even inflation to 3.5% from just above 3% currently.

2. South Africa: stay paid ZAR 5Y for now. With inflation and monetary policy uncertainty set to remain high in Q1, we continue to recommend payer positions in ZAR rates for at least the next month. Given the low level of 5Y and 10Y rates and long-tenor forwards, payer positions have attractive carry and roll. We target 10% in ZAR 5Y from 9.6% currently.

3. Hungary: HUF 2Y-10Y steepeners, target -20bp from -60bp currently. The recent re-pricing of the Hungary curve is overdone. The NBH remain keen to cut rates and the conservative pricing of the market means that 2Y receiver positions are attractive. Furthermore, the upside in rates is limited as the market will only begin to price hikes if EUR/HUF moves well above 260. We expect the curve to steepen in Q1 and combine our 2Y receiver with a 10Y payer with a target of -20bp from -60bp.

4. Poland: sell PLN 3X6 FRA at 6.1%, target 5.6%. Higher inflation and higher policy rates are the headline-grabbing reasons to be concerned about the direction of yields, but we believe that the current liquidity squeeze highlights problems with zloty-funding in the banking sector. This will at least partly abate after New Year. We find the FRA rates very high and whilst some of the pricing reflects tight liquidity, some of the implied monetary tightening is likely to unwind in Q1, if the ECB move towards a cut as expected by our economists.

5. Slovakia: pay SKK 2Y against EUR 2Y, target 20bp from -17bp. This is a risk-reward trade that we have recommended for some time. We expect Slovakia to enter the Eurozone, but given that this eventuality is fully priced, we recommend paying SKK rates and receiving EUR on an expectation of “divergence”.

Angus Halkett +44 20 7547 3512



Japan

Overview Last week has been fairly volatile with overseas

markets being the main driver

We recommend cutting remaining outright and the payer spread positions

We discuss defensive butterfly positions in the Relative Value section

In preparation for the budget discussions that are likely to start next week, we discuss our budget outlook for FY2008

This week saw significant market volatility on Wednesday. While the global market was apparently disappointed with the refusal of the Fed to do more than cut the funds target and discount rates by 25bp each on Tuesday, the joint liquidity injection announced on Wednesday by five central banks (Fed, ECB, BoE, SNB, BoC) failed to engender a reversal in rates. Swap spreads reacted more positively to the news, tightening between 5.5bp at the front end and 2.5bp in 30Y on Thursday. Our recommended outright short position fared badly due to the sharpness of the rally on Wednesday which left no time to stop out. The 8Y sector is now trading practically where we recommended shorting it, after having dropped up to 65 sen. The payer spreads we recommended as a safer alternative last week were largely unaffected by this week's rally.

Going forward, flow-of-funds considerations are likely to play an increasing role. Figure 1 shows that the trend towards higher deposit balances and subdued demand for bank loans is still ongoing.

Deposits, loans, and net balance (SA, JPY tr)

420

440

460

480

500

520

540

560

Dec-99 Dec-01 Dec-03 Dec-05

-20

0

20

40

60

80

100

120

140Total Lending (SA)Total Depo (SA)Deposits - Loans (rhs)

Source: BoJ, Deutsche Bank

This growth in the gap between deposits and lending is now beginning to affect regional banks more strongly after they had shown a declining trend up to October (figure 2).

Deposits-Loans (JPY tr)

60

62

64

66

68

70

72

74

76

Dec-04 Jun-05 Dec-05 Jun-06 Dec-06 Jun-07 Dec-07

40

42

44

46

48

50

52

Regional banks I and II (lhs) City banks (rhs)

Source: BoJ, Deutsche Bank

While banks are accumulating cash balances, we do expect further flows into the JGB market. However, our outlook for the 2008 budget suggests a more reflationary environment ahead so that we see more value in the front end of the curve than at longer maturities.

In this context, the hedge we recommended on 9 November against the market becoming bullish on the BoJ, the 3Mx4Y-3Mx20Y conditional bull-steepener, has behaved in an interesting fashion (figure 3).



JPY swap 4Y-20Y spread vs 10Y rate

90

95

100

105

110

115

120

1.6 1.7 1.8 1.9 2 2.1 2.2

15 Jun - 8 Nov 9 Nov - 4 Dec 5 Dec - now

Bearsteep

Bullsteep


While bull-steepening/bear-flattening had indeed been the trend movement until about one month ago, the market switched to a bull-flattening/bear-steepening behaviour as we recommended the trade. Moves in the front end of the swap curve appear to have been driven more by changes in the Libor-OIS basis rather than moves in OIS itself until the past month, while the steepener is now determined mostly by the long end of the curve (figure 4).

OIS-Libor basis and 4Y-20Y swap spread

10

15

20

25

30

35

40

45

50


80

85

90

95

100

105

110

115

120

1Y OIS-Libor basis

4Y-20Y (rhs, inverted)


We still believe that positioning for further falls in both the OIS rates and the basis are useful positions, especially when established through options, so we recommend holding this trade. The low point in the P/L of the trade was -20 cents and it is now trading at 8 cents profit.

Budget outlook Although the concrete discussions about the FY2008 budget are likely to start next week, we find it useful to form some expectations about the general picture towards the end of this calendar year. In our view, the coming year is likely to see a significant expansion of the central government budget relative to expectations, and

with it an increase in JGB supply. Prime Minister Fukuda has directed the MoF to produce a budget draft that will lead to a reduction in new JGB supply but we are somewhat sceptical that such a budget would pass the legislative process. We expect instead that increased net JGB supply will arise from the impact of both lower tax receipts as well as higher discretionary spending. The lower tax receipts, and to some degree the higher spending, are due to a significant undershooting of GDP growth relative to government estimates. We note that the 2007 Debt Management Report issued by the Ministry of Finance presents planned budget figures based on 2.2% nominal growth in FY2007 and 2.5% in FY2008 (2.2% in the risk scenario) whereas we expect only 0.5% in FY2007 and 1.6% in FY2008.

On the tax receipt side, we see the most significant uncertainty in the higher rate gasoline tax which, if repealed completely, would remove about JPY 1.4tr from the budgeted income while road construction is unlikely to be reduced. Furthermore, the slowdown in economic activity relative to expectations that started in the second quarter has had, and will continue to have, a significant negative impact on corporate and income tax receipts. While tax receipts in FY2007 are higher than last year, the growth rate is lower than expected and will undershoot expectations for the first time in 5 years. We have pointed out earlier that the impact of such a slowdown on the central government budget is not straightforward. Shortfalls in central government receipts during FY2007 largely affect the FY2007 supplementary budget and subsequent budgets mainly through debt financing costs. Shortfalls at the regional level affect the central government budget in FY2008 though an increase in Local Allocation Tax (LAT) demands. Projected tax receipts are one input for the calculation of the LAT amount that is transferred from the coffers of the central government to the municipalities (prefectures, cities, villages, etc.) each year. When the realised tax receipts differ from the projections, the municipalities expect to see the shortfall made up in the next year's allocation. The technical situation has changed over the last few years and the municipalities are now less able to borrow long term against such future LAT entitlements which would cushion the impact on the central government budget. Based on press reports, we estimate the impact on the FY2008 budget to be around JPY 0.8tr. Other planned fiscal measures, such as changes to the distribution of Local Corporate Tax, dividend taxation, etc., are unlikely to have a major impact on next year's budget.

On the expenditure side, a flurry of stimulus programmes have been announced, such as spending on regional revitalisation, improvements in energy efficiency, etc. We do not have sufficient details on these programmes to be



able to evaluate them properly but estimate these baramaki costs to add some JPY 1-2tr to the expenditure side of the government. Overall, therefore, we expect the primary balance to worsen by about JPY 3-4tr relative to projections.

In this context, the use of reserve accounts in the Special Budget has emerged as an important topic in the public discussion. Reserve accounts are dedicated line items in public accounts that track earmarked payment flows in the cameralistic accounting approach used in Japan. Although these accounts are allocated to particular purposes, there is significant administrative leeway in their use. At a time when legislation is made difficult by an opposition majority in the Upper House, the availability of funds that are not subject to detailed legislative oversight seems like an attractive option.

The particular reserve account that is relevant for the JGB budget is inside the Fiscal Investment and Loan Program (FILP) special account which is part of the Special Account of the Japanese sovereign. The purpose of this reserve account is to accumulate funds to cover the funding gap between the long-term, low interest loans extended to the relevant borrowers (mainly zaito institutions and municipalities) and the shorter maturity FILP JGBs that are issued to fund these loans. The size of this reserve account is linked by ministerial decree to the volume of existing loans and was expected to reach JPY 20tr by the end of 2008. The MoF announced on December 12 that they will look to use JPY 9.8tr to redeem JGBs. Such a transaction has occurred already in FY2006 when JPY 12tr were transferred from the FILP reserve account to the JGB management account in order to reduce the volume of outstanding JGB.

We are somewhat sceptical that a reserve transfer will occur in such a size. The legal basis for these transfers is §17(2) of the Administrative Reform Promotion Law enacted in May 2006. This article states that an amount of JPY 20tr should be targeted for the transfer of surplus amounts from the Special Account in order to improve fiscal stability in the years 2006 to 2010. If the amounts that have been transferred in FY2006 and FY2007 are added, about JPY 15.6tr have already been used up. This leaves a buffer of only about JPY 4.5tr that could be used to hold down new JGB supply in FY2008. This would be enough to compensate for our estimate of the deterioration in the primary balance, but would leave no reserves for the next two fiscal years. Needless to say, a legislative change could be made so that more of the reserve fund could be released, but that is probably not possible without the asset of the opposition DPJ.

Even if the limit was higher than we believe and the MoF could indeed release JPY 9.8tr, we believe that the market impact of this balance sheet operation depends on what assets the reserve funds are invested in. Like the US Social Security trust fund assets, reserve fund assets need not represent net claims of the government against another party and may instead be claims against the government itself. A transfer of reserve accounts in the FILP special account would therefore on balance increase FILP JGB supply by the same amount as debt repayment increases or new fiscal measure JGB supply is reduced, leading to a net impact of zero.

Alexander Düring, +81 (3) 5156 6199

Relative Value We recommend paying 7Y against 5Y and 10Y swaps

in an equal weighted butterfly

One of the interesting features of the market over the last week has been the apparent disconnect between strong futures versus a somewhat weaker picture in other parts of the curve. As a result, the 7Y sector now looks rather rich (figure 1).

5-7-10Y equal-weighted butterfly (bp)

-12

-10

-8

-6

-4

-2

0

2

4

Dec-05 Jun-06 Dec-06 Jun-07 Dec-07


This butterfly is directional as shown in figure 2. However, the directionality is less pronounced than for some other butterflies along the curve and the butterfly is rich even when viewed against its directional component. We also do not believe that a bullish consensus has re-emerged in the market so that we are not to worried about the directionality at this time.



5-7-10Y swap butterfly vs 7Y swap rate

y = 10.192x - 19.718

R2 = 0.3043

-12

-10

-8

-6

-4

-2

0

2

4

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1


This trade (notionals 68.9:-100:36.0) has marginally negative carry of -0.08bp over three months and negative roll-down so that the total carry and roll-down is -0.33bp over three months.

For investors who would like to remove the directional bias of the trade, the PCA-hedged butterfly (risk weights 36.3:-100:-70.1, notionals 50.1:-100:50.5) offers slightly less upside (figure 3).

5-7-10Y PCA-hedged butterfly residual (bp)

-5

-4

-3

-2

-1

0

1

2

3

4

5

Dec-05 Jun-06 Dec-06 Jun-07 Dec-07


The carry on this trade slightly positive (0.15bp over three months) but that is compensated by negative roll-down so that the overall carry and roll-down is virtually flat.

Alexander Düring, +81 (3) 5156 6199



Asia We provide a list of strong views held in several

markets around the region.

For further details see our forthcoming 2008 Country Outlook pieces due shortly.

Asia: 2008 Views

China: Warehouse CNY exposure by hedging CNY 1Yx2Y NDFs with 12M NDFs. Elevated inflation, USD weakness, and compressing USD/CNY interest rate differentials should create a swirl of interest in long CNY positions in 2008. We see 12M forwards as too aggressive, implying 8.5% CNY appreciation, while forward NDFs, for example 1Y Fwd 2Y, are a much cheaper way to take long CNY risk, currently pricing in appreciation at a pace of only 5.5%. For now we recommend hedging the latter (1Yx2Y NDFs) with the former (12M NDFs) in 1:2 ratios as a way of warehousing long CNY risk and arbitraging the NDF curve at the same time by taking in 2% in postive roll down per month (per unit 1Yx2Y NDF). A reduction in the pace of implied CNY appreciation in both tenors should provide an attractive opportunity to lift the 12M NDF hedge, leaving the trade exposed to faster CNY appreciation. [FZ/MH]

India: OIS Steepeners as a way to position for the turn in the interest rate cycle. Implement on 1M forward steepeners on the OIS curve (2Y/5Y). Current levels 8bp (mid). Liquidity remains primarily driven out of balance of payments inflows. With the overall liquidity management of RBI more streamlined now, and as RBI gets increasingly comfortable with the trajectory of monetary aggregates over the course of 2008 (along with slowdown in the real economy), we expect liquidity conditions to improve and front end yields to come off and the curve to steepen. We look to translate this position into outright long cash bonds position once RBI initiates its rate cutting cycle. [SG]

Korea: Buy Korean assets swapped into USD Buy 2Y MSBs swapped into USD, target Entry Libor +150bp. While we still expect volatility in USD/KRW basis underlying this trade we think foreign investors are willing to match the weakening position of banks in the market at the right price. As the USD funding situation improves globally going into 2008 and exporters hedging needs recede, KRW assets swapped in USD should richen. For investors with longer horizons 5Y KTBs swapped in USD around Libor+100bp also provide attractive value. Implement forward steepeners on the KRW IRS curve.

The unwinding of structured has led to extreme flattening of the IRS curve leaving the 1Y and 3Y relatively cheap and the 5Y and 10Y rather rich. Combinations such as 3M forward 2Y/10Y below -20bp or 6m forward 1Y/5Y around -40bp seem attractive. [CC]

Hong Kong: Add to 2Y/10Y curve steepeners on weakness: The conflicting forces of higher long-term US and China rates and increasing liquidity in the HKD money market have steepening implications. We believe that flattening to around +50bp provides an excellent opportunity for HKD curve steepening positions. We target a move above +100bp Pay the belly of HKD 2Y/5Y/10Y IRS: While 2Y/10Y IRS shows significant flatness the pressure on the 5Y point is even more extreme in relation to the wings of this butterfly. We suggest paying the 5Y point between -5bp and flat to take profit around +15bp while 2Y/10Y remains below 100bp. We look for a break above +30 bp as 2Y/10Y slope steepens further. [CC]

Cross Rates: Receive SGD rates against MYR on contrasting inflation implications. Neighbors Singapore and Malaysia seem to be sharing some similar underlying inflationary pressures, with two key distinctions that imply contrasting market implications. First, Singapore's transparent use of currency policy to suppress inflation means that an aggressively appreciating currency will attract heavy capital inflows from model-oriented hedge funds. As we saw in Q2 this year, these excessive inflows with eventually lead to lower interest rates as the MAS slows down sterilization to discourage inflows. Secondly, in the case of Malaysia CPI inflation is muted by subsidies on fuel prices, whereas Singapore's energy costs are showing up more clearly in headline CPI. Eventually we should see Malaysian authorities reducing subsidies, once fiscal costs mount, sending inflation higher. Heavier supply and the threat of higher interest rates should cause 5Y MYR swap rates to underperform SGD over the course of 2008. [MH]

Christian Carrillo, Tokyo, +81 3 5156 6206 Martin Hohensee, Singapore, +65 6423

Sameer Goel, Singapore, +65 6423 Feng Zhao, Singapore, +65 6423



Peripheral Dollar Bloc Strategy The outcomes for the $-bloc markets will depend

critically in developments in the US and globally. A global soft-landing will eventually see the RBA and the BoC tighten, with possibly even the RBNZ having to move upward if housing recovers. This will have implications for yield curves in particular.

A hard-landing by the US and the implications this has for global growth will be very negative for the $-bloc markets given the importance of the commodity boom to their recent success. In this case the BoC will keep easing through 2008 and both the RBA and RBNZ could join it by year-end.

Fate of the income shock is a key unknown in 2008 A key “big picture” trend this decade has been the emergence of China, the strength of global demand and the associated boom in commodity prices. This boom has provided a strong boost to the peripheral $-bloc economies, though the extent of the boost has varied across the three and over time. In total since 2000 Australia has benefited the most as a whole, though NZ’s terms of trade has risen the most in the past two years.

The terms of trade have risen across all the peripheral

$-bloc

90

100

110

120

130

140

150

Mar-00 Mar-01 Mar-02 Mar-03 Mar-04 Mar-05 Mar-06 Mar-07

CAN terms of trade

AUD terms of trade

NZD terms of trade

All set to 100 at Mar-00

Source: DB Global Markets Research, Stats CAN, ABS, SNZ

To a large extent the income boost from the positive terms of trade shock explains much of the recent economic history of these three economies. For instance, it explains why income growth has been so strong in Australia even when there have been times in the past few years when GDP growth has appeared to be relatively subdued. In our view the recent strength of NZ’s terms of trade, with dairy prices making a key contribution, helps

explain why the NZ economy has proved reasonably resilient despite aggressive RBNZ rate hikes.

How the external environment develops in 2008 is, in our view, the key uncertainty facing these three economies. If the global economy merely slows to trend, as seems to be the consensus forecast at present, then we suspect that by the end of 2008 the RBA and BoC will have tightened from current levels and the RBNZ may well have as well. If, however, the global economy slows sharply then the BoC’s overnight rate will likely be much lower than present levels and the RBA and RBNZ will likely have an easing bias (if they haven’t already moved to cut their cash rates). The outcomes for yield curves, swap spreads and so forth will be very different depending on which of these scenarios pans out.

BoC priced to keep easing, RBA priced to hike, with the RBNZ seen as on-hold In terms of current market thinking, 2007 draws to a close with the three markets taking very different views on the outlook for 2008. The CAN front-end is pricing a series of rate cuts from this point, with the Dec-08 BAX contract implying an overnight rate of less than 3.75%. The AUD IB contracts are pricing a cash rate of 7% by the second half of 2008 (i.e. one rate hike), while the NZ front-end effectively sees the RBNZ on-hold for 2008.

From a purely domestic standpoint we are broadly comfortable with market pricing for the RBA and RBNZ at this point, while that for the BoC looks out of step with the strength of the economy. Yet last week’s actions by the BoC make it clear that domestic economic developments are not foremost in its mind at present. Rather, its complete focus is on the downside risks created by the US economy and developments in credit markets. While this remains the case the BoC may continue to ease even if the domestic data provides little support.

If US outlook improves then short the CAN front-end, implement the flattener and look for the 10Y CAN to trade well above the 10Y UST At this stage we think there is simply too much uncertainty about the US outlook to take a bearish stance on the Canadian front-end. If the US uncertainty is resolved in a favourable way and the downside risks diminish then we would look to go short the CAN front-end. In which case, a companion trade would be to look for the curve to flatten as the BoC returns to a tightening



bias. But this stance is unlikely to make sense while the Fed is still in the process of cutting rates.

The CAN curve and the cash rate

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Jul-05 Dec-05 May-06 Oct-06 Mar-07 Aug-07

2.0

2.5

3.0

3.5

4.0

4.5

5.0

2Y/10Y CAN spread (LHS)

BoC cash rate - inverted (RHS)

Source: DB Global Markets Research, Datastream

As far as the 10Y CAN/UST spread is concerned, we think it should head into positive territory regardless of which scenario turns out to be correct. When we review domestic conditions in Canada and the US we think the market should have a clear expectation that the Fed Funds rate is going to fall well below the BoC’s overnight rate. While this is priced to some extent into the front-end of the curve, it is not appropriately reflected in the 10Y spread in our view. As such we see 10Y CAN as expensive versus the US.

The RBA will retain a tightening bias until evidence emerges of a sustained slowing in domestic demand As part of its revamped communication policy the RBA has published minutes for all the board meetings since Glenn Stevens became the Governor of the Bank, i.e. back to October 2006. In all these minutes the RBA had a tightening bias, which we don’t find overly surprising given the strength of the economy, the tightness of the labour market and so forth. And so long as these factors remain in play we think the Bank will retain a tightening bias.

But clearly the Bank hasn’t acted on its tightening bias at every meeting since October 2006. Whether the bias is acted on in the first half of 2008 will depend critically on global developments, in our view. If the global outlook remains highly uncertain then we think it will take especially strong domestic news to get the RBA over the line for further rate hikes. This means our central view is one which has AUD front-end pricing moving in a relatively narrow range until the current uncertainty is resolved one way or the other.

Absent an expected start to an RBA easing cycle the AUD curve will remain inverted in 2008 In thinking about the likely evolution of the AUD yield curve in 2008, the key in our mind is the market’s expectations for the RBA. Unless the market starts to price an easing cycle we think the yield curve will likely remain inverted for the duration of the year. Indeed, if our official forecast for more rate hikes is correct then there will likely be pressure on the curve to make new lows during the year.

3Y/10Y ACGB futures curve

-0.70

-0.60

-0.50

-0.40

-0.30

-0.20

-0.10

0.00

0.10

May-06 Aug-06 Nov-06 Feb-07 May-07 Aug-07 Nov-07

3Y/10Y ACGBfutures spread

Source: DB Global Markets Research, Reuters

Clearly expectations for the RBA will be sensitive to the global news flow, especially developments in the US. But as we have seen over the past few months bad news from the US may not be enough to dissuade the RBA from tightening if the news on domestic inflation pressures remains negative. This means that the Q4 CPI in late January is a key event for the curve outlook in the early part of 2008. A poor CPI result that cements in a February rate hike will pressure the curve to flatten further almost regardless of global developments.

AUD swap spreads exceed 100bp and then pullback In the first half of 2007 AUD swap spreads barely moved. 10Y spreads started to widen in June as rate markets sold-off, and then from July spreads have been subject to the impact of the credit crunch. Swap spreads widened sharply in November as credit fears re-surfaced, with spreads ultimately pushing above the levels reached in 1998 during the height of the LTCM crisis. Since that peak they have narrowed, especially at the front-end of the curve.



AUD swap spreads

30

40

50

60

70

80

90

100

110

Apr-07 May-07 Jun-07 Jul-07 Aug-07 Sep-07 Oct-07 Nov-07 Dec-07

10Y ACGB ASW spread

3Y ACGB ASW spread

Source: Deutsche Bank, Reuters

The influence of credit on long-end swap spreads is clear We would typically argue that the yield curve is the key influence on AUD swap spreads over time. Given that the yield curve reached a new low in early November some upward pressure on swap spreads from this source would have been expected. But the curve has steepened from that point, yet swap spreads pushed sharply wider for the next few weeks. This is because credit concerns have taken over as the dominant influence on 10Y swap spreads, with funding pressures an added factor at the front of the curve. The link between credit and swap spreads is evident in the following chart.

10Y AUD swap spread and Itraxx

45

55

65

75

85

95

105

115

Jun-07 Jul-07 Aug-07 Sep-07 Oct-07 Nov-07 Dec-07

15

20

25

30

35

40

45

50

55

60

65

7010Y AUD swap spread(LHS)5Y AUD Itraxx (RHS)


This suggests that if we are still bearish credit we should be looking for swap spreads to widen further.

We look for short-dated swap spread to narrow in early 2008 As we see it short-dated swap spreads are being impacted by liquidity pressures, as happened in August, as well as credit concerns. We think these funding

pressures will ease in early 2008, which should see short-dated swap spreads narrow even if credit concerns remain. To take advantage of this we will look to receive 3Y swap versus the March-08 bond futures contract. As far as 10Y swap spreads are concerned, we think the key is the direction of credit spreads.

Swap spreads to finish 2008 below current levels As far as our outlook for all of 2008 is concerned, we expect to see swap spreads quite a bit narrower than current levels. This is because we think the extreme risk aversion priced into bank spreads will likely ease as investors gain comfort that the worst of the loss disclosures is past. But with the curve likely to remain inverted, swap spreads should remain wider than their long-run equilibrium levels of around 40bp.

Current 10Y ACGB/UST spread is the widest since the RBA gained formal independence 2007 has been relatively unusual in that the 10Y UST has rallied quite strongly over the year as a whole but as of 12 December the 10Y ACGB was actually above its starting point for the year. In 2007 the movement in the 10Y ACGB/UST spread overwhelmed the direction of the US long-end. But the 10Y spread had to double to its widest level in more than a decade to offset the strong Treasury rally.

10Y ACGB/UST spread

-0.50

0.00

0.50

1.00

1.50

2.00

2.50

3.00

Jan-96 Sep-97 May-99 Jan-01 Sep-02 May-04 Jan-06 Sep-07

10Y ACGB/UST spread

Source: Deutsche Bank, Datastream

Of course, the 10Y spread doubled because the market went from expecting a 12 month ahead RBA cash/Fed funds spread of a little more than 150bp at the start of 2007 to one of almost 400bp at the time of writing. This gives us a sense of what it will take for the 10Y ACGB to again miss out completely if 2008 proves to be as good a year for the 10Y UST as 2007.

That is, if the 10Y UST was to put in another strong rally in 2008 the 10Y ACGB will only miss out on at least some of this rally if the market concludes that another couple of



hundred basis points of Fed rate cuts, on top of what is already priced, has no implications for the RBA. We struggle to believe the market will draw this conclusion if the US economy has a hard landing. So in the event we have another strong rally in the US we think it likely that the long-end of the ACGB curve will participate more fully in the gains than was the case in 2008 (though the 10Y ACGB/UST spread will still widen well beyond 200bp in this instance).

Spread compression to the US is our central expectation A 10Y UST rally to 3% is not our central case for 2008, however. Rather, we expect the market to eventually conclude that aggressive Fed action has lessened the severe downside risks to the outlook sufficiently to push long-end yields higher. Thus by the end of 2008 our thinking is that the 10Y UST yield will be higher than now.

The question then is whether the movement in the 10Y spread will dampen or amplify the transmission of the move in the 10Y UST to the 10Y ACGB? In our view the spread is likely to dampen the transmission of the US move. That is, we see the 10Y spread narrower by the end of 2008 than it is now. But just as we needed a massive widening in the 10Y spread to completely offset the 10Y UST rally in 2007, we will need a massive compression in the spread in 2008 to allow the 10Y ACGB to rally if the US long-end does sell-off. We think such a move in the spread is possible, but unlikely. Our central scenario for yields and spreads over 2008 is presented in the table below.

10Y ACGB yield and spread forecasts

10Y yield

10Y ACGB/UST spread

Mar-08 5.75% 200-225bp Jun-08 6.25% 150-200bp Dec-08 6.50% 100-150bp Source: DB Global Markets Research.

The scenario under which the 10Y ACGB yield can fall even if the US long-end is selling-off is one in which the US sells-off quite modestly and, for some reason, the AUD front-end prices a series of RBA rate cuts regardless of the move in US rates. While this combination is not out of the question it strikes us as reasonably unlikely. Much more likely, in our view, is the growing belief in an eventual US turnaround (as the Fed funds rate falls sharply) against fairly stable expectations for the RBA. This produces a much steeper UST curve, a higher 10Y UST yield, a sharp narrowing in the 10Y ACGB/UST spread and a modestly higher 10Y ACGB yield.

Unchanged OCR most likely scenario unless the global economy weakens sharply or housing recovers Our forecast is for the RBNZ to be on hold in 2008. Given the slowdown in the housing market that is well underway, the forecast of an unchanged OCR would seem to be overly conservative. But last week’s RBNZ Policy Statement highlighted that a weak housing market does not necessarily mean a substantial improvement in the inflation outlook. Indeed, the RBNZ’s inflation forecasts for 2008 actually deteriorated even as the Bank revised down its growth forecasts somewhat and lifted its assumed level for the TWI.

In our view the OCR will only end 2008 lower than its current level if the outlook for inflation improves dramatically. And it needs to be more than just an improvement in headline inflation. We think domestic price pressures have to ease in a sustained fashion.

The thing that would most likely change the inflation outlook for the better in a material fashion, in our view, is a sharp global slowdown. This would remove the strong commodity prices that have underpinned the economy for a number of years and are set to boost it further in 2008. A down-turn in commodity prices associated with a sharp global slowdown would likely impact on NZ quite severely.

Another scenario that could see a change in the OCR in 2008 is one in which the global economy remains strong, underpinning the income side of the economy (both directly and via expansionary fiscal policy) and allowing the housing adjustment to work its way through without a serious impact on the broader economy. In this scenario NZ gets to late 2008 with signs that housing is starting to recover without the inflationary imbalances having been corrected to any degree. At the same time the start-up of the NZ Emissions Trading Scheme in 2009 adds further upward pressure to the inflation outlook. In this scenario the RBNZ may feel it has little choice but to tighten further.

Given current pricing and the range of possible scenarios we are neutral the NZ front-end at present. We think the next big test domestically will be the Q4 CPI data on 17 January. A reversal of the downtick in annual non-tradable inflation recorded in the Q3 data, which we think is quite likely, will underscore the challenging inflation environment facing the RBNZ and keep the market from pricing an early start to the easing cycle.

Stable OCR suggests little scope for material curve steepening in 2008 In our view the key determinant of the 2Y/10Y slope in NZ, and the rest of the $-bloc for that matter, is the direction



of short-rates. Unless short-rates start to decline we doubt the NZ curve can sustain anything other than a modest steepening from this point.

The yield curve and short-rates

-1.25

-1.00

-0.75

-0.50

-0.25

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

Apr-99 Jun-00 Aug-01 Oct-02 Dec-03 Feb-05 Apr-06 Jun-07

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

2Y/10Y swap slope (LHS)

3M rate - inverted (RHS)


Some modest steepening is possible even if the OCR doesn’t move. Against stable expectations for the RBNZ a US sell-off could see a steeper curve. There have been occasions this year when the direction of the US long-end has been important for the curve. But of its own the US long-end can only cause a relatively small shift in the NZ curve. The key to a material steepening of the curve in NZ is a lower cash rate. Until we are confident of this we will not be recommending the curve steepener.

What about looking for a flatter 2Y/10Y curve? The curve has actually flattened notably in the past few months as the US long-end has rallied. It is a challenge to consider the flattener when the curve is already as inverted as it is. Yet this would have been the right strategy for a number of years, other than for a brief period at the start of 2006.

Given the global backdrop we don’t think implementing the 2Y/10Y flattener makes sense from a risk/reward perspective. We think the prospect of a hard landing globally is too high to support such a trade. But if this risk starts to fade and we see any signs of stabilization in the NZ housing market then looking for a flatter curve becomes a more attractive prospect.

With regard to the front of the curve, we note two prime influences: the direction of the cash rate and the direction of the 2Y swap. Which has the most influence depends on the stage of the tightening cycle. Early in the cycle the direction of the cash rate matters the most, with the curve flattening as rates move higher. However, in the later stages of the tightening cycle, when the market thinks it is drawing to a close, the direction of the 2Y swap takes over.

The 1Y/2Y slope – often a function of the 2Y direction

-0.50

-0.40

-0.30

-0.20

-0.10

0.00

Jan-06 Apr-06 Jul-06 Oct-06 Jan-07 Apr-07 Jul-07 Oct-07

6.4

6.6

6.8

7.0

7.2

7.4

7.6

7.8

8.0

8.2

8.4

8.6

8.8

9.0

1Y/2Y slope (LHS)

2Y swap rate (RHS)


We see this in the chart above, where since the start of 2006 the direction of the 2Y swap has had a big impact on the direction of the 1Y/2Y slope. We think this directionality will continue in 2008. Thus we would express a bullish view on NZ rates via a 1Y/2Y flattener – at least so long as we think that any RBNZ easing is a long way off. This is a trade we currently have on our book.

We think rate spreads with the US will widen in the early part of 2008 and then narrow by year-end The 10Y NZD/USD swap differential remains closely tied to the evolution of the expected short-end spread. As we write the market is looking for the expected short-end spread to widen by almost 100bp in 2008 from its current level. This is entirely due to the amount of Fed easing priced into the US front-end. So the issue for the 10Y swap differential is whether the market will price additional widening in the short-end gap than is already priced.

We think there is a high likelihood of this in the next few months. That is, we think the US front-end will price more than another 100bp of easing into 2008 at some point in the first few months of the year (though the Fed may not ultimately deliver this), while the NZ front-end will be relatively stable. For this reason we are forecasting additional modest widening in the 10Y NZD/USD swap differential in the first three months of 2008. From the middle of the year we are looking for the US market to start pricing a turnaround in the economy and thus Fed policy.

[email protected] (612) 8258 1475



Dollar Bloc Relative Value

Looking to 2008: A return to normalcy in credit should restore some order

The AUD implied vol surface has risen over the year –particularly at the short end – as the credit crunch has injected vol at shorter tenors. Once the credit issues fade, we think the yield curve slope will return as the key driver.

ACGB linker breakevens – particularly the 2020 – are at extremely wide levels. As more supply comes to the market from Semi-government issuers we look for breakevens to contract.

The AUD/USD basis swap remains wide. When issuers regain confidence we think it will begin to narrow. We would see a material widening in the basis as an opportunity to enter a received position.

AUD vol back to 2004 levels The long decline of implied volatility in Australia since 2003 has reversed course during 2007. Even with the continued flattening of the yield curve through the year, average normalised implied volatility through Q4 2007 to date has risen to be on par with its levels in Q4 2004.

Quarterly average of AUD normalised implied vol

55

65

75

85

95

105

115

125

135

Dec-99 Feb-01 A pr-02 Jun-03 A ug-04 Oct-05 Dec-06

1m 2m

3m 6m

12m


We have observed in the past that the AUD vol surface tends to steepen as the curve flattens. However, through the course of 2007 this link has been tested, with the vol surface flattening despite the increasingly inverted curve. Long tenor and long maturity volatility has risen negligibly over the year, with 1Y*10Y normalized implied vol climbing just 9 basis points since January. During the same period, vols 3M and shorter in tenor have increased by nearly 40bp.

Looking forward to 2008, we think the credit crunch will continue to have an impact on vol markets. In early October, when it looked as if credit might be returning to

some semblance of normality, the level of vol dropped and the surface steepened. However, short maturity vols were quick to rise again as swap spreads again began to widen through late October and the “second phase” of the credit crunch began.

If credit markets do return to “normal” for an extended period of time, we think this will see vol falling again to its pre-August 2007 levels. We do not expect credit pressures to persist too far beyond Q1 next year.

Changes in normalised implied vol since 2-Jan-07

1Y2Y

3Y5Y

7Y10Y

1M2M

3M6M

12M

0

5

10

15

20

25

30

35

40

S wa p T e no r

O pt io n M a t urit y


The other possibility (albeit seemingly a small one) for a significant shift in vol levels would be a change to the market pricing RBA easings. This could steepen the yield curve significantly and would point towards levels of vol more like what was seen in the last easing cycle through 2001.

Until this happens, it seems that vol movement in the short term is that long-tenor vols will rise to fairer levels as selling pressure we have been seeing in that region of the curve dissipates. Buying long-tenor and selling short-tenor vol is an attractive strategy in the short term. In the medium term, however, we think it is likely that all vols will fall back as the credit crunch fades.

2020 breakeven inflation at record wides The inflation-linked bond market has had a big year in 2007. Breakevens collapsed at the end of 2006 as actual inflation fell, but the 2015 and 2020 breakevens began a steady upward trend in 2007, largely ignoring soft CPI prints in January and April. The two longer-dated breakevens peaked in mid-October, with the 2020 reaching 3.60%.

The 2010 breakeven broke the trend early in the year, splitting away from the 2015 and 2020 breakevens. With less than three years to go to maturity, it was expected



that this bond would move to track real inflation more closely over the remainder of it life. As CPI continued to print softly through the early of part of the year, 2010 breakevens did indeed fall, but the higher CPI prints through the latter part of 2007 drove a sustained widening that lasted for six months – from May to November.

The ACGB linker market remains highly vulnerable to significant flows. The strong demand for inflation linked debt through 2007 has been the primary force responsible for keeping real yields bid low and pushing breakevens wider, particularly at long maturities.

ACGB breakeven inflation

1.80

2.00

2.20

2.40

2.60

2.80

3.00

3.20

3.40

3.60

3.80

Jan-06 A pr-06 Jul-06 Oct-06 Jan-07 A pr-07 Jul-07 Oct-07

A ug-10 B reakevenA ug-15 B reakevenA ug-20 B reakevenRB A Range


Some relief has come to the market of late in the form of Semi-government issuance, and there is currently $1b of TCorp and QTC linkers to mature between 2025 and 2035. This helped tighten breakevens for nearly a month before end-of-November flows (which we attribute to portfolio managers restructuring to match benchmark indices)

Through 2008 we are looking for a tightening in breakevens. Bootstrapped inflation from the breakeven curve for the period 2015 to 2020 is now 4.15% - an amount which is clearly extreme given the RBA’s inflation target of 2-3%. The broader trend, however, exclusive of moves driven by heavy flows will be for breakevens to remain correlated with bond yields.

Basis swaps: Waiting for a credit market turn The AUD/USD basis spread pushed tighter in the early part of 2007 on the back of a small flurry of Kangaroo issuance. Ultimately, however, issuance was not significant enough through the first half of the year to produce further large moves with spreads gradually drifting tighter.

The credit crisis impacted the basis in August, pushing the 2Y basis nearly 3bp wider over two days. The 5Y and 10Y basis were also affected. However, early in September it

began to look as if the credit crisis might be abating, and a few tentative issuers stood up with some Kangaroos. The 2Y and 5Y basis recovered to near their July levels, but the 2Y remained significantly higher.

AUD/USD basis swap curve inverts during 2007

4.0

5.0

6.0

7.0

8.0

9.0

10.0


2 Year 5 Year 10 Year


The second phase of the credit crunch in November saw the basis again lurch wider, but the recovery has not happened yet. We are hearing word of issuance gradually beginning to pick up again in offshore markets, but there has not been much activity in Kangaroos yet (the typical driver of a basis tightening). When this picks up we think the basis will return to its “normal” flows.

2Y/10Y basis spread slope

-3.5

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

Dec-06 Feb-07 A pr-07 Jun-07 A ug-07 Oct-07

2Y/10Y basis spread slo pe


Given that 2008 is likely to start with some uncertainty about the extent to which the AUD market is “open” for Kangaroos, we expect volatility in the basis swap market to persist in the short term. Our bias is to use any material widening that might occur to enter into a received position.

[email protected] (612) 8258 1361



Global Linkers Update

TIPS Outlook 2008

The dynamics of intermediate and long breakevens this year was particularly interesting. B/Es were virtually unchanged although the level of real yields declined significantly. We believe the market is pretty efficient to reflect diverging views on growth and inflation. Whether 2008 will be the year of the demand shock or the year of the supply shock will determine the dynamics of real yields and of B/E.

Front breakevens have continued to behave in line with the expected path of CPI MoM, i.e. with the projected path of carry. We expect seasonal effects and energy prices to continue playing a major role, although the materialization of a recession could overlay a significant downward trend on breakeven levels in the 2Y to 5Y sector.

We believe that the correction experienced by front TIPS B/E over the past few weeks has fully captured the expected slowdown in CPI-U MoM between Nov and Dec. We expect the pace of CPI MoM to accelerate from January onwards and would expect some outperformance particularly given quite attractive forward breakeven levels. We are mildly bullish on front B/E for the first quarter of 2008.

Although the depreciation of the US dollar in 2007, combined with elevated commodity prices and accommodative monetary policy could create conditions for high headline inflation in 2008, the downside risk to growth and the softness of core inflation could annihilate these bullish factors for B/E.

We believe that the uncertainty regarding the future direction of intermediate and long TIPS B/E is extreme and we favor long inflation risk premium trades, i.e. 10Y-30Y B/E steepeners. Still, our relative optimism on growth for 2008 and our view on flows would make us mildly bullish strategically on 30Y B/E.

The wide level of Libor-GC spread offers attractive opportunities to buy TIPS ASW spreads. Against UST ASW, the trade offers an exposure on tighter nominal swap spreads with mildly positive carry. We also identify a cross country opportunity following the strong divergence of EUR vs USD

forward real yields. We would recommend selling the USD vs EUR 10Y15Y forward real yield.

Lessons from the TIPS market in 2007

TIPS breakevens in the 10Y to 30Y sector experienced only limited volatility this year as they traded in a pretty tight range (see chart below).

Long and intermediate B/E unchanged this year…

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

Jan-07 M ar-07 M ay-07 Jul-07 Sep-07 Nov-07 Jan-081.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

TIPS 10Y BETIPS 30Y BETIPS 2Y BE (rhs)


This dynamics was at odd with the dynamics of nominal yields and real yields. While the current level of 10Y breakeven is virtually the same as it was at the beginning of the year, the 10Y real yield is lower by 60bp (see chart below). In other words, the beta between real yields and nominal yields has been 1 this year. The story is the same in the 20Y and 30Y sector. The typical directionality of breakevens did not apply, despite a very sharp rally in yield and recurring flight to quality periods following the Summer of Subprime.



… despite an impressive richening of real yields

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

Jul-07 Aug-07 Sep-07 Oct-07 Nov-07 Dec-071.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

TIPS 10Y BE (lhs)

TIPS 10Y real yield (rhs)


We believe that the recent dynamics of the real yield curve and of the breakeven curve has been consistent with the macroeconomic context. The strong real yield rally has been a reflection in our view of the downward revisions of market expectations for real GDP growth (on the back of the housing market meltdown and of the financial crisis). On the other hand the resilience of long breakevens in the face of the strong rally suggests that the market has remained worried about the long term inflation outlook. The combination of loss of credibility of the Fed, of dollar depreciation, and of a continued rally in commodity prices, has dominated market fears about inflation. The market has been pricing a supply shock scenario where inflation can coexist with subdued growth. Whether breakevens will remain well bid in 2008 or whether they will decline and catch up with real yields will mainly depend on the magnitude of the economic slowdown in our view. We will develop in detail the outlook for intermediate/long B/E in a subsequent section.

Front breakevens on the other hand have continued to behave in line with the expected path of CPI MoM, i.e. with the projected path of carry. However we have started seeing a disconnection between the underlying trend of 2Y to 5Y breakevens and the trend of headline CPI YoY in the second half of the year as growth expectations started declining.

Divergence of front B/E vs. headline CPI YoY

1.6

1.8

2

2.2

2.4

2.6

2.8

M ar-06 Jul-06 Nov-06 M ar-07 Jul-07 Nov-07-1%

0%

1%

2%

3%

4%

5%

TIPS 5Y BE (lhs)

YoY Headline Inflation (rhs)

Disconnection B/E vs YoY CPI

Downward trend on front B/E


A 2Y breakeven has to reflect realistic expectations about the future average rate of headline inflation in the next two years and although short term inflationary shocks related to sharp increases in food or energy prices lead to temporary/seasonal increases in front B/E (on the back of the improvement of carry), front breakevens also have to trend downward somewhat to reflect higher probabilities of a disinflationary recession scenario taking place in the next 2 years. Historically, in the 90’s, headline inflation YoY has reached levels below 1.5% in weak economic environments.

In 2008 we therefore expect seasonal effects and energy prices to continue playing a major role in the monthly dynamics of front breakeven, although the materialization of a recession could overlay a significant downward trend on breakeven levels in the 2Y to 5Y sector.

Outlook on front TIPS breakevens

Practically, we believe that the market has now fully priced in the likely upcoming slowdown between November CPI MoM and December CPI MoM. Given that November CPI is likely to come out particularly strong at 210 i.e. +0.50% MoM (on the back of a 10% increase in average gasoline prices between October and November) and given the typical negative seasonals associated with December and the correction in gasoline prices so far since the start of the month, the slowdown is almost a certainty and is going to lead to a deterioration of carry.

Comparing the historical dynamics of CPI-U NSA MoM and of the TIPS 09 breakevens, it seems that the strong tightening experienced since the end of November has brought the spot breakeven close to fair value. Given the presence of positive carry in the next 3 months, and our view that CPI MoM is likely to accelerate again from its



December level in January, we can observe a disconnection between the forward levels of the TIPS 09 breakeven and the expected future levels of spot B/E implied by the projected path of CPI MoM in Q1 2008 (see chart below).

Cheapness of front B/E in forward space

0.5

1.0

1.5

2.0

2.5

3.0

3.5

M ar-06 Jul-06 Nov-06 M ar-07 Jul-07 Nov-07 M ar-08-1.5%

-1.0%

-0.5%

0.0%

0.5%

1.0%

1.5%

BE CPI 2009M oM US CPI-U NSA (1M lag)

1M , 2M and 3M fwd BE

Discon-nection


This tends to show in our view some cheapness in forward breakevens, a potential opportunity to become long TIPS breakevens in the front-end at good levels in anticipation of the start of the typical period of positive seasonals in Q1 08. Moreover the recent increase in import prices is likely to create upward pressures on headline inflation in the next 2 to 3 months (see section below for more details on the link between dollar, import prices and headline inflation) and should therefore favor the typical acceleration of MoM inflation early in 2008.

The risks associated with the implementation of long positions in the front-end are threefold. First the realization of the positive seasonals and of the acceleration of inflation will depend on the short term dynamics of energy prices. As long as gasoline prices stabilize, the carry improvement should take place in Q1, but any collapse could jeopardize the expected rebound in front breakeven. The second risk is the materialization of a recession. Seasonal effects make breakevens move around their trend and should Q4 07 real GDP growth come out slower than expected and the outlook for Q1 deteriorate further, front breakevens could reprice the expected inflation for 2008 on the downside. Finally, although CPI MoM is expected to rebound in Jan vs. Dec, 2M carry is going to continue to decline until the end of January, which could delay the expected rebound of front breakevens.

All in all, taking into account the recent Fed cut and the new joint venture of central banks, united to solve the liquidity crisis, we feel that market sentiment could improve and support front breakevens. We recommend implementing a widening bias on front B/E for Q1 08.

Finally, we find that relative value opportunities involving the Jan-11 breakeven are still present and we would recommend monitoring them closely as liquidity could come back in the market in the upcoming days or weeks. The TIPS Jan-11 breakeven is extremely cheap relative to Jan-10 and Jan-12 B/E as shown by B/E butterflies. This cheapness is such that we find strong distortions along the term structure of forward breakevens. While 1Y1Y BE (implied by TIPS 09 and Jan-10) is close to 2.12%, the cheapness of the Jan-11 B/E has brought the 2Y1Y BE (implied by Jan 10 and Jan 11) to extreme lows around 1.9% and the 3Y1Y BE (implied by Jan 11 and Jan 12) close to extreme highs around 2.65% (see chart below).

3Y1Y BE disconnected vs. 2Y1Y BE

1.70%

1.90%

2.10%

2.30%

2.50%

2.70%

2.90%

M ar-06 Jul-06 Nov-06 M ar-07 Jul-07 Nov-07

2Y1Y BE (from TIPS Jan-10 and Jan-11)3Y1Y BE (from TIPS Jan-11 and Jan-12)


A 75bp spread between 3Y1Y BE and 4Y1Y BE does not make any sense and we would expect TIPS Jan-11 BE to richen going forward. We recommend buying TIPS Jan 11 breakeven against Jan 10 and Jan 12 BE, either as a butterfly or by buying the 2Y1Y BE vs 3Y1Y BE.

Outlook on 10Y to 30Y TIPS breakevens

We have argued in previous publications that long TIPS breakevens were somewhat caught between the weakness of the US economy and the presence of inflationary pressures generated by the recent depreciation of the US dollar in a context of higher commodity prices and the loss of credibility of the Fed as an inflation fighter. We review below the factors that could lead to persistent inflationary pressures in 2008, typically positive for long breakevens. We will then review the arguments for softer inflation and tighter breakevens.

Bullish factors for inflation and long B/E in 2008

- First, we believe the recent depreciation of the US dollar could put upward pressure on headline inflation in 2008 and potentially 2009 through the channel of higher import prices. Elasticity studies from the OECD show that



a 10% depreciation of the US dollar TWI typically leads to a 1% positive inflation shock one year later. We verify the relationship between the USD appreciation/ depreciation and subsequent inflation in the charts below. We first observe the lagged impact of the dollar depreciation on import prices with a 15-month lag (see chart below).

Weaker dollar leads to higher import prices

-15%

-10%

-5%

0%

5%

10%

15%

20%

Oct-95 Nov-97 Dec-99 Dec-01 Jan-04 Feb-06 Feb-08

-25%

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

US Import prices YoY all commodityUS dollar YoY depreciation (15M lag, rhs, inverted)


We also witness the lagged impact of the dollar depreciation on headline inflation itself, with a longer lag around 18 months. This relationship takes place via the strong link between headline inflation YoY and import prices YoY with a 3 months lag (see chart below). Based on these historical relationships, the depreciation of the US Dollar TWI, of about 10% since last year, could have a lasting impact on inflation at least until the end of 2008. The latest reading of import prices YoY for November came out at +11%, an all time high level, suggesting that we are going to continue to see continuous upward pressure on headline CPI in the next couple of months.

High import prices typically lead to high headline CPI

0%

1%

2%

3%

4%

5%

6%

Oct-95 Nov-97 Dec-99 Dec-01 Jan-04 Feb-06 Feb-08-15%

-10%

-5%

0%

5%

10%

15%

20%US CPI-U Headline YoY (lhs)US Import prices YoY all commodity (3M lag - rhs)


Given the lagged effect of the dollar on inflation, the evolution of the dollar in 2008 is likely to have an impact on inflation only over the course of 2009 and 2010, but

could play a role on long term inflation expectations. We believe the dollar could remain relatively weak in the medium run. With a current account deficit of 6% of GDP, the US is willing and likely to maintain its interest in a weak dollar for years to come, via a revaluation of China’s Renminbi in particular. Middle East countries have also started questioning the peg of their currencies to the dollar. These oil-exporting states and other emerging markets, which invested a large share of their foreign exchange reserves in US treasuries, have begun to refocus on international capital markets for diversification and return optimization reasons. Thus reserves are being transferred into sovereign wealth funds, and are increasingly being held in euros. The dollar could therefore continue to show signs of weakness in the medium run.

However, in the longer run, the improvement of the US economic situation, the end of rate cuts, and a reduction of the deficits should stabilize the dollar. Ultimately the positive demographic outlook in the US, relative to Japan and Europe, the flexibility of the US economic and financial system vs. Europe, could ensure that the US dollar remains a leading currency.

- Second, elevated energy commodity prices have been one of the key factors of upside risk to inflation cited by the Fed and other observers. In the short term the rise in energy prices typically boosts headline CPI on a contemporaneous basis since energy CPI is a very significant component of the overall CPI-U index. The recent increase in gasoline prices is going to push headline Nov CPI YoY above 4% and CPI-U YoY should remain above 3.5% until the end of Q1 08.

In 2007, the near doubling of oil prices from about $50 a barrel to levels close to $100 has been the result of strong global growth, especially in emerging markets, and of various constraints on supply, in a very tense geopolitical context. In our main scenario where the US economy experiences a slowdown but does not enter into a deep recession, we would expect growth in emerging markets to remain sustained and demand for energy commodities to be relatively strong. Moreover should the Fed continue to cut rates on the back of the deteriorating financial conditions and of the downside risk to growth, the dollar could continue to depreciate further, increasing the cost of imported energy commodities in dollar terms.

- Third, food and other non-energy commodity prices could also continue to represent a threat to price stability. Their increase was driven by the global growth and demand and by new policies encouraging the development of biofuels. Analysts expect falling inventories, growing shortages in Asia and rising US



ethanol production to continue to push corn and wheat prices higher over the coming year. The charts below show how higher food commodity prices translate contemporaneously into higher food PPI YoY and how higher food PPI leads to higher food CPI (15% of the global CPI) with a 3-month lag.

Continuous increase in food commodity prices should

put upward pressure on Food CPI

-30%

-20%

-10%

0%

10%

20%

30%

40%

93 94 95 96 97 98 99 00 01 02 03 04 05 06 07-6%-4%

-2%0%

2%4%

6%8%

10%12%CRB Food Index YoY (lhs)

PPI processed food YoY

0%

1%

2%

3%

4%

5%

6%

93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08-4%

-2%

0%

2%

4%

6%

8%

10%

12%Food & Beverages CPI (lhs)PPI processed food YoY 3m lagS i 3


We have also started observing a substantial increase in Chinese import prices in the US. More precisely China has increasingly been exporting inflation as higher food prices have been causing increased wage pressures and therefore have been leading to higher costs for exported goods (see chart below).

- Finally the Fed has been facing a complex situation of simultaneous downside risk to growth and upside risk to inflation from food and energy components. The deterioration of the credit and liquidity conditions following the subprime crisis played the role of referee and pushed the Fed to start cutting its target rate by 50bp in September. At that time the Fed clearly chose to support growth and financial markets at the expense of inflation control (unlike the ECB). When the Fed initiated its rate cut process, it itself acknowledged the presence of inflationary pressures, in marked contrast to previous rate cut cycles, when the FOMC described inflation as being constrained. To justify the rate cuts despite the presence of inflationary pressures, the Fed relied on soft measures of core inflation. It simply argued that recent inflationary pressures from commodity prices had not reached core prices and that therefore, they should not

persist over time. As long as financial stress remains, the housing market puts the economy at risk, and core inflation remains soft, the Fed is likely to continue to cut interest rates, and at the margin, to favor the development of inflationary pressures in the longer run.

Increase in Chinese import prices in the US

-2.0%

-1.5%

-1.0%

-0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

Nov-04 M ay-05 Nov-05 M ay-06 Nov-06 M ay-07 Nov-07

US Import prices (all commodities)from China YoY


- Moreover the concerted liquidity injections (announced on Wednesday) from the Fed, the ECB, the BOE, the BoC, and the SNB, to alleviate the pressure in short-term funding markets is somewhat inflationary in our view. The improvement of borrowing/refinancing conditions for banks, the likely return of liquidity, and the easing of credit rationing should be positive for the US economy, and should therefore create a more favorable environment for inflationary pressures to persist.

Bearish factors for inflation and long B/E in 2008

- Recession risk represents the key downside risk to inflation and breakevens in 2008. Although TIPS breakevens are typically directly connected to MoM and YoY headline inflation on a monthly basis (at least mechanically for carry reasons) larger trends for levels of breakevens are rather determined by the evolution of US real GDP growth (see chart below). Moderation in real growth is a typical factor of future inflation moderation through the canal of the negative demand shock. In other words, low growth environments are typically associated with lower demand for core products, lower wages, higher unemployment, creating disinflationary pressures. Although our main scenario corresponds to a moderate slowdown of real GDP growth (in a range around 1.5% and 2%) in 2008, the downside risk is significant given the potential negative wealth effects of the housing market meltdown on US consumers.



Real GDP growth, key factor of breakeven trends

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Aug-99 Feb-01 Aug-02 Feb-04 Aug-05 Feb-07-1%

0%

1%

2%

3%

4%

5%

6%

TIPS 5Y BE

TIPS 10Y BE

Real GDP YoY (rhs)


Although a specific supply shock on a certain type of commodities could lead to short term upward pressures on headline inflation, we believe that a recession would prevent these external cost and price developments to lead to a sustained rise in inflation. Inflation can only rise on a sustained basis if these higher input prices lead to consumer price increases that in turn trigger further rounds of cost increases. Firms can only raise prices if aggregate demand is strong enough. In a recession, higher input prices or wage costs would reduce profit margins for companies. They could then cut production or lay-off workers. Simultaneously, as income growth is slowing, higher food or energy prices could depress consumer spending on all other goods and services. This weakness in demand would be disinflationary.

- Core inflation should remain soft in the months to come in the context of moderating GDP growth even if headline inflation remains relatively high. One of the key components of core CPI, the owner’s equivalent rent (OER), which corresponds to about 30% of the core inflation index, should remain soft at least in the first half of 2008 according to our analysis and should maintain core inflation measures under control. The chart below shows indeed that trends in OER tend to drive trends in Core CPI. The OER measures how much an owner would receive for renting out his home. Interestingly though a more direct measure of rents is imbedded in the CPI, namely the “Rent of Primary Residence” but represents only 6% of the CPI-U. We find that OER growth has typically been slower than rent growth historically, maintaining core CPI at soft levels artificially. A key reason is that the OER measure corresponds to pure-shelter-service price and hence must exclude utilities. Given that rents in the US include utility costs most of the time these rents of utilities-included-units must be adjusted downward to remove the utilities-component (for a more detailed analysis of the OER vs Rent distortions, please see our FIW 2-Nov-07).

OER YoY tends to drive core CPI YoY

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

Sep-97 Sep-99 Sep-01 Sep-03 Sep-05 Sep-071.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

5.0%

Core inflation YoY

OER YoY growth (rhs)


As a consequence an increase in the cost of utilities (i.e. an increase in energy prices) typically leads OER inflation to slower levels than rent inflation. In order to develop an outlook for OER, we give a view on the future trend of Rents, and then on the spread between OER and Rents (i.e. on the change in utilities CPI). We find a very long and significant relationship between Rent CPI YoY and the 12M lagged headline CPI YoY since the mid 70s. The interpretation is as follows: landlords typically observe, ex-post, the level of inflation for the past year, and becoming aware of the increase in their cost of living and in utilities cost, they adjust their rent higher for the following year. The chart below shows this relationship recently.

Rent inflation react with a 12M lag to CPI YoY

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

5.0%

5.5%

6.0%

99 00 01 02 03 04 05 06 07 08 091.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

5.0%US CPI-URent o f Primary Residence YoY (rhs)

US CPI-U NSA YoY with 12M lag -smoothed (lhs)


Given last year’s decline in headline inflation due to the sudden crash in energy prices, we expect rent CPI to remain soft in the months ahead, with some potential upward pressure in Q4 in 2008 given the recent surge in headline inflation YoY. Finally given the strong link between Utilities CPI (cost from natural gas, fuel, etc…) and Energy CPI and the recent increase in Energy CPI YoY (likely to persist in the next few months because of a base



effect), we are likely to observe in the next few months OER CPI YoY growing at slower levels than Rent CPI YoY.

We would therefore argue that core inflation is likely to remain soft for most of 2008, which could limit the performance of TIPS 5Y5Y B/E, very correlated historically to Core CPI MoM (see chart) and thus of the 10Y BE.

Soft Core CPI should keep 5Y5Y BE under control

2.1%

2.2%

2.3%

2.4%

2.5%

2.6%

2.7%

2.8%

2.9%

Jun-04 Dec-04 Jun-05 Dec-05 Jun-06 Dec-06 Jun-070.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

0.35%

0.40%5Y5Y TIPS B/ECore CPI M oM


Simple macro model and conclusions

We have just shown that we could argue either for a positive or for a negative environment for TIPS over the coming year. On the one hand the unique context of strong dollar depreciation, of higher commodity prices, and of accommodative monetary policy, could create a durably favourable environment for long TIPS breakevens on the back of high headline inflation. On the other hand, the materialization of a US recession could bring down inflation to much lower levels, not only because of the negative demand shock but also because a US recession would likely bring down global growth and global demand for commodities. With no clear view on the US growth we could argue that inflation going forward could surprise significantly on the upside and on the downside. The distribution of TIPS breakevens could therefore be characterized by fat tails i.e. by high kurtosis. In this context we would favour trades that would benefit from high inflation uncertainty in the months to come, i.e. inflation risk premium trades.

We have recommended two weeks ago implementing a TIPS 10Y-30Y breakeven steepener trade to capture the inflation risk premium. Although the trade has started performing, we still expect risk premia to remain high for most of 2008 and we believe the 10Y-30Y BE spread has more room to widen. Given that the 10Y B/E is more sensitive to carry and seasonality than the 30Y B/E, there is an element of seasonality in the trade. It is possible to implement a beta weighted 10Y-30Y B/E

widener to remove the directionality of the spread with respect to the 10Y breakeven.

More room on 10Y-30Y B/E steepener

0

5

10

15

20

25

30

35

40

45

50

Jan-07 M ar-07 M ay-07 Jul-07 Sep-07 Nov-0730

40

50

60

70

80

90

100

110

120

130TIPS 10Y-30Y BE slope (lhs)

DGX implied vo latility (rhs)

Trade rec


In terms of outright view on long TIPS breakevens, the decision depends mainly on the perception of the most likely economic scenario and on current valuations. We therefore run a simple macro model on 10Y and 30Y breakevens involving 3 simple key variables: real GDP growth YoY, USD TWI YoY (14M lag), and Crude oil MoM (see chart below for the link between GDP, USD, and BE).

Lagged dollar depreciation and real GPD explain B/E

1.20

1.70

2.20

2.70

3.20

3.70 -30%

-20%

-10%

0%

10%

20%

30%

TIPS 30Y BE

USD Dollar TWI YoY inverted (14-months lag)

1.20

1.70

2.20

2.70

3.20

3.70

Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07-2%-1%0%1%2%3%4%5%6%TIPS 30Y BE

Real GDP Growth YoY


We run the model for 3 distinct economic scenarios. The first one is our base scenario, of moderate slowdown of real GDP growth in 2008 (the eventual weakness in the US consumer being somewhat compensated by stronger net export, boosted by the dollar). We assume an average real GDP growth rate around 2%. Given the lagged impact of the dollar depreciation on inflation and on long breakevens (about 14



months on breakevens), the dollar outlook in 2008 does not matter. In other words, long breakevens should benefit in 2008 from the 10% depreciation of the US dollar experienced in 2007, irrespective of the GDP growth scenario in 2008. As for crude oil as a proxy for commodity prices, we assume that a moderate growth scenario could lead to a moderate price increase by 10%.

Our second scenario is our most pessimistic one and the model output should give us what we believe are floor levels for TIPS breakevens. In this scenario, we assume that the US economy experiences a strong recession leading to 0% real GDP growth in 2008. The US recession leads to a global slowdown in growth in emerging market, and to lower demand for commodities, hence an assumed decline by 30% of crude oil prices. As for the dollar, the lagged impact of the dollar depreciation in 2007 would be the only positive factor for breakevens.

Finally, our third scenario is our most optimistic one and corresponds to surprisingly strong real GDP growth, around 3.5%, leading to a new surge of commodity prices by +30%. Again the recent dollar depreciation by 10% would have a positive impact on B/E over 2008.

The tables below summarize the results:

Macro model output

R eal GD P

C rude Oil

Lagged USD

T IP S 10Y B E

T IP S 30Y B E

M ain scenario (mild slo wdo wn)

+2.0% +10% -10% 2.18% 2.57%

Stro ng US recessio n

+0.0% -30% -10% 1.78% 2.28%

R esilient eco no my

+3.5% +30% -10% 2.48% 2.80%

10Y 30Y Slo pe

C urrent levels 2.31% 2.57% 0.26%

M ain scenario (mild slo wdo wn)

-0.13% +0.00% +0.13%

Stro ng US recessio n -0.53% -0.29% +0.24%

R esilient eco no my +0.17% +0.23% +0.06%Source: Deutsche Bank

The model shows that the current level of 30Y B/E (TIPS 32 B/E) is currently fully consistent with our base/main scenario while the 10Y B/E is currently pricing in a slightly more optimistic scenario. The downside risk on the 10Y B/E appears more important than that on the 30Y B/E in case of a recession while the upside risk is higher on the 30Y B/E than on the 10Y B/E in case of a resilient economy. The model therefore confirms our view that the 10Y-30Y BE spread should widen irrespective of the

economic scenario and that the performance of the 10Y sector could be particularly limited in 2008, as suggested by the 5Y5Y B/E chart shown previously.

Moreover 30Y breakevens in particular could benefit at any point from structural demand emerging from US pension funds or insurance companies willing to hedge their long-dated inflation-linked liabilities. Recently we have started seeing strong real money flows in the long-end of the TIPS curve as investors have been focusing on alternative and credit-risk-free investments. Fundamentally we still believe that for a long term real money investor, TIPS remain an attractive asset class compared to nominal Treasuries. The probability of outperformance of TIPS vs. nominal UST in the 10Y or 30Y sector is significant in the long run, in our view. We indeed believe that the probability of US headline inflation running faster than 2.30% YoY on average for the next 10 years is quite high. Ideally in 08, buyers of long B/E should simply hedge their position with a recession trade (i.e. OTM bull steepeners).

Given our view that the inflation risk premium should increase, that our main economic scenario is one of moderate economic slowdown only, and that flows could support longer breakevens, we maintain a mild strategic widening bias on the 30Y B/E outright. We mainly recommend holding 30Y vs.10Y B/E.

Opportunities on TIPS ASW

The extreme wideness of the LIBOR-GC spreads leads to very attractive opportunities on trades involving TIPS asset swap wideners. We present them below.

Buy TIPS ASW vs. UST ASW (convergence trade)

Historically, TIPS asset swap spreads have traded at much cheaper levels than nominal US Treasuries ASW. The reason for this phenomenon has been structural. On the demand side US pension funds and insurers for instance are end-users of inflation and are significant buyers of inflation swaps. Supply of inflation, on the other hand, only comes on the bond side in the form of TIPS issuance. This supply/demand mismatch has richened significantly ZC inflation swaps (ZCIS) vs. TIPS B/E. Given that nominal cash flows of TIPS are estimated using the forward CPI curve implied by the ZCIS, the richness of the CPI swap market leads mechanically to some cheapness in the TIPS ASW relative to nominal ASW.



Typical cheapness of TIPS ASW vs. UST ASW

-100

-80

-60

-40

-20

0

0y 5y 10y 15y 20y 25y

TIPS ASW net-proceedsUST ASW net proceeds

Years to maturity

bp

Structural cheapness of TIPS ASW vs UST ASW


Interestingly TIPS ASW move with a beta lower than 1 against nominal ASW, therefore in the recent widening move of UST ASW, TIPS ASW have also richened but to a lesser extent (see chart below). Consequently, the spread between UST and TIPS ASW has widened to extreme levels, particularly in the 2Y-3Y sector where the widening of swap spreads has been the most important.

The spread between UST and TIPS swap spreads currently ranges between 40bp and 50bp in the 2Y to 5Y sector. This context generates an excellent opportunity to implement a tightener of the spread between the TIPS ASW and the UST ASW as a way to gain an exposure on the tightening of nominal ASW and on the resolution of the liquidity crisis.

Recent widening of TIPS and UST ASW

-120

-100

-80

-60

-40

-20

0

M ay-06 Aug-06 Nov-06 Feb-07 M ay-07 Aug-07 Nov-07

TII 3.50% 01/11 ASW

UST 5.00% 02/11 ASWNet-proceeds


Many investors have been reluctant to implement tightener trades on nominal 2Y ASW given the magnitude of the LIBOR-GC spread, source of significant negative carry (about -14bp on the UST Jan-10 and -10bp on the UST Feb-11 over 3 months). Implementing the tightener trade indeed involves receiving Repo GC vs paying LIBOR minus the spread.

By buying the TIPS ASW (spread widener) against selling the UST ASW (spread tightener), i.e. by implementing a TIPS vs UST ASW convergence trade in the 2Y to 3Y sector, the carry is mildly positive (even assuming a GC rate realistically lower on the UST than on the TIPS, thanks to the high differential between the TIPS ASW and the UST ASW) while the global exposure is virtually equivalent to that of a UST ASW tightener trade outright (see below the strong correlation between the UST Feb-2011 ASW itself and the spread of TIPS vs UST ASW).

3Y TIPS vs UST ASW trade is a carry efficient way to

implement a nominal swap spread tightener

-110

-100

-90

-80

-70

-60

-50

-40

-30

-20

M ay-06 Aug-06 Nov-06 Feb-07 M ay-07 Aug-07 Nov-07 Feb-08

10

20

30

40

50

60

UST 5.00% 02/11 ASW (lhs)

Spread TIPS 01/2011 vs UST 02/2011 ASW (rhs)

3M forward

-10bp

+1bp


We would recommend investors willing to take a tightening view on nominal ASW, i.e. to implement a market normalization trade, but finding the negative carry of the trade particularly off-putting, to implement a TIPS vs UST ASW convergence trade instead. It is possible to hold on to the trade, earning flat to positive carry and when the market normalizes, to benefit from the retracement of the TIPS ASW vs. UST ASW spread back to its historical level around 30bp. Given the lack of liquidity in the inflation ASW market, we also suggest a one-leg trade involving buying the 10Y TIPS ASW outright.

Buy 10Y TIPS ASW (widener trade)

We believe that the 10Y TIPS (Jul-17) ASW presents an attractive opportunity to implement a spread widener trade i.e. a trade benefiting from the continuation of the liquidity crisis in the next few months, a serious possibility given the disappointment of the market following the Fed action and its neutral tone this week.

The 10Y TIPS ASW has typically been much less volatile than the 10Y nominal ASW and has traded in a very tight range historically. With the subprime crisis and the widening of all spreads and risk premia, TIPS ASW also



widened, but very moderately compared to nominal ASW. Moreover the expectation by the market that the Fed would commit to inject liquidity and to continue cutting rates led to a corrective tightening of spreads in the second half of November and in the first half of December. In this process, the 10Y TIPS ASW tightened back to levels that were relatively close to their pre-subprime crisis levels, particularly when compared to nominal ASW (see chart below, the chart shows Jan-16 ASW data where substantial history is available but the picture is the same on the Jul-17 ASW).

Forward 10Y TIPS ASW is very tight

-80

-70

-60

-50

-40

-30

-20

-10

0

M ar-06 Sep-06 M ar-07 Sep-07 M ar-08

TII 2.00% 01/16 ASW UST 4.50% 02/16 ASW bp

3M forward

Low downsiderisk when buying

TIPS ASW


Interestingly the current disconnection between Repo GC and 3M Libor creates a positive carry situation for TIPS ASW wideners. Thus, not only the 10Y TIPS ASW looks particularly tight compared to its nominal counterpart, but its 3M forward level is about 3bp tighter. The forward TIPS ASW is therefore even closer to its pre-subprime range around 20bp. With the 3M forward ASW close to 24bp, the downside risk of the 10Y TIPS ASW widener position is limited to 4bp in our view, which is very limited for a spread widener trade. The 10Y TIPS ASW widener trade may be one of the cheapest ways to buy some risk premium today. We also believe that in the long run, TIPS ASW should converge somewhat towards their wider nominal UST ASW counterparts as natural payers of inflation could appear.

Sell TIPS vs. EUR 10Y15Y real yield

TIPS real yields have rallied strongly over the past couple of months on the back of the continuous correction in the housing market leading to downward revisions by the market of the real GDP growth outlook based on the view that US consumption will slowdown in the coming months. The rally was reinforced by the flight to quality engendered by the liquidity crunch and the lasting effects of the subprime crisis. As a consequence, the Fed has started easing its

target rate and it is expected to continue as long as financial conditions remain tense and as long as it sees downside risk to growth.

Moreover the dollar has already depreciated strongly, which should support exports and might prevent the US economy from going into a recession. Further cuts could continue to weaken the dollar and to improve the competitiveness of US exports. All in all we expect the easing of monetary conditions to lead to a resolution of the liquidity crisis at some point in the months to come and to support somewhat future US growth. Interestingly, despite the steepening of the real yield curve, long TIPS forward real yields richened to low historical levels at a time when EUR forward real yields where selling-off, leading to an extreme USD vs EUR spread (chart below).

Richness of TIPS vs EUR 10Y15Y real yield

-0.60%

-0.40%

-0.20%

0.00%

0.20%

0.40%

0.60%

0.80%

Jun-04 Dec-04 Jul-05 Jan-06 Aug-06 Feb-07 Sep-07

USD vs EUR 10Y15Y real yield spread from Linkers


On the other hand, the economic situation could deteriorate relatively quickly in Europe. First of all, the dependence of European real GDP growth on exports is crucial (see chart below) and the appreciation of the Euro is likely to be a drag on growth in the next few months.



Strong Euro should affect Eurozone exports and GDP

0

1

2

3

4

5

6

M ar-95 M ar-97 M ar-99 M ar-01 M ar-03 M ar-05 M ar-07-3

0

3

6

9

12

15

18Eurozone real GDP YoY growth (lhs)

Eurozone exports growth YoY (rhs)


Moreover the appearance of worrying inflationary pressures in Europe has prevented the ECB from easing monetary conditions despite the liquidity crunch in financial markets and the stronger Euro, equivalent to implicit hikes. In this environment, we are likely to observe a significant turn in economic cycle in Europe in the months to come and ECB actions might come too late to mitigate the economic slowdown. Despite this negative economic outlook in Europe, long dated EUR forward real

yields have remained close to historically high levels due to the strong steepening of the EUR real yield curve.

As a consequence the spread between USD and EUR 10Y15Y real yields is now historically negative, implying that the long term equilibrium real GDP growth in the US will be lower than the equilibrium real GDP growth level in Euroland. Given the monetary policy and currency mismatch in the US vs Europe we would expect this spread to become positive in the months to come. We recommend selling TIPS vs EUR 10Y15Y real yield.

Trade recommendations Bullish bias on front breakevens in Q1 08

Mild strategic widening bias on 30Y TIPS B/E

Maintain 10Y-30Y breakeven steepener

Buy TIPS ASW outright or vs. short nominal ASW

Sell TIPS 10Y15Y vs. EUR 10Y15Y real yield

Jerome Saragoussi (1) 212 250 6846

Europe Linkers Outlook 2008

With energy and food prices moving sharply higher, the HICP increase this autumn looks set to be the strongest on record. As a result, favourable carry prospects have benefited short-end breakevens and rising inflation risk premia have supported longer-term B/Es; the B/E curve has flattened. Breakevens are also likely to have been supported by reduced inflation supply.

We believe this movement may have run its course. Despite further expected upward pressures from food prices, a hawkish ECB, an expected further softening in economic data and declining year-on-year inflation in early 2008 may weigh on B/Es, especially since supply is likely to pick up in Q108. We enter the new year with a negative bias.

After the upside surprise on French November CPI OATei/OATi B/E spreads have tightened further. While the move looks exaggerated from a macro perspective, the possibility of a significant Livret A rate reset on 1-Feb and the associated potential increased hedging needs prevent us from taking long OATei/OATi positions. We would advocate B/E steepeners on the OATi curve though.

With risks to the UK economy to the downside and the RPI/CPI wedge expected to narrow significantly through 2008 (slowing house price inflation, BoE rate cuts), UKTi breakevens look vulnerable. Given the structural demand overhang at the long-end any cyclical downgrade in inflation expectations may put some steepening pressure on the breakeven curve, but a recovery in corporate inflation supply may mean some B/E tightening in the long-end too.

EUR inflation outlook

While the first half of this year has been difficult for real bond markets, linker total return indices have been rallying since the middle of the year and, except for the very long-end, total performance indices are in positive territory year-to-date. Compared to nominal bonds, linkers have out-performed this year (chart below).



Linkers outperform nominals in 2007

-1.5

-1.0

-0.5

0.0

0.5

1.0

BTP

ei 0

8

BTA

Nei

10

OA

Tei 1

2

BTP

ei 1

4

OA

Tei 1

5

DB

Rei

16

BTP

ei 1

7

OA

Tei 2

0

GG

Bei

25

OA

Tei 3

2

BTP

ei 3

5

linker o ut-perfo rmance (linker to tal return -no minal to tal return), 2007 ytd


The macro backdrop Economic growth in the first half of this year surprised to the upside as the fiscal tightening turned out to be less of a harm to economic activity than feared. As a result, policy rate expectations were successively revised higher and 10y real yields increased by 60bp in seasonally adjusted terms between March and June. With above-trend growth meaning tightening labour markets and above-average capacity utilization inflation expectations and breakeven inflation rates increased in line with real yields (below).

A stagflation scenario?

2.00

2.05

2.10

2.15

2.20

2.25

2.30

Jan-07 M ar-07 M ay-07 Jul-07 Sep-07 No v-071.6

1.7

1.8

1.9

2.0

2.1

2.2

2.3

2.4

2.5

2.6

breakevenreal yield (rhs)

DB Rei-16, seaso nally adjusted


The re-pricing of credit and liquidity crisis this summer brought a turning point for linker markets as the re-allocation of portfolios towards less risky assets and the parallel downgrading of prospects for growth and central bank rates led to a rally in real yields and easing breakeven inflation rates between July and September.

Real rates and breakevens thus moved in sync through most of this year, but the situation has changed significantly this autumn. While a re-newed intensification

of credit uncertainties has meant increasing concerns about the economy and an ongoing rally in real yields, a sharp rise in energy and foodstuff commodity prices has pushed headline HICP inflation to 3% in November and breakeven rates towards new highs (chart above) and has triggered a discussion of potential ‘stagflationary’ risks.

While the risks for an extended period of low growth and high inflation are limited in our view, the macro outlook remains uncertain. Crude oil prices have very recently come off their highs and our commodity analysts expect brent to move towards USD80/b in mid-2008. Barring a rebound in oil prices consumer energy prices are unlikely to add to headline inflation in the coming months, even though gas and electricity prices may be adjusted in January 2008, while base effect will subtract form the headline year-on-year rate through most of next year.

Recent survey indicators and producer price trends strongly suggest that processed food prices will continue to put upward pressure on HICP inflation at least in the coming couple of months and we expect processed food price inflation to move towards 7% y/y early next year and headline inflation to rise to 3.1% y/y in December. In fact, the rise in HICP between August and December is expected to reach 1.9% this year which would be the strongest price increase between these two months on record (since 1995, see chart below) and thus mean more favourable carry prospects than what the typical seasonal pattern would imply.

Strong price increases at the end of this year

-0.5

0.0

0.5

1.0

1.5

2.0

1995 1997 1999 2001 2003 2005 2007

Rise in HICP xt between A ugust & December

%


Understandably, shorter-end breakevens benefited from the improved carry outlook, but longer-term spot as well as forward breakevens also moved towards highs with 5y5y forward and 10y10y forward B/Es trading around 2.50% and 2.60% respectively, i.e. around 25bp above the level reached on average in H12007. With recent price developments as such not implying a deterioration in longer-term inflation prospects, the outlook for inflation



valuations will not least depend on how underlying inflation trends react to the run-up in food and energy prices.

HICP pushes B/Es towards highs

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Jan-98 Jan-00 Jan-02 Jan-04 Jan-06

HICP xt current 10y B /E

min B /E max B /E%


Core HICP inflation has remained more contained this year than could have been expected given the macro-backdrop (it has trended sideways at 1.9% since February), but some indirect effects—i.e. price effects resulting from higher production costs—are in fact likely to be inevitable. Gas and electricity prices for example are scheduled to be raised in January 2008 as a result of the recent rise in crude oil prices, but given the moderate trend in producer prices of core consumption goods HICP core goods inflation is unlikely to accelerate strongly in the months ahead. The risks are somewhat higher for services prices as catering—restaurant, café and canteen—prices (7.5% of the CPI basket) can be expected to be pushed higher next year as a result of the current (and coming) increases in food prices (chart below).

Higher food prices on the menu

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

Jan-96 Jan-98 Jan-00 Jan-02 Jan-04 Jan-06 Jan-08-2

-1

1

2

3

4

5

6

7

catering

fo o d (8M lead; rhs)

% y/y % y/yHICP co mpo nents


A more fundamental inflation risk is the emergence of second round effects, i.e. that the recent increase in prices would spill over into expectations of permanently

high inflation, thereby affecting price and wage setting behaviour. This risk appears to be relatively high in the current context since food prices typically have a disporportionally high weight in households’ inflation perceptions, growth and labour market trends have remained robust until recently, inflation has repeatedly surprised to the upside in recent years (and thus real wage gains have been less than expected) and there seems to be some dissatisfaction among households over trends in ‘purchasing power’.

Consumers sensitive to food prices

0

5

10

15

20

25

30

35

40

45

Jan-94 Jan-97 Jan-00 Jan-03 Jan-06-2

-1

0

1

2

3

4

5

6

7

8EC co nsumer inflatio n expectatio ns

fo o d price inflatio n, 6m saar (rhs) %


While these risks certainly justify some rise in the inflation risk premium, several factors argue against a significant upshift in the longer-term inflation trend: (i) the ECB has kept a hawkish line, relative to its peers but also relative to the changes in the economic backdrop, which should keep a lid on inflation expectations; (ii) growth appears to be slowing, may be more quickly than expected, and risks are likely to be on the downside; slowing growth would make it more difficult to raise output prices and output price balances in business surveys have recently stopped rising; (iii) this year’s moderate core price and wage trends suggest that it remains very difficult to push through price and wage increases given stern international competition; (iv) the exchange rate has appreciated by around 10% since the middle of last year and by around 5% since the middle of 2007 in trade-weighted terms; according to large scale macro models a 10% rise in the exchange rate could subtract as much as 0.7% from inflation after one year; (v) low cost producers continue to gain market share in the euro area, which given the lower price level of the imported goods means ongoing inflation dampening effects from trade, especially if import quotas are lifted further.



Headline inflation to slow in 2008

0.8

1.3

1.8

2.3

2.8

3.3

Jan-01 Jan-03 Jan-05 Jan-07

HICP

co re (to tal excl energy, fo o d, alc, to b)

% y/y

fo recast

base effect fro m German VA T


On balance, we would argue that the baseline macro scenario for early next year will be a challenging one for breakevens, especially given current trading levels. In fact, a decline in headline year-on-year inflation, more signs of easing economic growth, the market starting to price in possible ECB cuts (linkers tend to under-perform in an environment of a flattening money market curve) could prove to be a negative for breakevens in Q1 and our macro model suggests that the inflation risk premium currently is at exceptionally wide levels (chart below). However, a slow uptrend in core inflation (excluding the German VAT effect) should provide some support.

Macro backdrop points to lower B/Es

1.90

1.95

2.00

2.05

2.10

2.15

2.20

2.25

2.30

2.35

2.40

Jan-04 A ug-04 M ar-05 Oct-05 M ay-06 Dec-06 Jul-07

B /E 10Yfitted

%


Supply & demand trends With growth surprising to the upside, public finance consolidation has progressed somewhat more quickly than expected during the past couple of years. Tax receipts again were above expectations this year and issuance plans for H207 were cut half way through the year. About EUR30bn of the roughly EUR43bn of 2007 sovereign linker supply were issued in the first half of the

year and reduced supply is likely to have been a factor supporting breakeven levels towards the end of the year, in particular since the tightening in credit conditions also meant only very limited non-sovereign issuance.

Sovereign linker issuance to pick up again

0

5

10

15

20

25

30

2003 2004 2005 2006 2007 2008 (e)20

25

30

35

40

45

50

55

60

FR IT DE

GR EUR (rhs)

linker issuance, EURbn


Slowing economic growth as well as pressures for increased government spending are likely to mean that the process of public finance consolidation will lose momentum next year. Headline government deficits are projected to show a marginal decline at best in the big3 economies, and a heavy redemption schedule means that total gross bond issuance is likely to increase. With breakevens still high from a historical perspective, conditions remain relatively favourable for linker issuance and the share of linkers in total bond issuance is expected to remain high in Italy and France and rise in Germany. Sovereign supply is likely to be front-loaded into the first half of the year again and with Germany expected to (tap the OBLei-13 and) issue a new 30y bond, Greece to re-open the GGBei-30 and France reported to contemplate a new 15y or 30y OATi, supply may well be tilted towards the longer-end which raises the risk of the 30y sector under-performing on the curve early next year.

Concerning non-sovereign supply the uncertainty remains high. Press reports suggest that a significant amount of infrastructure debt placements in particular have been postponed because of the difficult market conditions. A normalisation of the latter could hence mean an increase in non-government supply early next year.

Demand for inflation products has remained robust or even increased. Assets invested in specialized inflation funds denominated in EUR have declined slightly this year, but this was offset by an increase in assets invested in USD inflation funds (which are often invested globally). There are also signs that some of the holdings have been switched to specialized LDI funds which have seen their volumes grow significantly this year.



Inflation funds

-5

5

15

25

35

45

55

65

75

H104 H204 H105 H205 H106 H206 H107 11/07

assets in USD inflation funds

assets in EUR inflation funds

EURbn

Source: BBG, Deutsche Bank

In general, LDI flows continued to develop with Dutch pension funds in particular increasing their holdings of inflation-linked assets and this trend is expected to continue and intensify next year. Overall, while an increase in supply may weigh on breakevens temporarily (and may amplify downward pressure on B/Es from macro drivers early in the year), more structurally increasing demand is expected to easily absorb the additional supply.

OATei vs OATi OATei - OATi spreads have seen another volatile year in 2007. Breakeven spreads tightened significantly in the wake of this year’s presidential election on news that the new French government was contemplating a hike in value-added tax, but re-widened later during the summer as it became clear that these plans were put on ice. Spreads then stabilized at levels slightly below the average prevailing during the first half of the year, before tightening again recently, in particular at the shorter-end.

EUR/FRF breakevens spreads

-5

0

5

10

15

20

25

30

35

40


ei12-i11ei10-i09

bp B /E spread, seaso nally adjusted


While some of the tightening may be due to expectations of stronger increases in food prices in France in December, expectations of a significant increase in the

Livret A rate in February 2008 and the associated increased hedging needs are likely to also have helped. The biannual adjustment of the Livret A rate is mechanical and based on EURIBOR and CPI data registered in December, although in ‘exceptional circumstances’ the BdF may decide not to apply the usual formula. Given current data trends, the rate may well increase to 4.00% on 1-February, from currently 3%.

This would be an exceptionally high rate for this type of risk-free and flexible savings vehicle and could lead to a significant rise in Livret A deposits (currently EUR116bn outstanding) early next year and thus increase demand for French inflation around the 10y point.

Moreover, plans to increase value-added tax have not been definitively abandoned and more support for OATi could emerge if the VAT debate re-surfaces once the regional elections of March out of the way. In all, while macro arguments would clearly argue in favor of OATei at current spread levels, the prospects of a significant rise in Livret A demand as well the possibility of a re-newed VAT discussion means we prefer neutral EUR/FRF positions going into the new year.

On the OATi curve, after the upside surprise on November CPI short-end breakevens widened markedly and the B/E curve now looks flat in forward space against our CPI forecasts. In fact, 3M forward breakevens are still at three-year lows (see chart below). Some pay-back for the sharp energy price rises in November can be expected in December and, despite more upside pressures from food prices, we would prefer B/E steepening positions (especially OATi-11/13), in particular given the potential increase in Livret A hedging.

French B/E curve looks flat now

-2

0

2

4

6

8

10

12

14

Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07 Jan-08

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

OA Ti 13 - OA Ti 11

cpi x to b, fr% mo m

bp




Credit revaluation & relative value The revaluation of credit in recent months has led to two noteworthy relative value trends in inflation markets. For once, lower credit risk paper has generally fared better with the DBRei-16 out-performing. For example, the DBRei-16/OATei-15 B/E spread has tightened since the spring, despite further evidence of the HICP seasonal pattern becoming more pronounced, i.e. pointing towards a widening of the spread. In other words, in seasonally adjusted terms, the DBRei-16/OATei-15 B/E spread has widened towards extreme values.

GGBei-25 B/E cheapens vs neighbors

-10

-8

-6

-4

-2

0

2

4

M ay-06 Sep-06 Jan-07 M ay-07 Sep-07

50/50

OA Tei 20/GGB ei 25/OA Tei 32

B /E butterfly


The spread widening has however been most visible between Italian and Greek linkers on the one hand and core paper on the other. The GGBei-25 B/E has cheapened markedly vs its OATei neighbors (chart above) while the BTPei-17 B/E has reached new lows in the DBRei-16/BTPei-17/OATei-20 fly or relative to the DBRei-16 B/E (chart below). The credit component in inflation can largely be explained by the fact that breakevens compare real and nominal bonds of similar maturity, but not of similar duration and that higher duration bonds are affected more by a steepening in the credit curve.

So does the BTPei-17

1

3

5

7

9

11

13

Jun-06 Sep-06 Dec-06 M ar-07 Jun-07 Sep-07 Dec-07

B TP ei 17 - DB Rei 16B /E spread

bp


The second trend has seen a richening in zero-coupon breakevens relative to bond B/Es. Indeed, the difficult credit conditions have meant a reduction in inflation supply in derivatives format while pension fund demand for inflation derivatives has at the same time remained robust. With Bank’s left to do the cash-derivatives recycling and balance sheet costs increasing, zero-coupon swap breakevens have out-performed cash breakevens since this summer (chart below).

Swap B/Es widen relative to cash B/Es…

2

4

6

8

10

12

14

16

18

20

22

Dec-05 A pr-06 A ug-06 Dec-06 A pr-07 A ug-07

B /E ZC swap - B /E OA Tei-12 (sa)bp


Mechanically, this has put cheapening pressure on linker ASW spreads relative to nominal ASW spreads, which has been particularly pronounced for BTPeis. As long as financial market conditions remain difficult and natural derivatives supply does not pick up, this trend may well remain in place, particularly if pension fund demand increases.

…and real/nominal ASW differentials widen

5

7

9

11

13

15

17

A ug-06 No v-06 Feb-07 M ay-07 A ug-07 No v-07

B TP ei-17

OA Tei-15

A SW differential, linker - no minal

bp

linker richens relative to no minal


To recap, from a macro perspective breakevens look high and first signs of stabilization in headline inflation may put downward pressures on breakevens early next year, especially if as expected the ECB sticks to a hawkish line, growth indicators continue to slow and the euro remains



strong. Until then short-term upward inflation pressures remain dominant however and upside risks on December CPI may provide support for B/Es in particular at the short-end. A pick-up in supply may also be a negative for breakeven levels in Q108 in particular at the longer-end, especially if non-sovereign supply recovers as well, but on balance over next year the risk seems to be that new demand outstrips additional supply. Still, we would prefer entering the year with a negative breakeven bias with long forward B/Es but also short-end B/Es looking particularly vulnerable.

UKTi outlook

UK linkers had another excellent year with UKTi generating returns of close to 7% and out-performing their peers (chart below). Also, in breakeven terms, the UK is likely to have been the best performer among the major markets this year.

2007 (ytd) returns of sovereign linker indices

0

2

4

6

8

10

12

go bal US UK EUR SEK

linker to tal return 2007 (ytd)%

Source: DBIQ, Deutsche Bank

In fact, after trending higher through the first half of this year, real yields gradually moved lower since early July on the back of successive downgrades of expectations for growth and central bank rates (yields are down more than 40bp for 30y, more than 70bp for 10y) and despite a very recent rebound, currently still trade close to the cyclical lows recorded in autumn 2006.

Breakeven inflation rates have trended gradually higher since mid-2005 and while the move has lost some momentum over the past few months and long-end breakevens currently trade just below the highs reached in late June, they have nonetheless proved remarkably resilient against the backdrop of a marked rally in nominals and deteriorating growth expectations (chart below).

Breakevens high

2.80

2.90

3.00

3.10

3.20

3.30

3.40

3.50

3.60

Jun-06 Sep-06 Dec-06 M ar-07 Jun-07 Sep-07 Dec-07

UKTi 16UKTi 35

breakeve rates

%


Indeed, while rising energy and food prices have meant some upward revisions to near-term inflation forecasts, core inflation has actually surprised on the downside since this summer and the downgrade in growth prospects has led to expectations of easing pipeline price pressures.

Core inflation has eased back

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5


CP I

Co re (to tal ex energy, fo o d alc & to b)

% y/y


We believe the UK economy will slow quite quickly as a result of tightening credit standards and that the risks to consensus growth forecasts are to the downside. With financial strain on consumers increasing, households will be sensitive to rises in prices and producers and retailers are likely to be reluctant to pass-through past cost increases. In these circumstances we expect the underlying inflation trend to remain contained next year.



Expected rate cuts point to narrower RPI/CPI wedge

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

Jan-97 Jan-99 Jan-01 Jan-03 Jan-05 Jan-07-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.53M LIB OR (12M change; lhs)

RP I-CP I (y/y spread; rhs)


While CPI inflation is expected to move broadly sideways, RPI inflation is projected to slow gradually and a significant narrowing of the RPI/CPI gap, currently at 2.1pp, seems to be on the cards. First, house price inflation is slowing and the DCLG measure which the ONS uses to proxy housing depreciation costs in the RPI may well slow towards zero at the end of 2008, from over 10% currently. As a result, the contribution to the wedge from the ‘other housing’-component could also quickly wane, from currently 0.4pp. Moreover, with BoE policy rates expected 75bp lower in autumn 2008 than in autumn this year, the contribution to the RPI/CPI gap from mortgage interest payments could also fade (chart above) from currently 1.1pp.

RPI to converge towards CPI next year?

0.0

1.0

2.0

3.0

4.0

5.0

6.0


CP I

RP I

% y/y fo recast


As a result, we expect RPI and CPI inflation to converge through next year and RPI year-on-year inflation to reach 2¼% in Q408 (chart above). The expected cyclical slowdown in RPI inflation may prove to be a negative for breakevens going forward.

While the level of breakevens thus may look stretched against the macro backdrop and a BoE target of 2%, the reduction of non-sovereign inflation supply in the wake of this summer’s credit crunch is likely to have been a factor supporting linkers in the context of a structural demand overhang. Indeed, while gilt issuance in Q4 was in line with expectations, corporate supply declined sharply in the last quarter of this year (chart below).

Non-sovereign supply sharply lower in Q4

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

Q107 Q207 Q307 Q407

so vereignno n-so vereign

GB P minflatio n supply


Total gilt issuance is expected to fall by GBP8bn to GBP50bn during the next fiscal year, and although the proportion of linkers may increase somewhat from the current close to 26%, total linker issuance may end up slightly below this year’s GBP15bn.

There remains significant uncertainty about the development of corporate supply. Clearly there is potential for some catch-up with deals having been postponed during the last quarter of this year. While the effects on corporate issuance of financial difficulties at monocline insurers remain unclear and markets are far from normalizing, the timid pick up of activity towards the end of this quarter (Walsall and Essex hospital PFI, Amey Lagan Roads) is encouraging. A more meaningful recovery in corporate supply in the context of decelerating RPI inflation may mean some downside risks for breakevens even at the long-end. More generally however, the structural demand/supply imbalance is likely to continue to support breakevens in particular at the long-end and any cyclical downgrade in inflation expectations may put some steepening pressure on the breakeven curve.

Markus Heider (44) 20 7545 2167



Auction Calendar Market Ticker/Coupon/Maturity Date Tap/New Issue Size

AUSTRALIA Nothing Expected

AUSTRIA Nothing Expected

BELGIUM Nothing Expected

CANADA Nothing Expected

DENMARK Nothing Expected

FINLAND Nothing Expected

FRANCE BTF

BTF

Mon, 17 Dec 2007

Mon, 24 Dec 2007

Tap

Tap

TBA

TBA

GERMANY Nothing Expected

GREECE Nothing Expected

IRELAND Nothing Expected

ITALY ICTZ

BOTS 0% 06/08

Thu, 27 Dec 2007

Thu, 27 Dec 2007

Tap

Tap

TBA

TBA

JAPAN JGB 0% 12/08

JGB TBA% 12/27

JGB 0% 04/08

PORTB 0% 03/08

JGB TBA% 01/10

Mon, 17 Dec 2007

Tue, 18 Dec 2007

Wed, 19 Dec 2007

Wed, 19 Dec 2007

Fri, 21 Dec 2007

Tap

New Issue

Tap

Tap

Tap

JPY

JPY 800 bn

JPY

JPY

JPY

NETHERLANDS Nothing Expected

NEW ZEALAND Nothing Expected

NORWAY NGTB Mon, 17 Dec 2007 Tap TBA

PORTUGAL Nothing Expected

SOUTH AFRICA Nothing Expected

SPAIN SGLT

SGLT

Wed, 19 Dec 2007

Wed, 19 Dec 2007

Tap

Tap

TBA

TBA

SWEDEN Nothing Expected

SWITZERLAND Nothing Expected

UK Nothing Expected

US 4 Month

3 Month

6 Month

Tue, 18 Dec 2007

Mon, 17 Dec 2007

Mon, 17 Dec 2007

Tap

Tap

New Issue

TBA

TBA

TBA Source: Deutsche Bank



Contacts Name Title Telephone Email

FRANKFURT

Bernd Volk Covered Bonds/SSA 49 69 9103 1967 [email protected]

LONDON

Aditya Challa Euroland & UK RV 44 20 754 51600 [email protected]

Alessandro Cipollini Euroland & UK RV 44 20 754 74458 [email protected]

Markus Heider Euroland RV 44 20 754 52167 [email protected]

Ralf Preusser European Strategy 44 20 7545 2469 [email protected]

Soniya Sadeesh Euroland & UK RV 44 20 7547 3091 [email protected]

Gopi Suvanam Euroland & UK RV 44 20 7545 1600 [email protected]

Francis Yared Euroland & UK RV 44 20 7545 4017 [email protected]

NEW YORK

Ralph Axel US RV 1 212 250 7104 [email protected]

Lei Chen US RV 1 212 250 9830 [email protected]

Mustafa Chowdhury Head of US Rates Research 1 212 250 7540 [email protected]

Marcus Huie US Derivatives Strategy 1 212 250 8356 [email protected]

Anish Lohokare US RV 1 212 250 2147 [email protected]

Aleksandar Kocic US Quantitative Strategy 1 212 250 0376 [email protected]

Jerome Saragoussi US RV 1 212 250 6846 [email protected]

SYDNEY

David Plank $ bloc Strategy 61 2 8258 1475 [email protected]

Kenneth Crompton $ bloc RV 61 2 8258 1361 [email protected]

TOKYO

Christian Carrillo Asia Strategy 81 3 5156-6206 [email protected]

Alexander During Japan & Asia RV 81 3 5156 6199 [email protected] Source: Deutsche Bank



Appendix 1 Important Disclosures

Additional information available upon request

For disclosures pertaining to recommendations or estimates made on a security mentioned in this report, please see the most recently published company report or visit our global disclosure look-up page on our website at http://gm.db.com.

Analyst Certification

The views expressed in this report accurately reflect the personal views of the undersigned lead analyst(s). In addition, the undersigned lead analyst(s) has not and will not receive any compensation for providing a specific recommendation or view in this report. Mustafa Chowdhury, Ralf Preusser Francis Yared

Deutsche Bank debt rating key

CreditBuy (“C-B”): The total return of the Reference Credit Instrument (bond or CDS) is expected to outperform the credit spread of bonds / CDS of other issuers operating in similar sectors or rating categories over the next six months. CreditHold (“C-H”): The credit spread of the Reference Credit Instrument (bond or CDS) is expected to perform in line with the credit spread of bonds / CDS of other issuers operating in similar sectors or rating categories over the next six months. CreditSell (“C-S”): The credit spread of the Reference Credit Instrument (bond or CDS) is expected to underperform the credit spread of bonds / CDS of other issuers operating in similar sectors or rating categories over the next six months. CreditNoRec (“C-NR”): We have not assigned a recommendation to this issuer. Any references to valuation are based on an issuer’s credit rating. Reference Credit Instrument (“RCI”): The Reference Credit Instrument for each issuer is selected by the analyst as the most appropriate valuation benchmark (whether bonds or Credit Default Swaps) and is detailed in this report. Recommendations on other credit instruments of an issuer may differ from the recommendation on the Reference Credit Instrument based on an assessment of value relative to the Reference Credit Instrument which might take into account other factors such as differing covenant language, coupon steps, liquidity and maturity. The Reference Credit Instrument is subject to change, at the discretion of the analyst.



Regulatory Disclosures

SOLAR Disclosure

For select companies, Deutsche Bank equity research analysts may identify shorter-term trade opportunities that are consistent or inconsistent with Deutsche Bank's existing longer term ratings. This information is made available only to Deutsche Bank clients, who may access it through the SOLAR stock list, which can be found at http://gm.db.com

Disclosures required by United States laws and regulations

See company-specific disclosures above for any of the following disclosures required for covered companies referred to in this report: acting as a financial advisor, manager or co-manager in a pending transaction; 1% or other ownership; compensation for certain services; types of client relationships; managed/comanaged public offerings in prior periods; directorships; market making and/or specialist role.

The following are additional required disclosures:

Ownership and Material Conflicts of Interest: DBSI prohibits its analysts, persons reporting to analysts and members of their households from owning securities of any company in the analyst's area of coverage. Analyst compensation: Analysts are paid in part based on the profitability of DBSI, which includes investment banking revenues. Analyst as Officer or Director: DBSI policy prohibits its analysts, persons reporting to analysts or members of their households from serving as an officer, director, advisory board member or employee of any company in the analyst's area of coverage. Distribution of ratings: See the distribution of ratings disclosure above. Price Chart: See the price chart, with changes of ratings and price targets in prior periods, above, or, if electronic format or if with respect to multiple companies which are the subject of this report, on the DBSI website at http://gm.db.com.

Additional disclosures required under the laws and regulations of jurisdictions other than the United States

The following disclosures are those required by the jurisdiction indicated, in addition to those already made pursuant to United States laws and regulations. Analyst compensation: Analysts are paid in part based on the profitability of Deutsche Bank AG and its affiliates, which includes investment banking revenues Australia: This research, and any access to it, is intended only for "wholesale clients" within the meaning of the Australian Corporations Act. EU: A general description of how Deutsche Bank AG identifies and manages conflicts of interest in Europe is contained in our public facing policy for managing conflicts of interest in connection with investment research. Germany: See company-specific disclosures above for holdings of five percent or more of the share capital. In order to prevent or deal with conflicts of interests Deutsche Bank AG has implemented the necessary organisational procedures to comply with legal requirements and regulatory decrees. Adherence to these procedures is monitored by the Compliance-Department. Hong Kong: See http://gm.db.com for company-specific disclosures required under Hong Kong regulations in connection with this research report. Disclosure #5 includes an associate of the research analyst. Disclosure #6, satisfies the disclosure of financial interests for the purposes of paragraph 16.5(a) of the SFC's Code of Conduct (the "Code"). The 1% or more interests is calculated as of the previous month end. Disclosures #7 and #8 combined satisfy the SFC requirement under paragraph 16.5(d) of the Code to disclose an investment banking relationship. Japan: See company-specific disclosures as to any applicable disclosures required by Japanese stock exchanges, the Japanese Securities Dealers Association or the Japanese Securities Finance Company. Russia: The information, interpretation and opinions submitted herein are not in the context of, and do not constitute, any appraisal or evaluation activity requiring a licence in the Russian Federation. South Africa: Publisher: Deutsche Securities (Pty) Ltd, 3 Exchange Square, 87 Maude Street, Sandton, 2196, South Africa. Author: As referred to on the front cover. All rights reserved. When quoting, please cite Deutsche Securities Research as the source. Turkey: The information, interpretation and advice submitted herein are not in the context of an investment consultancy service. Investment consultancy services are provided by brokerage firms, portfolio management companies and banks that are not authorized to accept deposits through an investment consultancy agreement to be entered into such corporations and their clients. The interpretation and advices herein are submitted on the basis of personal opinion of the relevant interpreters



and consultants. Such opinion may not fit your financial situation and your profit/risk preferences. Accordingly, investment decisions solely based on the information herein may not result in expected outcomes. United Kingdom: Persons who would be categorized as private customers in the United Kingdom, as such term is defined in the rules of the Financial Services Authority, should read this research in conjunction with prior Deutsche Bank AG research on the companies which are the subject of this research.

GRCM2007PROD012106

David Folkerts-Landau Managing Director

Global Head of Research

Global Company Research Global Fixed Income Strategies & Economics

Global Equity Strategies & Quantitative Methods

Ross Jobber Chief Operating Officer

Guy Ashton Global Head

Marcel Cassard Global Head

Stuart Parkinson Global Head

Europe Germany Asia-Pacific Americas

Pascal Costantini Regional Head

Andreas Neubauer Regional Head

Michael Spencer Regional Head

Karen Weaver Regional Head

Principal Locations

Deutsche Bank AG London 1 Great Wincester Street London EC2N 2DB Tel: (44) 20 7545 8000

Deutsche Bank AG New York 60 Wall Street New York, NY 10005 United States of America

Deutsche Bank AG Hong Kong Cheung Kong Center, 2 Queen’s Road Central China Tel: (52) 2203 8888

Deutsche Bank AG Japan 2-11-1 Nagatacho Sanno Park Tower Chiyoda-ku, Tokyo 100-6171 Tel: (81) 3 5156 6701

Deutsche Bank AG Frankfurt Grosse Gallusstrasse 10-14 Frankfurt am Main 60311 Germany Tel: (49) 69 910 0

Deutsche Bank AG Boston 225 Franklin Street Boston, MA 02110 United States of America Tel: (1) 617 217 6100

Deutsche Bank AG Singapore One Raffles Quay South Tower Singapore 048583 Tel: (65) 6423 8001

Deutsche Bank AG Australia Deutsche Bank Place, Level 16 Corner of Hunter & Philip Streets Sydney NSW 2000 Tel: (61) 2 8258 1234

Subscribers to research via email receive their electronic publication on average 1-2 working days earlier than the printed version.

If you would like to receive this or any other product via email please contact your usual Deutsche Bank representative.

Publication Address: Deutsche Bank AG London 1 Great Winchester Street London EC2N 2EQ United Kingdom (44) 20 7545 8000

Internet: http://gmr.db.com Ask your usual contact for a username and password.

Global Disclaimer The information and opinions in this report were prepared by Deutsche Bank AG or one of its affiliates (collectively “Deutsche Bank”)for its clients. The information herein is believed by Deutsche Bank to be reliable and has been obtained from public sources believed to be reliable. With the exception of information about Deutsche Bank, Deutsche Bank makes no representation as to the accuracy or completeness of such information.

This published research report may be considered by Deutsche Bank when Deutsche Bank is deciding to buy or sell proprietary positions in the securities mentioned in this report.

For select companies, Deutsche Bank equity research analysts may identify shorter-term opportunities that are consistent or inconsistent with Deutsche Bank's existing, longer-term Buy or Sell recommendations. This information is made available on the SOLAR stock list, which can be found at http://gm.db.com.

Deutsche Bank may trade for its own account as a result of the short term trading suggestions of analysts and may also engage in securities transactions in a manner inconsistent with this research report and with respect to securities covered by this report, will sell to or buy from customers on a principal basis. Disclosures of conflicts of interest, if any, are discussed at the end of the text of this report or on the Deutsche Bank website at http://gm.db.com.

Opinions, estimates and projections in this report constitute the current judgement of the author as of the date of this report. They do not necessarily reflect the opinions of Deutsche Bank and are subject to change without notice. Deutsche Bank has no obligation to update, modify or amend this report or to otherwise notify a reader thereof in the event that any matter stated herein, or any opinion, projection, forecast or estimate set forth herein, changes or subsequently becomes inaccurate, except if research on the subject company is withdrawn. Prices and availability of financial instruments also are subject to change without notice. This report is provided for informational purposes only. It is not to be construed as an offer to buy or sell or a solicitation of an offer to buy or sell any financial instruments or to participate in any particular trading strategy in any jurisdiction or as an advertisement of any financial instruments.

The financial instruments discussed in this report may not be suitable for all investors and investors must make their own investment decisions using their own independent advisors as they believe necessary and based upon their specific financial situations and investment objectives. If a financial instrument is denominated in a currency other than an investor’s currency, a change in exchange rates may adversely affect the price or value of, or the income derived from, the financial instrument, and such investor effectively assumes currency risk. In addition, income from an investment may fluctuate and the price or value of financial instruments described in this report, either directly or indirectly, may rise or fall. Furthermore, past performance is not necessarily indicative of future results.

Derivative transactions involve numerous risks including, among others, market, counterparty default and illiquidity risk. The appropriateness or otherwise of these products for use by investors is dependent on the investors' own circumstances including their tax position, their regulatory environment and the nature of their other assets and liabilities and as such investors should take expert legal and financial advice before entering into any transaction similar to or inspired by the contents of this publication. Trading in options involves risk and is not suitable for all investors. Prior to buying or selling an option investors must review the "Characteristics and Risks of Standardized Options," at http://www.optionsclearing.com/publications/risks/riskchap1.jsp. If you are unable to access the website please contact Deutsche Bank AG at +1 (212) 250-7994, for a copy of this important document. Furthermore, past performance is not necessarily indicative of future results. Please note that multi-leg options strategies will incur multiple commissions.

Unless governing law provides otherwise, all transactions should be executed through the Deutsche Bank entity in the investor’s home jurisdiction . In the U.S. this report is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this report is approved and/or communicated by Deutsche Bank AG Frankfurt authorised by Bundesanstalt für Finanzdienstleistungsaufsicht. In the United Kingdom this report is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange and regulated by the Financial Services Authority for the conduct of investment business in the UK and authorised by Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin). This report is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. and in Singapore by Deutsche Bank AG, Singapore Branch. In Japan this report is approved and/or distributed by Deutsche Securities Inc. The information contained in this report does not constitute the provision of investment advice. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Deutsche Bank AG Johannesburg is incorporated in the Federal Republic of Germany (Branch Register Number in South Africa: 1998/003298/10) Additional information relative to securities, other financial products or issuers discussed in this report is available upon request. This report may not be reproduced, distributed or published by any person for any purpose without Deutsche Bank's prior written consent. Please cite source when quoting.

Copyright © 2007 Deutsche Bank AG 11/2007

2008 db fixed income outlook (12!14!07)

Documents

us economy

fixed income outlook

dbsis research

research reports

thirdparty research

certain companies

subject companies

uk economy