global fixed income weekly - dws...6 june 2014 global fixed income weekly deutsche bank ag/london...

Deutsche Bank Markets Research

Global

Rates Credit

Date 6 June 2014

Global Fixed Income Weekly

________________________________________________________________________________________________________________

Deutsche Bank AG/London

DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1. MCI (P) 148/04/2014.

Francis Yared

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

Dominic Konstam

Research Analyst (+1) 212 250-9753 [email protected] We think we are in a lower range for spot 10s for a while centering around

2 ½ percent. Assuming the economy bumbles along 10s should track their forwards into year end.

It’s not that the recovery isn’t good but it’s not good enough to warrant higher real rates. Central banks know that only too well. Bloated debt ratios mean they need to restore much higher nominal growth or be prepared to own a large amount of the same debt indefinitely.

There has been a lot of easing of late. It started with FX reserve managing stepping in to do “foreign QE” as the Fed has pulled away. The ECB is now on the same path. As we struggle to ramp nominal growth significantly higher this simply reinforces the view of stretching for yield and spread.

Flattening trades are troublesome for negative carry and we don’t think we are close enough to a Fed lift off to beat the forwards on the short leg. This biases us still to receiving the longer dated forwards, particularly on any dips.

Foreign investors have become a major source of demand for Treasuries during Fed taper. In the latest Federal Reserve Flow of Funds data, foreign investors bought 60% of Treasury supply last quarter, followed by the Fed and banks. Fed purchases accounted for 42% of Treasury supply, down from about 70% during QE3 in 2013. Bank holdings of Treasuries went up by $47 billion; the increase was consistent with other official data on banks’ Treasury holdings.

Not to our surprise, there was little evidence of strong demand for bonds by pension investors last quarter in the Flow of Funds report. Total pension holdings of Treasuries increased about $9 billion. The holdings of Credit Market Instruments among Defined Benefit Plans in Private Pension Funds almost held flat in Q1 2014 ($728.8 billion) compared to Q4 2013 ($729.1 billion).

Most bond funds outperformed the index in total return last week as bond yields rose. Many bond funds are still short relative to their benchmark index’s duration, according to surveys. However, most funds have outperformed the index in total return on a year-to-date basis, perhaps thanks to a credit overweight and spread tightening. Although they have tended to underperform in bull flatteners in price returns, their total return performance over a longer period should be fine as long as credit performs well

Table of contents

Bond Market Strategy Page 02

US Overview Page 08

Treasuries 6

Derivatives Page 22

Agencies Page 26

Mortgages Page 28

Credit Strategy Page 43

European ABS update Page 46

Covered Bond and Agency Update Page 47

A new 30Y Gilt Page 49

Japan Strategy Page 52

Japan Relative Value Page 56

Global Relative Value Page 59

Asia Page 65

Dollar Bloc Strategy Page 73

Global Inflation Update Page 77

Inflation-Linked Page 80

6 June 2014



Global

Rates Gov. Bonds & Swaps Inflation Rates Volatility

Bond Market Strategy

Carry on after the ECB The view: The ECB over-delivered in a move that will anchor the front-end, reduce volatility, steepen long-dated forwards, support peripheral spreads and be constructive for breakevens. The outright duration call is torn between expensive valuations and an environment which will be supportive for carry.

The rationale: The ECB's package exceeded expectations primarily thanks to the provision of up to a 4-year targeted LTRO (TLTRO) at effectively a fixed rate of 25bp. There is an initial cap based on the stock of loans and net-lending conditions associated with the TLTRO (modeled on the BoE's FLS). The cap will limit the initial take out to a maximum of EUR400bn in aggregate. The net lending constraint is difficult to evaluate without further information regarding the benchmark against which it will be evaluated. However, even if these conditions are restrictive, they would become relevant only after September 2016, which implies that the ECB has effectively offered at a minimum 2 years of unconditional LTRO (subject to an initial size cap and no material restrictions in terms of use of liquidity which although hinted at by Draghi have not been confirmed otherwise). In addition, the 4-year commitment together with the extension of the full allotment to December 2016 will reinforce forward guidance (at least until inflation starts normalizing). This will encourage front-end carry trades. The negative deposit rate is the least relevant in this policy arsenal. It is unlikely that it will materially affect the behavior of core banks, which is probably why the ECB has been so reluctant to go into negative territory and signaled that the rate cut avenue is exhausted.

The take up of the TLTRO is difficult to gauge at this stage. At a minimum, it should be used by banks to roll the maximum allowable amount from the existing vLTRO. Although the outstanding LTRO amount and the implied TLTRO cap are similar (~EUR400bn), the different distribution across countries implies that only ~EUR225bn can be rolled initially as the cap will constrain peripheral banks, while core banks cannot roll over more than the outstanding amount. Beyond rolling the existing LTRO, the following two considerations will determine the additional demand for the TLTRO. First, while core banks don't need liquidity, they do need term funding. At current pricing, the TLTRO compares favorably with alternative sources of term funding for core banks. The take-up will thus depend upon whether other issues trump the economic attractiveness. For peripheral banks, the net-lending test will be more relevant in determining the potential for additional take up next year. Given the lack of information on the relevant benchmark, this is difficult to assess for now. In any case, irrespective of the take up of the TLTROs the interest rate on the TLTROs can drag down funding costs for banks in core countries as the TLTRO interest rate acts as an alternative cost of funding against which to compare market funding costs. Finally, Draghi signaled that the ECB is likely to embark on an ABS purchase program (which does not seem to be restricted to SMEs assets) as soon as the appropriate conditions (regulation etc.) are in place. We estimate that the universe of ABS held by investor that meets current ECB repo eligibility criteria placed with investors is about EUR185bn.

There is the thorny issue as to whether any of this will actually support lending to the real economy. The process of balance sheet repair and compression of funding costs was anyway under way as evidenced by the latest ECB's bank lending survey and general improvement in funding conditions. The ECB support will at least support this process even one can be skeptical as to whether it will materially impact lending to the real economy.

Francis Yared


Alexander Duering, CFA


Markus Heider


Abhishek Singhania


Jerome Saragoussi


George Saravelos


6 June 2014



Finally, Draghi’s comments suggest that the ECB hopes to be done easing for now. He clearly signaled that further rate cuts were unlikely and only paid lip service to the prospect of large scale asset purchases. Moreover, the downward revisions to the inflation forecast have effectively “provisioned” for potential downside surprises.

The trades: We were positive peripheral assets, even without aggressive ECB support. The ECB decision reinforces our positive bias (e.g. 10Y BTP-Bund tightener) and the prospect of ABS purchases will be particularly favourable for peripheral ABS. The front-end of the curve should be anchored, but assuming that eonia will not easily trade in negative territory, the 2s5s curve is more likely to flatten in a rally. This could be expressed via 1m forward 2s5s bull flatteners. Long dated forwards were already pricing significant QE and/or deflation risks, which should be priced out over time. However, the positive carry environment may delay this repricing for now, and we exit the 2s10s steepener recommended last week. A careful expression of a bearish bias on longer dated forwards could be expressed conditionally via mid curve 5y5y payer structures. Alternatively, the long dated forward steepeners suggested last week (3-year forward 10s30s via caps) offer an attractive carry opportunity. Finally, the environment should be favourable for linkers, which can be expressed in real rates term in 5Y or in breakeven terms in 10Y.

What did the ECB deliver?

As expected the ECB announced a whole package of measures. The measures include cuts to its policy rates, the decision to take the deposit facility rate into negative territory, extension of the fixed rate full allotment procedure until end 2016, ending the SMP sterilization, targeted LTROs and announcement of preparatory work on ABS purchases. Whether this has the desired impact on inflation and credit dynamics will be seen only in the medium-term. However, the near term expected impact on financial markets is also highly uncertain given the lack of details regarding some elements of the ECB’s decision and the complex forces in play with regard to banks’ behaviour in uncharted territories.

Policy rates: On the policy rate front, the ECB delivered as expected. The main refi and deposit facility rate were cut by 10bp each to 15bp and -10bp respectively. The marginal lending facility rate was cut by 35bp to 40bp which should help contain the recent volatility in Eonia. However, the ECB did indicate that a floor for rates has been reached as it dropped the reference to “lower rates” in its forward guidance statement. This is only marginally negative for the front-end as the market was not in any case expecting the ECB to take rates deeply into negative territory. Note that, the negative deposit facility rate was made fully effective by the decision to charge the negative deposit facility rate on banks’ average reserve holdings in excess of the minimum requirements.

Extension of Fixed rate full allotment procedure: The fixed rate full allotment procedure for MROs, 1M and 3M LTROs was extended from mid 2015 to end 2016. This was slightly longer than expected and effectively provides banks in the Eurosystem funding certainty (as long as they have sufficient collateral) until the end of 2016. It is also a way for the ECB to strengthen its forward guidance. If the market fully believes that the ECB will not hike rates until Dec-16 then, forward Eonias until Dec-16 should trade below or close to 15bp. However, the fact that forward Eonias trade above 15bp by around mid 2016 suggests that this is as yet not fully priced in.

End of SMP sterilization: The liquidity injected via the SMP will no longer be sterilized. This will result in a permanent injection of liquidity of EUR 164.5bn

6 June 2014



(in practical terms only 119bn as only this amount was sterilized at the last operation). The increase in excess liquidity should push Eonia lower in the refi rate - deposit facility rate corridor. Given that a negative rate will be charged on the deposit facility rate, the response of banks in these countries will be critical. The options for the banks include

Reducing their borrowings, if any, from the ECB

Bearing the negative interest rate charged on the deposit facility.

Passing on the cost to depositors

Passing on the cost to borrowers

Buying any asset which is considered to be safe and can be treated as an alternative to the ECB deposit facility

In the first case, the excess liquidity in the Eurosystem will decrease and therefore the lowering of Eonia might be short lived. The table below shows the distribution of excess liquidity and outstanding ECB operations by country within the Eurozone. France and Germany stand out as countries where there is an overlap between excess liquidity and outstanding ECB operations. We could see some reduction in borrowings via the ECB by French and German banks in the coming weeks although this could be limited by fragmentation within the domestic banking systems.

SMP sterilization is likely to result in more excess liquidity in the countries

which currently have excess liquidity while the ability to repay will depend on

the amount of outstanding ECB operations in these countries Country Excess Liquidity (EUR bn) Outstanding ECB operations (EUR bn)

Germany 47 51

Netherlands 30 7

France 24 56

Luxembourg 18 3

Finland 12 0

Austria 8 8

Belgium 7 16

Italy 4 210

Ireland 2 31

Portugal 2 45

Greece 0 59

Spain 0 184

Sum 154 672 Source: Deutsche Bank * The excess liquidity and outstanding ECB operations are based on latest data available from individual national central banks and is a mix of data as of March and April 2014

As far as the last option is concerned, if banks buy the asset from another entity within the same banking system or from an entity within a banking system which is also sitting on excess liquidity the problem will be passed on from one bank to another. Buying of assets can be a solution to the problem only if it results in the liquidity being transferred to a banking system which does not have excess liquidity. The banks in the system which get this liquidity could then reduce their borrowing from the ECB and excess liquidity will decline. However, it is not clear that a charge of 10bp on the deposit facility will be sufficient to encourage a change in the credit appetite of the banks in the countries with the excess liquidity. Therefore, it remains to be seen whether the liquidity injected via the end of the SMP sterilization is repaid or remains in the system for an extended period of time.

6 June 2014



Any acceleration in the pace of LTRO repayments announced each week or a reduction in the take up at the weekly MROs could be indicative of the liquidity injected via the end of the SMP sterilization declining over time.

Targeted LTROs: The announcement in regard to the targeted LTROs (TLTROs) is the biggest discussion point from this ECB meeting. The key elements of the TLTROs are

— Maturity: All the TLTROs will expire in September 2018

— Take up window: Two TLTROs will be conducted in September and December 2014. In addition, from March 2015 to June 2016, all counterparties will be able to borrow quarterly. Therefore, there will be in all 8 TLTROs

— Limit on amount borrowed: Counterparties will be entitled to borrow, initially, 7% of the total amount of their loans to the euro area non-financial private sector, excluding loans to households for house purchase, outstanding on 30 April 2014. This is applicable for the TLTROs conducted in September and December 2014.

For the quarterly window from March 2015 to June 2016, all counterparties will be able to borrow, up to three times the amount of their net lending to the euro area non-financial private sector, excluding loans to households for house purchase, over a specific period in excess of a specified benchmark.

— Interest rate: The interest rate on the TLTROs will be fixed over the life of each operation, at the rate on the Eurosystem’s main refinancing operations (MROs) prevailing at the time of take-up, plus a fixed spread of 10 basis points. In effect, at the moment this would imply a cost of 25bp

— Repayment options: Starting 24 months after each TLTRO, counterparties will have the option to make repayments

— Conditionality on lending to the real economy: Those counterparties that have not fulfilled certain conditions regarding the volume of their net lending to the real economy will be required to pay back borrowings in September 2016

The maturity of the TLTROs was clearly greater than our and consensus expectations. Although the fixed interest rate is clearly an additional attractive feature the spread of 10bp does mean that the regular ECB refinancing operations (one week MRO, 1M and 3M LTROs) will be cheaper than the TLTRO funds.

The biggest uncertainty remains about the potential size of take up of these TLTROs. For the peripheral banks who have outstanding 3Y LTROs which are maturing in Jan/Feb 2015 the Sep-14 & Dec-14 TLTROs offer an opportunity to rollover some of the maturing LTRO funding into longer term funding. Even if these peripheral banks are unable to meet the conditions regarding lending to the real economy they will have rolled over the LTRO funding until at least Sep-16.

Banking systems in core countries are not constrained by funding given that they have excess liquidity and they have been prepaying their 3Y LTROs. However, to the extent that the 4Y term funding is a cheaper alternative to market based funding or term-deposits it would make economic sense for these banks to take up funds at the TLTROs.

The table below provides the country breakdown of the limit on funding available under the TLTROs in the Sep-14 and Dec-14 rounds based on the

6 June 2014



limit of 7% of stock of loans to the real economy. We also provide the country breakdown of the estimated outstanding 3Y LTROs and outstanding debt securities which are maturing over the next two years. For banks in peripheral countries the outstanding 3Y LTROs exceeds the funds available as per the 7% limit. For banks in core countries the demand for the TLTRO funding could largely be driven by replacement of upcoming redemptions of debt securities. The issue will be whether despite the attractive terms of the TLTRO funding, the potential stigma limits the take up from these banks. In addition, replacement of maturing term deposits could be another source of demand for TLTRO funds for banks in core countries.

Limit on the first two rounds of the TLTRO and the possible sources of

demand at these TLTROs for the major Eurozone economies TLTRO limit based

on 7% stock of loans

Estimate of outstanding 3Y LTROs (EUR bn)

Outstanding bank debt maturing over

next 2 year (EUR bn)

Possible source of demand for TLTRO

(EUR bn)

Italy 75 161 32 193

Spain 54 139 23 162

Portugal 8 35 2 37

Ireland 8 20 10 30

Belgium 10 12 12 24

Luxembourg 4 1 13 14

Finland 7 0 19 19

Greece 10 1 0 1

Austria 15 5 13 18

Netherlands 29 3 35 38

France 77 48 310 358

Germany 95 11 35 47

Sum 392 437 503 940 Source: Deutsche Bank, ECB *For the countries highlighted in grey the demand due to rolling over 3Y LTRO funding exceeds the TLTRO limit based on 7% stock of loans

For the banks in core countries the demand for TLTRO funding will be driven largely by replacing market based/deposit funding for central bank term funding at attractive levels. The potential economic advantage of such a switch is highlighted in the table below which shows the average interest rate paid on deposits with maturity exceeding 2 years and the average cost & maturity of senior bank debt. The 25bp cost of funding under the TLTRO compares very favourably with the alternative sources.

Interest rate paid by banks on term deposits and senior bank debt is

significantly higher than the TLTRO funding cost Interest rate on new

deposits with maturity >2 years (%)

Senior bank debt data from iBoxx Investment Gradeindex

Yield on senior bank debt (%)

Average maturity of senior bank debt

Germany 1.24 1.36 4.07

France 2.24 1.46 5.00

Netherlands 2.69 1.37 4.41

Austria 1.49 1.80 3.42

Finland 1.08 1.15 3.87 Source: Deutsche Bank

Beyond the first two rounds of the TLTRO, the take up will depend on the definition of the benchmark level of lending which “takes into account” the net lending over the 12 month period upto April 2014. However, given the lack of clarity about the definition of the benchmark level it remains very hard to make

6 June 2014



any assessment of the possible size in these TLTROs. The details regarding the forced early repayment in September 2016 for banks who do not meet certain lending criteria to the real economy also remains unclear. Therefore, it is probably safer to consider these TLTROs as unconditional LTROs with size limits until September 2016 and conditionality which is as yet unclear for the remaining 2 years.

ABS purchases: The decision to intensify the preparatory work related to outright purchases of ABS remains lacking in details. Further details will be announced in due course and the ECB will work with other relevant institutions especially in regard to the regulatory environment.

6 June 2014



United States

Rates Gov. Bonds & Swaps Rates Volatility

US Overview

We think we are in a lower range for spot 10s for a while centering around 2 ½ percent. Assuming the economy bumbles along 10s should track their forwards into year end.

It’s not that the recovery isn’t good but it’s not good enough to warrant higher real rates. Central banks know that only too well. Bloated debt ratios mean they need to restore much higher nominal growth or be prepared to own a large amount of the same debt indefinitely.

There has been a lot of easing of late. It started with FX reserve managing stepping in to do “foreign QE” as the Fed has pulled away. The ECB is now on the same path. As we struggle to ramp nominal growth significantly higher this simply reinforces the view of stretching for yield and spread.

Flattening trades are troublesome for negative carry and we don’t think we are close enough to a Fed lift off to beat the forwards on the short leg. This biases us still to receiving the longer dated forwards, particularly on any dips.

Foreign investors have become a major source of demand for Treasuries during Fed taper. In the latest Federal Reserve Flow of Funds data, foreign investors bought 60% of Treasury supply last quarter, followed by the Fed and banks. Fed purchases accounted for 42% of Treasury supply, down from about 70% during QE3 in 2013. Bank holdings of Treasuries went up by $47 billion; the increase was consistent with other official data on banks’ Treasury holdings.

Not to our surprise, there was little evidence of strong demand for bonds by pension investors last quarter in the Flow of Funds report. Total pension holdings of Treasuries increased about $9 billion. The holdings of Credit Market Instruments among Defined Benefit Plans in Private Pension Funds almost held flat in Q1 2014 ($728.8 billion) compared to Q4 2013 ($729.1 billion).

Most bond funds outperformed the index in total return last week as bond yields rose. Many bond funds are still short relative to their benchmark index’s duration, according to surveys. However, most funds have outperformed the index in total return on a year-to-date basis, perhaps thanks to a credit overweight and spread tightening. Although they have tended to underperform in bull flatteners in price returns, their total return performance over a longer period should be fine as long as credit performs well.

And you thought risk premium was already dead

The overriding conclusion from recent policy actions, investor preferences and economic data is that risk is still very much “on” and the search for yield carries on with a vengeance. At one level this may appear to befit a resigned response to a path of least resistance in a web of confusion! At another level it is plainly obvious and makes sense of everything from the corporates’ lack of enthusiasm to invest to beggar thy neighbor exchange rate manipulation and an FOMC that seems to be getting more hawkish day by day.

Dominic Konstam

Research Analyst (+1) 212 250-9753 [email protected]

Aleksandar Kocic


Alex Li


Stuart Sparks


Daniel Sorid


Steven Zeng, CFA


6 June 2014



We have refreshed our official forecasts for rates and expect 10s to trade around current levels to slightly lower before rising modestly in q4 towards 2.8 percent. The curve should have a flattening bias so 5y5y can still make new lows (3 ½ percent implied) in coming months.

On the economic outlook we are reminded of former ECB President Trichet when asked to sum up in one word the performance of the European economy. “Good”, he responded. He was asked to elaborate, perhaps in two words. “Not good”, he responded. And that is true not just for Europe but for the US and the global economy generally.

Our standards may be high after the pre-crisis boom. But they are high because growth was strong for good reasons: productivity. Productivity makes growth feel better because by definition you don’t work so hard to achieve the same results. Worse, labor is a finite resource. Full employment limits growth. The US is driving at a brick wall and at 200k+ jobs a month the brick wall is on the horizon for 2015. Taking the household survey of around a 155m labor force, if the potential workforce grows at ½ percent, as the CBO estimates, monthly employment wouldn’t even be 100k. If we assume NAIRU is 0.8 percent lower and theirs is another 1 ½ percent discouraged (adjusting for participation rate cohorts), we are two years away from full employment if job growth is 200k – a year and half away if it is 250k. So it is quite reasonable to think about Fed tightening next year assuming the Fed might want to be off first base before we reach full employment albeit subject to inflation.

Capital labor ratio doesn’t suggest stronger productivity

on the horizon

Labor input vs. productivity

-0.500.511.522.533.544.55

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

19901 20001 20101

capital labor ratio yoy, leads 4qtrsproductivity rhs

-2

-1

0

1

2

3

4

5

6

-12-10

-8-6-4-202468

10

19913 19961 20003 20051 20093 20141

labor input

productivity rhs

Source: SMR and Deutsche Bank Source: CME Group and Deutsche Bank

Productivity is weak because of weak capital spending. The capital labor ratio remains lackluster. The Fed being able to raise rates is therefore not the embrace of a strong recovery and job well done but a reluctant acceptance that this is as good as it gets for real growth and there is some responsibility for containing inflation (expectations). Equilibrium real rates are clearly lower than we would ideally like and the issue for whether the terminal Funds rate is around neutral or temporarily above is a function of how long the Fed may choose to delay raising rates and at what ultimate risk to inflation. The earlier the move the more like the rates will cap out lower. Our best estimate of terminal funds is still closer to 3 or below rather than 4 or above and we still think 5y5y OIS will define a low bound below 3 having been 4 ¼ at the start of the year.

6 June 2014



Low nominal growth is a major problem for servicing debt. We got through the housing and global financial crisis not by getting rid of the debt but by shuffling it around and mainly putting onto the government balance sheet. Since then deleveraging has seen some net improvements, particularly in the UK and US. But there is legitimate concern for sovereign debt load and especially in the vulnerable parts of Europe. Either way it is still a central bank problem. Moreover it constrains fiscal policy going forward in the event of a cyclical slowing.

QE is the ultimate solution to sovereign debt burdens. It literally is a “now you see it, now you don’t” moment. It is anathema to the days when governments relied on deficit financing. But this is ex post deficit financing. It is a quid pro quo for fiscal discipline. It recognizes the fiscal impasse in the US and Europe. It also can also be reversed at anytime. The ECB moving to join the BoE, the Fed and the BOJ represents an inevitable conclusion to finding the ultimate remedy for the crisis. For all the intricacies of the actual easing via TLTROs at four years etc.. the bottom line is that there is more liquidity for longer and the ECB is the backstop for risk in the periphery – even Greece. Most importantly it represents another succumbing to the need for lower for longer that necessarily pushes investors further out the credit and duration spectrum. In the US when short rates hadn’t hit zero curves could bearishly steepen. When the front end is already pinned scope for steepening is limited and bull flattening works further out the curve.

Foreign QE is as important as domestic QE. Domestic QE can’t work in a vacuum – negative deposit rates neither – if trading partners resist currency appreciation. In the race top the bottom currency manipulation is rife from China to the Swiss to Europe. But that also means lots of reserves. What the Fed buys in QE, China accumulates in cash that eventually needs to be reinvested. They can try to diversify but spread differentials become so extreme that there is a limit to how far Spain can trade through the US. FX reserve managers therefore reinforce, effectively replace domestic QE, when the central bank steps back - in this case the Fed. For all those bond bears who thought taper meant higher yields this is the cruel reality.

Domestic debt GDP ratios General Govt debt GDP ratios

200250300350400450500550600650700

20001 20034 20073 20112

UK US

Euro Japan

100

120

140

160

180

200

220

240

0

20

40

60

80

100

120

140

20001 20034 20073 20112

UK US

Euro - gengovt Japan rhs

Source: Haver and Deutsche Bank Source: Haver and Deutsche Bank

6 June 2014



General government less C bank assets as % GDP US Govt less C bk assets less EM FX reserves as % GDP

150155160165170175180185190195200

2030405060708090

100110120

20062 20101 20134

UKUSEuro - gengovt less cbk assetsJapan rhs

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

40

45

50

55

60

65

20062 20101 20134

US Gen govt less Cbk and ReservesEM Reserves

Source: Haver and Deutsche Bank Source: Haver and Deutsche Bank

If we are right, the outlook for rates and spreads is an ongoing compression in risk premium. The curve could become very flat which might have to be justified in still low volatility, which is consistent perhaps with a constrained nominal growth outlook. No doubt there will be bouts of doubt surrounding an inflation surprise or a central bank outburst but a more substantive shift in regime relies on a more meaningful adjustment to the real interest rate equilibrium view. This isn’t really about inflation but about real growth after all. Inflation might be important if the central banks were so slow and so tolerant that they allowed it to rise persistently and substantially over 2. The problem with that is that financial markets would unravel so quickly it seems unlikely that they could lean that way for long. So bear tail risk hedges are certainly warranted but should not represent the core strategy.

There are some more tangible endogenous risks to the view. No one really understands the corporate capex funk. Tobin’s Q is still around 1. The risks of capex outright M and A activity. Maybe real rates are simply still too high. So another round of lower yields and higher equities could deliver the capex we need that segways labor led to productivity led growth. It may even need to take place over a token Fed tightening cycle – just when those wage increases look likely to be passed into higher prices, the Fed raises rates/ A Greenspan moment akin to opportunistic disinflation.

Exogenous risks are possible too. We think a unified (Republican) Congress could actually strike an oil-financed fiscal package that re rates growth expectations and releases capex spending and a new leverage cycle. Even without the oil, anything fiscal represents a similar risk. Larry Summers bemoans the fact that governments won’t borrow at sub 1 percent real for 30 years. These are risks to monitor and hedge but for now not the immediate focus.

Beyond the domestic economic view an additional risk factor for the US rate outlook is clearly persistent global – and in particular European – disinflation. To its credit, the ECB has gone to great lengths to eliminate any impediments to the supply of credit to the European economy, and while the toolkit is perhaps running low after last week, the ECB still has private sector and public sector asset purchases at its disposal. The market response was however somewhat muted as HICP swaps were only modestly higher and the curve steepened but bullishly and in the end 5y5y nominal rates fell.

6 June 2014



There are several points of relevance to the US. Given the terms of the TLTRO Europe is potentially looking at four years or more of financial repression in which large excess reserve balances chase core rates lower and the subsequent hunt for yield steadily drives significant outperformance in peripheral sovereign debt and risk assets more generally. As we noted recently, EURUSD basis swap compression has on balance cheapened US assets in euro terms, and to the extent that capital leaves the Eurozone in search of yield some is likely to end up in Treasuries.

The second point is that beggar-thy-neighbor currency practices on balance make it more difficult for Fed policy to de-couple from the rest of the world. If the Fed begins its exit the ECB can ease further by doing essentially nothing, or at least nothing as hawkish as the Fed. This would provide some fillip to European growth but at a cost to growth in the US on the margin. The point here is that it will be increasingly important that US growth accelerates over time as our colleagues in economics and their counterparts at the Fed forecast.

The third point is that the efficacy of the latest unconventional measures remains uncertain. While the ECB has proved aggressive and innovative, there is some chance the new measures will have limited effect simply because the problem is demand for, rather than supply of, credit that is the problem.

Our colleagues in European economics have enumerated four points of uncertainty with the package of new measures. First, the mechanism for ensuring that TLTRO borrowing ultimately funds the non-financial private sector is unclear and the risk remains that borrowing could instead be used to repeat the carry trade whereby banks simply extend credit to high-yielding sovereigns. Moreover, lower funding costs might not be passed along to final borrowers to the extent that banks require higher intermediation spreads to compensate for credit risk or to rebuild capital (note this echoes one of Dudley’s recent arguments as to why long term equilibrium policy rates might need to be lower in the US). Third, the elasticity of demand for credit to changes in borrowing rates is unclear, and given the environment of de-leveraging this sensitivity might reasonably be expected to be lower than usual. Fourth, given ongoing repayments of the initial two vLTROs, one might construe liquidity creation of the latest measures simply as a (partial) replacement of maturing ECB loans to MFIs.

Quarterly change of total credit to non-financial private sector to changes in

non-financial borrowing spreads

-150

-100

-50

0

50

100

150

200

-50 -30 -10 10 30 50

Mon

thly

chg

non

fin c

redi

t, b

illio

n

Chg in Non-financial credit spreads

Pre Crisis

During/Post

Linear (Pre Crisis)

Linear (During/Post)

Source: Haver Analytics, Deutsche Bank,

6 June 2014



The chart above illustrates a regression of monthly changes in total MFI credit outstanding to the non-financial private sector (from the counterparts of M3 data) on monthly changes in non-financial borrowing spreads. The chart above segments the data at August 2007 as a proxy for the beginning of the crisis.

For the entire period, narrower spreads actually have been accompanied by lower levels of credit creation to the private sector. Segmenting the data at August 2007 as above, the pre-crisis fit is (only) mildly upward sloping while the post-crisis fit is more decidedly downward sloping. This latter perverse relationship is in fact stronger when looking at all-in borrowing rates rather than simply spreads. The cause for consternation is that there was not a stronger relationship pre-crisis. For all its efforts, the risk remains that the ECB is pushing on a string.

At its heart the risk is that inflation is low and growth sluggish due to a global common shock stemming ultimately from the after effects of the crisis and to some structural change in demand - higher precautionary savings by a workforce that is older and less productive on the margin.

The final point to make is that if the latest ECB measures are not accompanied by ongoing structural reform or worse yet if structural measures are reversed given the prospect for more ready financing of the public sector, then the monetary response of the ECB might in the end constitute another instance of kicking the can down the road. In this sense, there remains considerable political risk across the Eurozone.

Supply and demand for Treasuries during Fed taper

Foreign investors have become a major source of demand for Treasuries during Fed taper. In the latest Federal Reserve Flow of Funds data, foreign investors bought 60% of Treasury supply, followed by the Fed and banks. Fed purchases accounted for 42% of Treasury supply, down from about 70% during QE3 in 2013. Bank1 holdings of Treasuries went up by $47 billion; the increase was consistent with other official data on banks’ Treasury holdings.

Foreign investors bought 60% of Treasury supply in Q1 2014

Taper 4Q2013 1Q2014 Change % supplyHousehold sector 935$ 841$ (94)$ -36%Foreign investors 5,804$ 5,960$ 156$ 60%Fed 2,209$ 2,320$ 111$ 42%Banks 217$ 264$ 47$ 18%Insurance 267$ 271$ 4$ 1%Pension 713$ 722$ 9$ 4%Money market funds 488$ 454$ (34)$ -13%Mutual funds 633$ 676$ 43$ 16%Brokers and dealers 136$ 153$ 17$ 6%Other 926$ 930$ 4$ 2%Treasury Issues 12,328$ 12,591$ 262$ 100%

Source: Fed and Deutsche Bank

1 U.S.-Chartered Depository Institutions, Excluding Credit Unions

6 June 2014



Supply and demand for Treasuries during QE3

QE3 4Q2012 4Q2013 Change % supplyHousehold sector 941$ 935$ (6)$ -1%Foreign investors 5,574$ 5,804$ 230$ 30%Fed 1,666$ 2,209$ 543$ 71%Banks 243$ 217$ (26)$ -3%Insurance 271$ 267$ (3)$ 0%Pension 644$ 713$ 69$ 9%Money market funds 458$ 488$ 30$ 4%Mutual funds 573$ 633$ 59$ 8%Brokers and dealers 247$ 136$ (110)$ -15%Other 953$ 926$ (26)$ -3%Treasury Issues 11,569$ 12,328$ 759$ 100%


In order to meet the demand from foreign investors, the Fed and bank, households and money market mutual funds had to sell $94 billion and $34 billion Treasuries, respectively, last quarter.

Not to our surprise, there was little evidence of strong demand for bonds by pension investors last quarter in the Flow of Funds report. Total pension holdings of Treasuries increased about $9 billion. The holdings of Credit Market Instruments among Defined Benefit Plans in Private Pension Funds almost held flat in Q1 2014 ($728.8 billion) compared to Q4 2013 ($729.1 billion). There was a $2.6 billion increase in their holdings of Treasuries, offset by a decline of similar amount in their holdings of Corporate and Foreign Bonds.

Pension funded status further improved in Q1 2014

60%

70%

80%

90%

100%

110%

120%

130%

Jan-85 Jan-90 Jan-95 Jan-00 Jan-05 Jan-10 Jan-15

Funded StatusPotential risks of underfunding: * 7% on new mortality tables;* 6-8% due to not fully investable discount rates


Private Pension Funds funded status improved further from 94.0% in Q4 2013 to 94.7% in Q1 2014. Defined benefit pension plans hold about $194 billion Treasuries, almost double the size from five year ago. Still, Treasuries are only about 6% of the total financial assets of those plans, and Credit Market Instruments are about 24% of the total financial assets. As there is room for this percentage to rise, pension demand for long dated bonds could be sizable. In the mid 1990s, Credit Market Instruments are about 34% of the total financial assets.

6 June 2014



Bond fund performance update

Most bond funds outperformed the index in total return last week as bond yields rose. Many bond funds are still short relative to their benchmark index’s duration, according to surveys. The SMRA index dipped to 98.1 this week from 98.2. However, most funds have outperformed the index in total return on a year-to-date basis, perhaps thanks to a credit overweight and spread tightening. The credit index has generated a higher year-to-date total return than the aggregate index. Although most funds have tended to underperform in bull flatteners in price returns, their total return performance over a longer period should be fine as long as credit performs well.

Most bond funds outperformed their benchmark index in

total return last week

Most bond funds have outperformed their benchmark

index in total return year to date

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

1 2 3 4 5 6 7 8 9 10 11 12 13 14W

eigh

ted 15 16 17 18 19

Inde

x20

Tota

l ret

urn

(%)

Performance ranking, sorted by return

5/29/14 to 6/5/14

0.0

1.0

2.0

3.0

4.0

5.0

6.0

1 2 3 4 5 6 7 8 9 10 11 12 13W

eigh

ted 14 15 16 17

Inde

x18 19 20

Tota

l ret

urn

(%)

Performance ranking, sorted by return

12/31/13 to 6/5/14

Source: Bloomberg Finance LP and Deutsche Bank Source: Bloomberg Finance LP and Deutsche Bank

6 June 2014



United States

Rates Gov. Bonds & Swaps

Treasuries

Foreign investors have become a major source of demand for Treasuries during Fed taper. In the latest Federal Reserve Flow of Funds data, foreign investors bought 60% of Treasury supply last quarter, followed by the Fed and banks.

Bank holdings of Treasuries went up by $47 billion last quarter; the increase was consistent with other official data on banks’ Treasury holdings.

Not to our surprise, there was little evidence of strong demand for bonds by pension investors last quarter in the Flow of Funds report. Total pension holdings of Treasuries increased about $9 billion.

Many bond funds are still short relative to their benchmark index’s duration, according to surveys. The put/call ratio on Treasury options remains elevated at around 1.6. Specs are net long Treasury futures, but set a record short in Eurodollar futures.

Fed custody holdings of Treasuries went up by a sizable $29 billion this week; it was the second highest increase this year and included the May month-end note auction settlements.

Who have bought Treasuries in taper?

Foreign investors have become a major source of demand for Treasuries during Fed taper. In the latest Federal Reserve Flow of Funds data, foreign investors bought 60% of Treasury supply, followed by the Fed and banks. Fed purchases accounted for 42% of Treasury supply. Bank2 holdings of Treasuries went up by $47 billion; the increase was consistent with other official data on banks’ Treasury holdings.

Foreign investors bought 60% of Treasury supply in Q1

2014

Supply and demand for Treasuries during Q1 2013

Taper 4Q2013 1Q2014 Change % supplyHousehold sector 935$ 841$ (94)$ -36%Foreign investors 5,804$ 5,960$ 156$ 60%Fed 2,209$ 2,320$ 111$ 42%Banks 217$ 264$ 47$ 18%Insurance 267$ 271$ 4$ 1%Pension 713$ 722$ 9$ 4%Money market funds 488$ 454$ (34)$ -13%Mutual funds 633$ 676$ 43$ 16%Brokers and dealers 136$ 153$ 17$ 6%Other 926$ 930$ 4$ 2%Treasury Issues 12,328$ 12,591$ 262$ 100%

QE3 4Q2012 1Q2013 Change % supplyHousehold sector 941$ 966$ 25$ 7%Foreign investors 5,574$ 5,722$ 148$ 44%Fed 1,666$ 1,796$ 130$ 39%Banks 243$ 230$ (13)$ -4%Insurance 271$ 267$ (3)$ -1%Pension 644$ 661$ 17$ 5%Money market funds 458$ 470$ 12$ 4%Mutual funds 573$ 628$ 55$ 16%Brokers and dealers 247$ 203$ (44)$ -13%Other 953$ 962$ 9$ 3%Treasury Issues 11,569$ 11,905$ 337$ 100%

Source: Fed and Deutsche Bank Source: Fed and Deutsche Bank

2 U.S.-Chartered Depository Institutions, Excluding Credit Unions

Alex Li


Steven Zeng, CFA


6 June 2014



In order to meet the demand from foreign investors, the Fed and bank, households and money market mutual funds had to sell $94 billion and $34 billion Treasuries, respectively, last quarter.

Defined Benefit Pension Plans

Equity and bond holdings by Defined Benefit Pension

Plans

$500

$1,000

$1,500

$2,000

$2,500

$3,000

$3,500

15%

20%

25%

30%

35%

Jan-84 Jan-94 Jan-04 Jan-14

Credit Market Instruments/Total FA

Defined Benefit Pension Plans: Total Financial Assets (NSA, Bil.$)

10%

20%

30%

40%

50%

60%

70%

Jan-06 Jan-08 Jan-10 Jan-12 Jan-14

Equities % total fin: Private Pension DB

Bonds % total fin: Private Pension DB

Source: Fed and Deutsche Bank Source: Fed and Deutsche Bank

Not to our surprise, there was little evidence of strong demand for bonds by pension investors last quarter in the Flow of Funds report. Total pension holdings of Treasuries increased about $9 billion (Private Pension Funds, State & Local Govt Retirement Funds, and Federal Government Retirement Funds). The holdings of Credit Market Instruments among Defined Benefit Plans in Private Pension Funds almost held flat in Q1 2014 ($728.8 billion) compared to Q4 2013 ($729.1 billion). There was a $2.6 billion increase in their holdings of Treasuries, offset by a decline of similar amount in their holdings of Corporate and Foreign Bonds.

Private Pension Funds funded status improved further from 94.0% in Q4 2013 to 94.7% in Q1 2014. Defined benefit pension plans hold about $194 billion Treasuries, almost double the size from five year ago. Still, Treasuries are only about 6% of the total financial assets of those plans, and Credit Market Instruments are about 24% of the total financial assets. As there is room for this percentage to rise, pension demand for long dated bonds could be sizable. In the mid 1990s, Credit Market Instruments are about 34% of the total financial assets.

Positioning update

Most bond funds outperformed the index in total return last week as bond yields rose. Many bond funds are still short relative to their benchmark index’s duration, according to surveys. The SMRA index dipped to 98.1 this week from 98.2. The put/call ratio on Treasury options remains elevated at around 1.6.

Most funds have outperformed the index in total return on a year-to-date basis, perhaps thanks to a credit overweight and spread tightening. The credit index has generated a higher year-to-date total return than the aggregate index. Although most funds have tended to underperform in bull flatteners in price returns, their total return performance over a longer period should be fine as long as credit performs well.

6 June 2014



Fed custody holdings of Treasuries went up by a sizable $29 billion this week; it was the second highest increase this year and included the May month-end note auction settlements.

The SMRA index suggests money managers are still

short

The put/call ratio on Treasury options remains elevated at

around 1.6

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5 96

97

98

99

100

101

102

103

Jun-11 Jun-12 Jun-13 Jun-14

SMR Weighted Index 10-year yield (rhs, inverted)

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.25

3/1/07 3/1/09 3/1/11 3/1/13

Put/call ratio Average

Source: SMRA and Deutsche Bank Source: CME Group and Deutsche Bank

Spec positions in Treasury futures Spec positions in Eurodollar futures

-600,000

-400,000

-200,000

0

200,000

400,000

1/1/08 1/1/09 1/1/10 1/1/11 1/1/12 1/1/13 1/1/14

Aggregate net spec positions in all Treasury futures in TY equivalents

-2,000,000

-1,500,000

-1,000,000

-500,000

0

500,000

1,000,000

1,500,000

2,000,000

1/1/08 1/1/09 1/1/10 1/1/11 1/1/12 1/1/13 1/1/14

Net spec in ED

Source: CFTC and Deutsche Bank Source: CFTC and Deutsche Bank

Fed custody holdings of Treasuries went up by a sizable $29 billion this week

-120

-100

-80

-60

-40

-20

0

20

40

60

1/1/09 1/1/10 1/1/11 1/1/12 1/1/13 1/1/14

$ Bi

llions

Treasuries: $2,970.7B, last change $28.9B


6 June 2014



Auction preview: 3s, 10s, Bonds

Treasury will sell $62 billion notional securities worth $58 billion in ten-year equivalents through three- and ten-year notes, and 30-year bond auctions next week. These auctions will settle on the following Monday, June 16 against $32 billion coupon debt maturing on the same day. As expected three-year note auction size has been reduced further by a billion to $28 billion. 3s and 10s were well received at May auctions, but customer demand for the 30-year bond was relatively poor. Direct bidder’s average participations in recent three auctions were solid as compared with the below average takedowns by indirect bidders.

3-year note Direct bidder participations were solidly above the average 17.5% in last two auctions. Indirect bidders however have been weak since last February averaging 28.4% as compared with their one-year average of 33.7%. Combined customer participations were decent in the last two auctions as compared with the average 51.2%. The bid-to-cover ratio in May beat the average 3.30. Notably, dealer net shorts in two- to three-year Treasuries have increased by $3 billion since the last auction to $5.5 billion as of May 28.

3-year note auction statistics

Size ($bn)

Primary Dealers

Direct Bidders

Indirect Bidders

Cover Ratio

Stop-out Yield

1PM WI Bid

BP Tail

1yr Avg $30.5 48.8% 17.5% 33.7% 3.30 -0.2

May-14 $ 29.0 47.3% 24.5% 28.1% 3.40 0.928 0.93 -0.2

Apr-14 $ 30.0 48.8% 24.0% 27.3% 3.36 0.895 0.895 0.0

Mar-14 $ 30.0 54.6% 15.5% 29.9% 3.25 0.802 0.801 0.1

Feb-14 $ 30.0 41.3% 16.6% 42.0% 3.42 0.715 0.72 -0.5

Jan-14 $ 30.0 49.4% 22.6% 28.0% 3.25 0.799 0.797 0.2

Dec-13 $ 30.0 49.6% 12.0% 38.4% 3.55 0.631 0.637 -0.6

Nov-13 $ 30.0 47.3% 19.4% 33.3% 3.46 0.644 0.645 -0.1

Oct-13 $ 30.0 45.8% 19.7% 34.4% 3.05 0.710 0.718 -0.8

Sep-13 $ 31.0 46.8% 20.0% 33.1% 3.29 0.913 0.918 -0.5

Aug-13 $ 32.0 44.7% 14.0% 41.4% 3.21 0.631 0.636 -0.5

Jul-13 $ 32.0 51.5% 13.0% 35.6% 3.35 0.719 0.721 -0.2

Jun-13 $ 32.0 58.4% 8.4% 33.1% 2.95 0.581 0.577 0.4 Source: US Treasury, SMRA, and Deutsche Bank

10-year note (re-opening) Except for the one in April, every auction since January this year has recorded solid customer participations. Customers have averaged 65.6% in the last five auctions as compared to their one-year average of 63.3%. Indirect bidders beat their average 45.2% takedown in four of the last six auctions while direct bidders beat theirs (18.1%) in just two. Bid-to-cover ratio of 2.63 at the May auction was slightly below the average 2.65, but the auction stopped through by 0.6bps. Previous two auctions, however, had recorded solid bid-to-cover ratios. Dealer net shorts in seven- to 11-year Treasuries hit the highest level of $11.9 billion on May 28 since June 19 last year.

6 June 2014



10-year note auction statistics

Size ($bn)

Primary Dealers

Direct Bidders

Indirect Bidders

Cover Ratio

Stop-out Yield

1PM WI Bid

BP Tail

1yr Avg $ 22.0 36.7% 18.1% 45.2% 2.65 -0.3

May-14 $ 24.0 29.1% 21.6% 49.3% 2.63 2.612 2.618 -0.6

Apr-14 $ 21.0 40.1% 15.2% 44.7% 2.76 2.720 2.712 0.8

Mar-14 $ 21.0 29.1% 27.5% 43.4% 2.92 2.729 2.743 -1.4

Feb-14 $ 24.0 34.1% 16.2% 49.7% 2.54 2.795 2.799 -0.4

Jan-14 $ 21.0 39.8% 13.6% 46.6% 2.68 3.009 3.008 0.1

Dec-13 $ 21.0 40.4% 10.6% 48.9% 2.61 2.824 2.816 0.8

Nov-13 $ 24.0 33.8% 18.6% 47.7% 2.70 2.750 2.754 -0.4

Oct-13 $ 21.0 40.2% 21.2% 38.6% 2.58 2.657 2.667 -1.0

Sep-13 $ 21.0 33.8% 29.6% 36.6% 2.86 2.946 2.966 -2.0

Aug-13 $ 24.0 38.5% 15.2% 46.3% 2.45 2.620 2.62 0.0

Jul-13 $ 21.0 45.2% 16.3% 38.6% 2.57 2.670 2.668 0.2


30-year bond (re-opening) May auction recorded lowest customer participation of 48.8% (avg. 56.7%) since last June. Direct bidder takedown was the lowest 8.4% (avg. 15.5%) in as many months while indirect bidders too were slightly below the average at 40.4%. Bid-to cover ratio of 2.09 was the lowest since August 2011 and compares with its one-year average 2.35. The hefty 3bps tail was also the largest of any auction in last 11 months. On May 28, dealer net longs in long-term Treasuries hit the record $13 billion level of the year.

30-year bond auction statistics

Size ($bn)

Primary Dealers

Direct Bidders

Indirect Bidders

Cover Ratio

Stop-out Yield

1PM WI Bid

BP Tail

1yr Avg $14.0 43.3% 15.5% 41.1% 2.35 0.3

May-14 $ 16.0 51.2% 8.4% 40.4% 2.09 3.440 3.410 3.0

Apr-14 $ 13.0 38.8% 17.9% 43.3% 2.52 3.525 3.531 -0.6

Mar-14 $ 13.0 48.6% 12.6% 38.8% 2.35 3.630 3.615 1.5

Feb-14 $ 16.0 40.8% 13.9% 45.3% 2.27 3.690 3.699 -0.9

Jan-14 $ 13.0 38.1% 17.5% 44.4% 2.57 3.899 3.909 -1.0

Dec-13 $ 13.0 41.4% 12.5% 46.0% 2.35 3.900 3.894 0.6

Nov-13 $ 16.0 46.5% 18.3% 35.3% 2.16 3.810 3.795 1.5

Oct-13 $ 13.0 35.5% 22.6% 41.9% 2.64 3.758 3.779 -2.1

Sep-13 $ 13.0 41.7% 20.6% 37.7% 2.40 3.820 3.827 -0.7

Aug-13 $ 16.0 42.7% 17.1% 40.2% 2.11 3.652 3.644 0.8

Jul-13 $ 13.0 43.4% 16.3% 40.2% 2.26 3.660 3.674 -1.4


6 June 2014



June Fed buyback schedule

Fed plans to remove about $25 billion of notional worth $28.6 billion in ten-year equivalents from the markets through 16 purchase operations in June. The frequency and weights of the operations have been kept the same as in May. TIPS purchases of the month are scheduled on Thursday, June 12 for about $0.7 billion takeout in notional or approximately $0.9 billion in ten-year equivalents.

Duration takeout of next week’s four purchases is about $5.5 billion in notional or $6.7 billion in ten-year equivalents. No operation is scheduled on Wednesday, June 18 when FOMC will release its next policy statement.

Fed Treasury buyback schedule for June 2014

Date Operation type Maturity Range Par Amt. ($b)

Average Duration

Avg DV01

10y equ. ($b)

2-Jun Treasury 15-Nov-24 15-Feb-31 0.475 9.6 13.28 0.73

3-Jun Treasury 31-Mar-20 31-May-21 1.75 5.8 6.23 1.26

4-Jun Treasury 15-Feb-36 15-May-44 0.975 16.8 18.32 2.06

5-Jun Treasury 15-Aug-21 15-May-24 2.5 7.3 8.05 2.32


10-Jun Treasury 31-Mar-19 29-Feb-20 2.875 4.9 5.09 1.69


12-Jun TIPS 15-Jul-18 15-Feb-44 0.675 11.3 12.70 0.90

16-Jun Treasury 30-Jun-18 28-Feb-19 2.625 4.2 4.39 1.33



23-Jun Treasury 31-Mar-20 31-May-21 1.75 5.8 6.23 1.26





Total 24.480 9.29 10.14 28.56 Source: Deutsche Bank, NY Fed

Distribution of monthly Treasury purchases by frequency and weight

Nominal Coupon Securities by Maturity Range TIPS

4 - 4¾ 4¾ - 5¾ 5¾ - 7 7 - 10

10 - 20

20 - 30 4 - 30 years

1 1 2 3 1 7 1 frequency 11% 12% 16% 29% 2% 27% 3% weight

Source: NY Fed, Deutsche Bank

6 June 2014



United States


Derivatives

The second stage of the unwind of monetary policy, return of volatility, continues to face strong resistance. The collapse of realized vol is no longer only the rates story, but is becoming a dominant feature across a broad range of asset classes. In less than 12 months, realized vol has declined to only a fraction of last year’s levels with the most dramatic decline in credit – HY by 80%, IG by more than 65% and rates, equities and FX by 50%. Although this is seen as potentially harmful in the long run, it is difficult to see a mechanism that could break the circuit.

In comparison, implieds, although lower for the year, are more resistant. On one hand, the status quo should persist as long as there is no urgency to make a change in monetary policy. On the other, there is a sense that things could, in principle, turn around at any moment.

We believe that in the short run, market conditions will continue to favor the carry trade, but would not rule out a breakout of the range and possible sell off in rates in the case of strong data. We are sellers of short dated straddles, but as we see both rates and risky assets as vulnerable to a gapping move in rates, we would cover the tails with high strike payers. We recommend short straddles overlaid with high strike payers:

Sell $100mn 3M5Y straddles at 29bp vs. buy $100mn 3M5Y 22bp OTM payers at 7bp, net takeout 22bp (107c).

Hearing the mermaids sing

The decline in rates volatility, which has been a signature mode of post-2008 monetary policy, is getting a broader dimension. The second stage of the unwind of monetary policy, return of volatility, continues to face strong resistance. Such a state of affairs, if it persists, could develop to be a siren’s song as it fosters misallocation of capital and causes long-term problems that will be difficult to manage. Yet, it is difficult to see a clear mechanism that could break the circuit. As a consequence of years of accommodative monetary policy and excess transparency, having given rise to the entire “culture of carry”, market positioning has evolved in such a way that it resists the change. The collapse of realized vol is no longer only the rates story, but is becoming a dominant feature across a broad range of asset classes.

To put things in perspective, we take the end of Jun 2013 as a reference point. This is the time when the dialogue about policy unwind began. Figure 1 shows current realized and implied vols across different assets in units of their respective Jun 2013 levels (aligned in order of their duration). Current realized has declined to only a fraction of last year’s levels with the most dramatic decline in credit – HY is currently one-fifth of what it was a year ago, IG about one-third – followed by rates and equities which have dropped to about half, together with FX, which is showing the same pattern.

Aleksandar Kocic


6 June 2014



Figure 1: Implied and realized vols across different assets expressed in units

of their Jun-2013 levels

Source: Deutsche Bank

In comparison, implieds, although lower for the year, are more resistant. Their story reflects several layers of non-linearity. On one hand, the status quo should persist as long as there is no urgency to change monetary policy. On the other, there is a sense that things could, in principle, turn around at any moment. A few strong numbers could ignite volatility as repricing of the new information could trigger a disorderly unwind of the carry trade, a mode that could be disruptive for risky assets. While there is a considerable amount of unresolved tension and a buildup of tail risk, it is difficult to convince investors that long gamma is worth the pain. This regime of market operation bears strong resemblance to previous episodes of complacency. Although the memory of their violent resolution is still fresh, it appears to be difficult to resist the seduction of the carry trade.

This dichotomy is articulated in two ways, a high volatility risk premium and a bid for forward vol. These are two different takes on the current situation. Figure 2 shows two snapshots of the risk premia across different assets, current and Jun 2013. We note a high divergence across two dates. While implieds have undergone a downward revision, it is really the collapse of realized that stands out. This is especially pronounced for risky assets whose vulnerability is exactly a function of persistent low realized vol. High vol risk premia reflect the fear that things can turn around at any moment. The current environment of low realized volatility and persistence of range-bound rates is an insufficient incentive to push vol lower. Every time gamma, e.g. 3M10Y, declines below 60bp, there are buyers. The right strategy, in our view, is to sell gamma but spend a fraction of the premium to cover the tails.

Figure 2: Volatility risk premia across different assets: current and 2013


6 June 2014



Another take on this story is forward vol. The market is essentially held hostage by monetary policy. As long as accommodation is in place, the carry trade and mortgage replication are keeping rates in a range and provide support for risky assets. Return of volatility and repricing across the board will happen when the stimulus is withdrawn. We believe the right strategy is to use the current conditions of excessive transparency and liquidity as a leverage to buy a long vol position in the future.

In the absence of any unanticipated shift in monetary policy, it is difficult to argue a massive turnaround in risky assets. A gapping move in rates that could be catalyzed by stronger-than-expected inflation or generally faster and more aggressive Fed hikes could have an adverse effect across the board, but rates would most likely lead the way in that case. We are facing tail risk along two main directions. Risky assets are likely to be most reactive in risk-off trade. Rates are already low and further complications along the lines of global growth slowdown or escalation of geopolitical tensions are likely to see a relatively modest repricing, within one-sigma, while a decline in risky assets could be substantial, possibly a multiple standard deviation in size.

The trade: Selling straddles with partial tail covering We consider another twist on selling gamma along the lines of straddle/strangle switches. As short vol position is more vulnerable to higher rates, our strategy would be to cover the tails only partially at higher rates. We choose the strike on the OTM payer so that the takeout reduction from the naked straddle flattens the exposure in high rates. The figure below shows the terminal profile of the resulting position together with its building blocks.

Figure 3: Overlaying short straddles with OTM payers

-30

-20

-10

0

10

20

30

-50 -40 -30 -20 -10 0 10 20 30 40 50

Straddle

OTM payer

Net


To achieve this payout, the OTM payer and the strangle must have the same breakevens, which implies that the strike of the OTM payer is the breakeven of the combined trade (Fig). If S denotes the price of a straddle, and P(K) is the price of a K bp OTM payer, when expressed in bp running, the strike satisfies:

)(KPSK −= . This always has a solution since the two lines (left & right hand side) have different slopes and intercepts. For 3M5Y options, the payer strike is 22bp OTM and the trade is specified as follows:

Sell $100mn 3M5Y straddles at 29bp vs. buy $100mn 3M5Y 22bp OTM payers at 7bp, net takeout 22bp (107c).

6 June 2014



The breakevens on the trade are 22bp away from the forward on both sides. The ageing and P&L profile across different time horizons is shown in Fig 4. It is assumed that curve rolls to forwards.

Figure 4: Ageing and P&L of partially covered short straddle position


While there are no losses in high rates, in terms of terminal payout, the trade is long gamma in a sell off, so the MTM exposure is favorable as vol is likely to increase with rates. In a rally, vol is likely to decline which is in tune with the short gamma exposure. The position is vulnerable to rally below the left breakeven with theoretically unlimited downside.

6 June 2014



United States

Rates Gov. Bonds & Swaps Credit Sovereigns

Agencies

Two dominant themes in the US agency market this year have kept valuation rich and agency spreads at or near all-time tights. The first is the decline in supply. Year-on-year new issuance of GSE and $SSA paper is down more than 30% from 2013 levels, totaling just $250 billion vs. $360 billion through the first five months of the year. The other main story is regulatory reform of the GSEs, which could potentially see Fannie Mae and Freddie Mac shut down in five years time and stop issuing debt all together. Other reform scenarios could keep Fannie and Freddie in the business, but their balance sheet limited to a much smaller size on an ongoing basis, and their credit worthiness enhanced through the companies’ renewed profitability and additional government support measures. We take a cautious view on spreads and think they are vulnerable to modest widening pressures. Investors should stay neutral in allocation to agencies.

$SSA and GSE new issuance (>1y maturity) down 30%

from 2013 so far this year

Yoy issuance vs. change in 3y agency spread

$0

$10

$20

$30

$40

$50

$60

$70

$80

$90

$100

$0

$10

$20

$30

$40

$50

$60

$70

$80

$90

$100

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

bnbn$SSA 2013 GSE 2013

$SSA 2014 GSE 2014

(20)

(15)

(10)

(5)

0

5

10

15

-60%

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

Jan-11 Jul-11 Jan-12 Jul-12 Jan-13 Jul-13 Jan-14

yoy issuance (lhs) chg in agy spread, 3m/3m (bp, rhs)

Source: Bloomberg and Deutsche Bank Source: Bloomberg and Deutsche Bank

Supply remains tight but danger is if demand leaks Investors looking for GSE or $SSA issuance to pick up probably won’t find it this year. Fannie and Freddie have been reliably reducing their outstanding debt at the regulators-required rate of 15% a year. Their issuance of new debt has declined at even a greater rate of nearly 40% per year, in part due to the lower redemption of callables because yields have risen from last year. Growth in Federal Home Loan Banks and Federal Farm Credit, whose outstanding debt is about four-fifths of Fannie and Freddie’s, can help offset some, but not all, of the falling GSE issuance. $SSA new issuance is also down slightly in 2014 after sovereigns and supranationals issued 35% more debt in 2013 than the previous year. The two charts above show new issuance per month for 2014 vs. 2013, and also the yoy change in issuance against the three-month change in agency spreads3. There is an observable relationship between the decline in supply and the compression of spreads.

3 Agency spread is for a theoretical 3y par agency bond over a 3y par Treasury from Deutsche Bank spline models.

Steven Zeng, CFA


6 June 2014



While GSE and $SSA supply remains tight, which should be supportive of spreads, the pool of investable US fixed income securities will expand drastically after QE ends later this year. In a recent note, we wrote that the duration-adjusted net supply of Treasuries and mortgages available to private investors has turned positive in the current quarter and is expected to grow in 2015 and 2016. If the demand for agencies leak over to alternative fixed income products, especially those offering higher risk-adjusted returns, it could pressure spreads wider.

Johnson-Crapo is bullish for spreads, but (when) will it pass? Agency spreads have tightened considerably since the March introduction of the Johnson-Crapo GSE reform bill. The spread curve is also now at its flattest level in three years. The difference between 2y spreads and 5y spreads dropped to just 9.5bp this week, giving investors little incentive to extend out the maturity curve. The reasons agency debt investors like the Johnson-Crapo bill are clear: if made into law, the bill would require Fannie and Freddie to exit the mortgage financing business in five years, further accelerating their wind down and scaling back their issuance. After five years, outstanding Fannie and Freddie debt will also gain the government’s explicit guarantee, thus removing all their credit risk. Whether these securities will continue to publicly trade is unclear; one possibility is that the GSEs will try to buy back as much of their debt as possible in the open market, possibly paying a slight premium, just prior to their dissolution. At which point, it is not unreasonable to see Fannie and Freddie debt trade in a low single digit spread to or even through Treasuries, depending on how the government decides to deal with Fannie and Freddie’s remaining obligations. This could tighten the 5y sector another 5-8bp and 10y+ sector possibly as much as 30bp.

Difference between 2y spread and 5y spread dropped to

the lowest in three years

Agency spread reaction to Fannie/Freddie news

0

5

10

15

20

25

30

Jun-11 Dec-11 Jun-12 Dec-12 Jun-13 Dec-13 Jun-14

bp5y spread minus 2y spread

0

5

10

15

20

25

30

35

40

Jan-13 Apr-13 Jul-13 Oct-13 Jan-14 Apr-14

bp

5y par agency spread

GSEs reports record profits, investors bid up common shares and files lawsuits seeking to reverse the net-worth sweep agreement.

Investment fund Fairholme proposes in a presentation to take Fannie and Freddie private, but makes no mention of treatment of existing bondholders.

Johnson-Crapo bill introduced, proposes to eliminate the GSEs within five years

FHFA director Watt speaks

about preserving the GSEs' role

Source: Deutsche Bank Source: Deutsche Bank

All this sounds hunky-dory for agency investors, except the Johnson-Crapo bill at the moment is nowhere close to becoming law. The odds of it getting through both chambers of Congress this year is low, according to industry analysts. If the bill doesn’t maintain the momentum needed to get the debates going between parties and the modifications to assuage its opponents, it could go the way of the other forgotten GSE reform proposals. Moreover, if Fannie and Freddie prove to be successful businesses under the conservatorship statute, Congress could lose the will or the incentive to push for a reform. It’s interesting to see the remarks given last month by FHFA Director Mel Watt had widened spreads (see chart on right), even though his message was largely about preserving Fannie and Freddie’s footprint and continuing to reduce their risk. The reaction in spreads could very well be investors expressing displeasure at extending the GSE status quo. It leaves open the possibility of Fannie and Freddie eventually exiting the conservatorship or being allowed to releverage, and these are risk events that would no doubt require a much higher risk premium than what’s currently being priced.

6 June 2014



United States

Credit Securitization

Mortgages

Originally published on June 4 in The Outlook in MBS and Securitized Products.

MBS reinvestment: basis positive

The Fed has gone back and forth over the last few years on the best way to get out of its $4.37 trillion in securities, and the latest thinking once again is MBS friendly. Vice Chair Bill Dudley’s remark a few weeks ago dropped the first hint. Raise rates first, he suggested, and deal with the portfolio later. That way if the market has a problem with tighter money, the Fed will have only one piece of the policy machine in motion. Fixing things becomes much easier.

This is the fourth time that the Fed has thought out loud about dealing with MBS, and each time the news for the long side gets better. In June 2011 the Fed said that it would stop reinvestment, raise rates and then sell MBS. In June 2013, it backpedaled on selling. Since then it has suggested that it would stop reinvestment and raise rates at roughly the same time. Dudley’s idea now to raise rates first and stop reinvestment later would revise the plan again.

For MBS investors, Dudley’s revision would extend the Fed influence on the market and strengthen the case for adding to MBS positions. By the time the Fed finishes taper, it should hold $1.75 trillion in MBS or nearly 34% of the total outstanding fixed-rate market. If the Fed then treads water, that means at least a couple of important things:

More stable spreads on MBS

Less MBS deliverable for the dollar roll market

The combination is enough to push us from neutral to overweight the MBS-Treasury basis.

Impact on spreads Take the impact of the Fed on spreads. That impact depends either on changing balances in Fed hands or a changing balance of outstanding securities. And with balances in securities up less than $20 billion so far this year and very uncertain over the intermediate and longer term, the Fed has been all the action. Using a new way to predict supply in the short term—described in detail in Modeling monthly net and gross MBS supply elsewhere in these pages—net supply in June should come in light again. Fed demand still looks set to rule.

Diana Hancock and Wayne Passmore, researchers at the Fed, estimated in January that a 1-point rise in the Fed’s share of MBS tightens spreads by 2.29 bp. Fed purchases through the end of taper, all else equal, should tighten the basis from current levels by 3.5 bp. And the projected 34% share will represent a cumulative 78 bp of tightening from the time the Fed put the first $1 of MBS on its balance sheet.

If the Fed holds onto its MBS, those tighter spreads should persist—at least if supply cooperates. Any change in spreads, absent a shift in demand from

Steven Abrahams Research Analyst (+1) 212 250-3125 [email protected] Christopher Helwig Research Analyst (+1) 212 250-3033 [email protected] Ian Carow Research Analyst (+1) 212 250-9370 [email protected] Jeff Ryu Research Analyst (+1) 212 250-3984 [email protected]

Steve Abrahams


6 June 2014



another major buyer, would depend entirely on changes in outstanding supply. That makes the Fed a significant source of ballast for MBS. Supply moves slowly. Spread volatility should be relatively low, and much lower than if the Fed lets its MBS go free.

If the Fed ignores Dudley’s advice and allows amortization and prepayments in its MBS to flow back into the market, then, all else equal, spreads should go wider. Based on the Fed’s projected portfolio at the end of June 2015—after taper ends and at the point where rates might lift off—the Fed’s share of today’s outstanding market would start dropping by roughly one to two percentage points every quarter depending on prevailing mortgage rates (Figure 1). For example, holding today’s mortgage rates constant, the Fed MBS portfolio would drop by an estimated $77 billion each quarter and its share would drop by 1.5 points. According to the Fed’s model, MBS spreads each quarter would widen by 3.4 bp. The market would likely price that in advance, of course, widening spreads before Fed secondary supply actually shows up.

But before a falling Fed share could push spreads wider, something that might not come until 2016, there’s carry.

Figure 1: Projected quarterly Fed MBS run-off if reinvestment stops

Note: assumes Fed finishes taper and stops reinvestment after June 2015. Source: Deutsche Bank

Impact on carry Fed reinvestment would keeps MBS out of the dollar roll market since the Fed generally buys and holds its pools. And to see evidence of the Fed’s impact, just look at dollar roll financing rates. TBA 30-year 3.5% pools finance from June to July at 1-month LIBOR – 78 bp, and 4.0% pools at LIBOR – 51 bp. Those rates signal that the Street needs to deliver more pools than it can find in the natural floating supply of TBA pools. If that special financing persists for a year, it will be enough to offset 12 bp of widening in the 3.5% pass-throughs and 10 bp in the 4.0%. That calculation uses the pass-through spread duration to calculate the breakeven levels. That’s a reasonable cushion.

That carry is also enough to outstrip rates. Hedge-adjusted carry on 30-year MBS has improved substantially through most of this year largely because of the flattening yield curve and the falling impact of getting short swaps or other instruments that roll down the curve.

The carry in TBA pass-throughs and the indication from Dudley that the Fed will stay invested in MBS a lot longer than anticipated makes a potentially powerful combination. The Fed’s continuing purchases and likely steady

6 June 2014



balances after taper reduce although does not eliminate the risk of wider spreads. But the risk is muted. The carry from dealers financing scarce MBS at special rates creates a compounding advantage. On the expectation that Dudley speaks for a broadening constituency at the Fed, it’s time to move some capital off the sidelines of the MBS-Treasury game and into MBS.

Figure 2: Hedge-adjusted carry on TBA MBS

TBA

Price

Front $Roll (32s)

Carry (32s) net of:

Dur and Cnvx and Vega

FNCL 3.0 98-14 2 8:5 -0:5 -0:7 -0:1

FNCL 3.5 102-19 4 11:3 2:2 1:2 2:0

FNCL 4.0 105-20 4 11:3 2:6 0:6 1:2

FNCL 4.5 107-30 5 9:2 4:5 2:7 3:1

FNCL 5.0 110-15 4 9:7 8:0 6:7 6:6

FNCL 5.5 111-13 0 6:6 6:0 5:3 5:2

FNCI 2.5 101-13 0 8:2 2:5 2:3 2:4

FNCI 3.0 103-25 2 8:1 3:0 2:1 2:1

FNCI 3.5 105-20 4 7:6 3:4 2:5 2:5

FNCI 4.0 106-08 3 3:4 1:0 0:4 0:3

FNCI 4.5 106-07 2 3:0 1:0 0:7 0:7

FNCI 5.0 106-10 6 2:0 0:0 0:0 -0:0 Note: all market levels as of COB 06/02/2014. Source: Deutsche Bank

* * * The view in rates Falling real rates have accounted for most of the drop in US Treasury yields so far this year, a sign of continuing concern about global growth. Concerns about growth and inflation have clearly dragged down rates in the EU, making US yields look relatively attractive. Central bank flows into the US have followed. As long as EU rates remain low, it should be hard for US rates to rise too far or too quickly except in the unlikely event that US inflation gets out of hand. The US stands to enjoy low real rates for a while even though the economy is picking up—further fuel to a US rebound.

The view in spread markets The increasing likelihood of stable or slowly moving rates and the Fed tilt toward holding its MBS longer than expected adds to the return potential in spread product. The flattening yield curve also has lowered the cost of hedging MBS. Although surveys of MBS investors suggest that many are still underweight, that seems like an increasingly expensive prospect—one where an eventual widening of MBS spreads will have to be dramatic to overcome forgone carry. A decline in the Fed’s share of outstanding MBS could take a long time. We go to overweight on the MBS-Treasury basis this week.

NEW: overweight the MBS-Treasury basis

NEW: overweight 30-year 3.5%s and higher, underweight 3.0%s

Overweight 15-year pass-throughs against 30-year

Underweight Ginnie Mae/conventional

Overweight LLB 3.0%s and 3.5%s

Overweight conventional 30-year 5.0%s of 2010

The view in mortgage credit US home prices have gone up by 4.4% through the first four months of 2014, according to data released Tuesday by CoreLogic, and are on track to meet or exceed Deutsche Bank’s forecast of a 7.8% gain this year. Our most recent forecast anticipates a 5.2% gain in 2015 and a 3.6% gain in 2016.

6 June 2014



The rise in home prices this year comes despite weakness at times in new and existing home sales, although both series have rebounded in their most recent reports. Home prices historically have been relatively insensitive to changes in rates while residential construction has been among the most sensitive. The data so far this year continue to bear this out.

Modeling monthly net and gross MBS supply

Net supply has never been more important to MBS performance. And although predicting net supply over long horizons remains difficult, some simple methods may allow prediction with increased accuracy over short horizons. Those methods now suggest that net supply for May was around $11 billion and that June looks likely to come in as another light month.

The approach takes advantage of tremendous stability in originators' patterns of pooling loans and issuing new MBS. Originators can start pooling loans into MBS weeks before the monthly TBA settlement date and can continue pooling until nearly the last day of a month. The percentage of each month's total supply that gets pooled and issued in the run-up to regular settlement and afterwards follows a very predictable pattern (Figure 3).

Figure 3: FNMA 30-year daily issuance follows stable pattern in recent

months

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

-26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14

Cum

ulat

ive %

of t

otal

ME i

ssua

nce

Days before/after notification

Feb-14 Mar-14 Apr-14 May-14

Note: chart shows cumulative issuance as a percentage of eventual total month’s issuance on each business day relative to the class notification date, where 0 = the notification date. Source: Deutsche Bank, eMBS

As might be expected, the pattern differs between agencies and product types, but within each individual sector the pattern proves remarkably stable from month to month.

Within Fannie Mae 30-year fixed-rate MBS, a small percentage of each month's supply starts getting issued as early as 26 business days before notification day, the day where sellers in the TBA market have to disclose the pools they intend to deliver. So far into 2014, that accumulating percentage jumps to an average of 54% of the ultimate total by notification day. It rises slowly after notification day, and then, roughly a week or so before the end of the month, takes a final leap toward 100%.

The pattern makes sense given the operational details of MBS production. Originators hedge their loan pipeline against rate risk by selling TBA forward positions. So, for instance, mortgage rate offers made to borrowers at the beginning of May might be hedged by selling TBA contracts for June

Steven Abrahams Research Analyst (+1) 212 250-3125 [email protected] Ian Carow Research Analyst (+1) 212 250-9370 [email protected]

6 June 2014



settlement. On the June notification date, the originator will need to disclose the CUSIPs of the pools they intend to deliver into the TBA short. Class A notification day typically falls somewhere between the 8th to 11th day of the month. Allowing for the four to five days that Fannie Mae and Freddie Mac typically need to take in loans and create pools, the originator will need to bring loans to Fannie or Freddie within the first few business days of the month. And, in fact, we see a large surge in pool issuance, primarily in large single-issuer pools, in the first few business days of each month. These large issuers are the ones most likely to hedge origination with TBA contracts.

For the remainder of the month, issuance continues at a gradual pace, up until the last few business days, when a second large surge covers roughly the remaining 20% of the total. Looking at the pool-level detail, the majority of this final push seems to come from multiple issuer pools. One plausible explanation offered by industry experts is that this represents the GSEs cleaning up the remaining loans purchased from smaller lenders at the cash window. The GSEs can and do issue multi-issuer pools throughout the month. However, given the uncertainty around the timing of loan inflow, it seems reasonable that the GSEs would want to keep a few pools open and retain the option to close before month end.

Estimating monthly issuance in advance Based on the stable pattern discussed above, predicting month-end total issuance should be a fairly straightforward exercise of updating daily accumulated issuance and applying the relevant ratio given days to notification. Unfortunately, reality is never quite that simple. There is noise in daily reporting.

For example, looking back at historical data, as of May 1 eMBS reports that originators had created $3.5 billion in pools for May issuance. Since originators in February, March and April had created an average of 21.5% of the ultimate supply by the first day of those months, we can estimate that the May volume represents the same share. That predicts ultimate May gross supply of $3.5 billion/21.5% or $16.2 billion, compared to a final actual total of $21.1 billion. By May’s Class A notification date the estimate improved to $19.8 billion (Figure 4).

Figure 4: Simple model approximates total gross issuance as early as 10

business days to notification in FNMA 30-year for May

-

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

50,000

-17

-16

-15

-14

-13

-12

-11

-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Fina

l cum

gros

s iss

($M

)

Days before/after notification

Low Avg High Act

Note: chart shows the estimated total month-end gross issuance for May at each point in time based on the reported cumulative issuance available at that point in time and the simple average of cumulative percentage issuance patterns from the prior three months. Source: Deutsche Bank, eMBS

6 June 2014



However, that simple test of our methodology ignores some nuance to the reality of daily issuance reporting. In particular, certain pool types—such as Fannie Mae Mega, Freddie Mac Giants and multi-sellers, and crucially, GNMAII multi-issuer—may be subject to upward revision throughout the month. The pool may be created and reported early on in the cycle, but continue to grow throughout the month. The pool may not close until the end of the month, but the report date will not change with added issuance.

While originators’ issuance patterns appear remarkably stable, daily reporting intra-month may contain a fair degree of noise. So at the end of May we can say with confidence that $3.5 billion was issued in pools reported by May 1. But on May 1 we may have had a much lower estimate.

This is a problem, but one that time should be able to heal. As we progress further into the issuance month, we would expect the variance from daily reporting noise to diminish. So for the time being we may not be able to project month end issuance with a high degree of confidence 10 days prior to notification, but confidence should still grow as we approach and pass notification. For a market hanging on the details of supply, this certainly is an improvement over waiting until the end of the month. And as we continue to track issuance on a daily basis, we will be able to further refine our model to adjust for daily noise in reporting. That should improve confidence in early month estimates in the future.

Bringing it all together We also looked at monthly issuance patterns in 2014 for Freddie 30-year, GNMA I 30-year, GNMA II 30-year, Fannie 15-year and Freddie 15-year fixed-rate MBS, and were able to develop similar simple models for estimating gross supply in each of those sectors. Combined, those products represent the vast majority of agency MBS issuance in a given month and should provide a fairly clear picture of total gross issuance as we move further into the month. Daily noise in reporting remains an issue in these other sectors as well, but we are equally optimistic that we will be able to improve estimation quality with time.

To get net supply, we also need to estimate the amount of gross issuance that represents simple amortization and prepayments. Any good prepayment model can predict that with reasonable accuracy, especially for the 30-year market overall. For May, our models predict $61 billion in combined principal pay-downs. That leaves our net at $71.5 billion gross minus $61 billion, or $10.7 billion of net supply.

Estimating individual sector gross issuance and pay downs we will also be able to compare relative supply/demand technicals between different agency MBS products (Figure 5).

Conclusion Instead of relying on macro models of new supply—ones that consider the broader economy, housing, foreclosures, seasonality and the like—investors can rely on a simpler approach that allows day-to-day updates with steadily improving accuracy. Since Fed demand has turned out to be very predictable this year and net supply has become the wildcard, cumulative pooling models can help investors see that wildcard further in advance.

Figure 5: Sector net supply for May

May Gross iss $bn Pay down est Net $bn

FN30 21.1 -20.2 0.9

GD30 21.3 -12.3 9.0

FN15 4.7 -7.0 -2.3

GD15 4.3 -4.3 0.0

GN1 0.5 -5.6 -5.1

GN2 19.6 -11.4 8.2

Total 71.5 -61.0 10.7 Source: Deutsche Bank, eMBS

6 June 2014



Observing all-cash home sale share

As the supply of agency MBS continues to remain light, MBS investors are digging through different factors that drive the supply to understand the phenomenon. One factor that influences the mortgage supply is all-cash home purchases. If all-cash buying share decreases, more mortgages will be issued even as total home sales are unchanged, and vice versa. A comparison of all-cash home transaction data from a few vendors shows different trends and levels, which are discussed below. Looking forward, we expect all-cash sale shares to trend lower from last year’s levels due to lower distressed inventory and pick-up in home purchase activity, which should be a positive to agency MBS supply.

The all-cash home sale share should trend lower Trends in all-cash home sales give more information to MBS investors about the potential supply of mortgages than projecting total home sales alone, thus its importance. Figure 6 compares the all-cash home purchase share data series for Campbell, RealtyTrac, National Association of Realtors (NAR) and CoreLogic since 2011. CoreLogic series shows year-over-year decreases in all-cash shares of home sales with seasonality. Campbell survey data also shows a similar pattern, although at a lower levels. NAR data shows a slight year-over-year increase during the first quarter of the year. Meanwhile, RealtyTrac data shows a much larger year-over-year gain. The pattern from the RealtyTrac data is clearly different, but many analysts gauging cash-share transactions are skeptical of this trend, especially given that three other major data sources show nothing like it.

Figure 6: All-cash home sale shares data for various vendors show different

trends and levels

0

10

20

30

40

50


All-

cash

sha

re (%

)

Campbell RealtyTrac (Qtrly) NAR CoreLogicSource: Deutsche Bank, Campbell, RealtyTrac, NAR, CoreLogic

According to Sam Khater, deputy chief economist for CoreLogic, one of the main reasons for the decrease in the all-cash sales in their data is the decline in distressed home sales. This is because nearly 60% of distressed sales since 2013 have been for all-cash, a much higher ratio than other types of sales. In comparison, around 35% of re-sales and short sales were for all-cash and 15% of new construction sales were for those without financing. During February 2014, CoreLogic data shows that there were around 30,000 distressed home sales, or around one-third less than the previous year (Figure 7). Assuming a 60% all-cash sale ratio, this implies that all-cash sales decreased by 9,000 just from the decline in distressed sales, or around 3.5% of total sales.

Jeff Ryu Research Analyst (+1) 212 250-3984 [email protected]

6 June 2014



Looking forward, the all-cash home sale share should decrease as distressed inventories of mortgages continue to shrink as it has been since 2010 (Figure 8). In addition, purchase activity, which has a lower ratio of all-cash transactions, has been fairly weak to start the year. As home purchase activity picks up in the coming months, this should also put downward pressure on the all-cash home sale share.

Calculation methodology impacts levels The latest data points for various vendors show different all-cash share levels. Using the February data points, the Campbell survey and NAR survey shows that all-cash share lies between 30% and 35%, while RealtyTrac and CoreLogic data are between 40% and 43%. The difference in the calculation methodology looks to explain the majority of these variations. Campbell and NAR data come from the survey of real estate agents, while RealtyTrac and CoreLogic data are computed by matching sales deeds and mortgage data (Figure 9). This means that the Campbell and NAR survey will miss sales that do not go through real estate agents. Some of these include bulk sales, institutional sales and lower priced home sales, which tend to have higher all-cash shares and, hence, surveys of real estate agents show lower all-cash transaction share.

Figure 9: Comparison of all-cash data series

Campbell/

IMF HousingPulse RealtyTrac CoreLogic NAR

Percent of market covered

Around 2,000 real estate agents

Around 80% of the home sale

Around 90%+ of the home sale

Around 3,000 real estate agents

Methodology Survey of real estate agents

Recorded sales deeds and loan

data

Recorded sales deeds and loan

data

Survey of real estate agents

Feb 2014 data point 30.4 42.7 (Q1 2014) 40.2 35.0

Feb 2013 data point 32.5 19.1 (Q1 2013) 43.7 33.0

YoY change -2.1 23.6 -3.5 2.0 Source: Deutsche Bank, Campbell, RealtyTrac, NAR, CoreLogic

Seasonality factor in all-cash sale share As mentioned above, the all-cash sale share shows a seasonality factor, peaking in January and bottoming out in June. For example, the all-cash sale share in January averaged 110% of the year’s average from 2000 to 2013, while it was 93% during June (Figure 10). This pattern comes from the fact that mortgage sales show a stronger seasonality than all-cash sales. Thus,

Figure 7: Distressed home sales has steadily gone down over the years

0

100

200

300

400

500

0

20

40

60

80

100


Sale

s (00

0s)

sale

s (00

0s)

Distressed sale Short sale New constr

Others Resale (RHS)

Source: Deutsche Bank, CoreLogic

Figure 8: Distressed inventory

continues to shrink

0

200

400

600

800

1000

1200

1400

2009

Q1

2009

Q3

2010

Q1

2010

Q3

2011

Q1

2011

Q3

2012

Q1

2012

Q3

2013

Q1

2013

Q3

Seri

ous

dq m

tg o

utst

($bn

)

Note: amount outstanding of 60+ day delinquent, foreclosed and REO mortgages Source: Deutsche Bank, CoreLogic

6 June 2014



mortgage sales tend to increase more during the summer than all-cash sales, but also decrease more during the winter.

Figure 10: All-cash sale share show a seasonality factor

80%

90%

100%

110%

120%

130%

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

Seas

on

alit

y fa

cto

r

2000 to 2013 AverageSource: Deutsche Bank, CoreLogic

Repricing MBS: the hidden convexity in OAS

The standard drill for predicting returns in MBS involves projecting cash flows over some future scenario and then figuring out where the security might price. That price usually comes from assuming that today’s spread—or, more precisely, its option-adjusted spread—predicts the future’s. It’s easy to guess, that with rare exception, that has to be wrong. OAS typically changes a lot as rates move and as uncertainty about prepayments shifts. Once an investor takes likely OAS change into account, projected MBS risk and return can change in surprising ways.

A straightforward exercise with IOS, the part of the market arguably most sensitive to prepayment risk, suggests that these instruments in practice can have greater negative duration and much less negative convexity than projected in traditional analysis (Figure 11). One implication then is that a much smaller amount of IOS can potentially be used to reduce a mortgage portfolio’s duration and that it will add much less negative convexity to the position. Another is that investors should consider hedge ratios carefully against the probability of large swings in rates.

Steven Abrahams Research Analyst (+1) 212 250-3125 [email protected] Ian Carow Research Analyst (+1) 212 250-9370 [email protected]

6 June 2014



Figure 11: OAS re-pricing increases negative duration, adds convexity

-50

-40

-30

-20

-10

0

10

20

30

-100 -75 -50 -25 0 25 50 75 100

1Y T

RR %

Gradual parallel curve shift (bp)

IFN34010 Mdl OAS IFN34010 OAS Const

Note: chart shows 1-year projected total returns for gradual, parallel shifts in the yield curve with reinvestment at 1-month LIBOR. “OAS Const” returns calculated holding OAS constant to the horizon. “Mdl OAS” returns calculated by gradually shifting spot OAS to the OAS projected at the horizon in our model given a set of horizon values fitting the scenario. All levels indicative, as of COB: 5/27/2014. Source: Deutsche Bank, YieldBook

Compensating for prepayment risk Spread in MBS and in any asset outside of Treasury debt compensates for risk. The key risk in agency MBS and the most plausible explanation for positive OAS is prepayment risk. Prepayment risk changes as rates and volatility move around. It also changes over time as origination and servicing conditions change. In general, refinanceable MBS trade at wider OAS than MBS subject mainly to housing turnover. And moderately seasoned, more negatively convex MBS trade at a wider OAS than newer securities.

It’s worth focusing on some of the details.

Prepayment risk changes as rates move primarily because loans and their opportunity to refinance slide from out-of-the-money to in-the-money. Borrowers without refinancing incentives tend to exhibit fairly low variance in prepayment speed. This stands to reason as the primary non-refinance drivers of prepayments—turnover and default—pose fairly substantial barriers to the borrower.

Borrowers with refinance incentive can vary dramatically in prepayment speed. In contrast to moving or losing the home, a borrower with rate incentive should theoretically have very low barriers to prepayment. However, in practice the transmission from rate refinance incentive to prepayment is far less straightforward. Indeed, one need look no further than the multiple rounds of FHFA targeted HARP marketing to see that borrowers do not always react to refi incentive in the rational, efficient manner that one would expect.

The underlying reasons for borrower refinance inefficiency can be many and varied. Some, like current FICO, LTV and geographical differences can be established over time and built into prepayment models. Others, like a borrower’s lack of information, may be extremely difficult to quantify or capture.

It consequently is easier to predict prepayments in out-of-the-money MBS and harder with in-the-money MBS. That is, model risk rises quickly as the loans backing an MBS become refinanceable.

6 June 2014



Prepayment risk also gets leveraged or deleveraged by the pricing and structure of a security. Returns on a security trading well above or well below par become progressively more sensitive to small differences in prepayments. Interest-only or principal-only securities are extreme versions of this. But structural features that stabilize cash flows, such as PAC protection, or destabilize cash flow, such as companion classes, also change risk.

Prepayment risk also changes over time as origination and servicing conditions change. Mortgage industry concentration and the mix between bank and non-bank servicers are examples of structural changes that may influence lender aggressiveness in refinancing. The adoption and interpretation of new regulatory frameworks also adds an additional layer of uncertainty to understanding lender behavior.

The difficulty of accurately capturing prepayment risk leads to some simple predictions:

OAS should widen as prepayment risk rises

OAS should tighten as risk falls

OAS should change over time with market views of prepayment dynamics

Test case: IOS In an effort to better approximate actual return performance, we set out to develop a simple model for projecting OAS based on refinance incentive. For a number of reasons, the IOS market is a good place to start.

IOS reference a broad and consistent set of collateral. The structure does not change over time as is the case with CMO. The market provides daily pricing and weekly indications of liquidity. And, critically, the dominant risk in IOS is levered prepayment exposure. All of these factors make IOS a good candidate for modeling OAS sensitivity against prepayment risk.

Indeed, a simple first approach plotting IOS OAS against refinance incentive suggested a strong relationship between the two on a number of trading days (Figure 12).

Figure 12: Strong relationship between select IOS OAS and refinance

incentive

IFN33012

IFN33512

IFN33510IFN34011

IFN34010

IFN34009

IFN34511

IFN34510

IFN34509

IFN35010

IFN35009

-100

-50

0

50

100

150

200

250

300

-1 -0.5 0 0.5 1 1.5

OAS (

bp)

Refi Incentive (%)

Note: chart shows each IOS contract OAS plotted against refinance incentive where refinance incentive is equal to IOS gross WAC less the reported primary market rate for that day. Scatter plot for those values as of Jan 24, 2014. Source: Deutsche Bank, Bankrate.com

6 June 2014



Our sample period ran from late September 2012 through the beginning of May 2014. On some trading days the positive relationship between refinance incentive was quite clear; on others, less so. That suggested the need for a more refined analytical approach incorporating additional variables.

Building a model Determining what those independent variables should be requires both strong a priori rationale and then some experimentation. Some variables that should clearly make the cut:

Refinance incentive

WALA has a clear effect in the scatter plots; for the same refi incentive, even in lower coupons, the older vintage trades wider

Par coupon 30-year OAS should help control for generally wider or tighter spreads in MBS as a whole

An interaction variable between refi incentive and WALA

However, simple refi incentive likely does not provide a complete picture of prepayment risk. In our view, some additional rate variables can serve as proxies for other dimensions of prepayment risk:

The shape of the yield curve may provide information on the attractiveness of refinance product alternatives and primary-secondary spread, amongst other factors

Implied 10-year rate volatility should speak to general variability in prepayments and thus the degree of uncertainty between periods

As a final layer of refinement, we removed IOS coupons 5.5% and higher, and 5.0% vintages 2008 and older, due to infrequent transactions and therefore materially lower quality of pricing (Figure 13). The values are highly significant, and directionally the coefficients make empirical sense. In particular, the model predicts that OAS should increase with refinance incentive, and that effect may be amplified by seasoning.

Figure 13: Simple OAS model regression results

Variable Coefficients Standard error t Stat P-value

Intercept -624.0 65.5 -9.53 0.0%

RefiInc 22.7 9.9 2.30 2.1%

WALA 6.2 0.2 24.75 0.0%

RefixWALA 3.8 0.3 13.76 0.0%

30Y CC OAS 14.0 0.4 39.33 0.0%

3m10Y Vol 0.7 0.3 2.38 1.7%

5Y10Y Vol 12.2 0.9 13.83 0.0%

2s10s -475.4 11.0 -43.40 0.0% Source: Deutsche Bank

The model explains roughly 67% of the OAS level, and projected OAS historically tracks actual levels fairly closely, particularly for the most liquid contracts (Figure 14).

6 June 2014



Figure 14: IOS 4.0%s 2010 projected OAS tracks actual levels fairly closely

-200-100

0100200300400500600700800

9/25

/201

2

10/2

5/20

12

11/2

5/20

12

12/2

5/20

12

1/25

/201

3

2/25

/201

3

3/25

/201

3

4/25

/201

3

5/25

/201

3

6/25

/201

3

7/25

/201

3

8/25

/201

3

9/25

/201

3

10/2

5/20

13

11/2

5/20

13

12/2

5/20

13

1/25

/201

4

2/25

/201

4

3/25

/201

4

4/25

/201

4

OAS

(bp)

Proj Act

Source: Deutsche Bank, Bloomberg Finance LP

Repricing using modeled OAS Given a simple model for projecting IOS OAS, we can incorporate OAS repricing into the standard total return analysis. For the most liquid IOS contracts, we calculated 1-year total returns using the standard OAS constant assumption for gradual parallel curve shifts up and down 100 bp in 25 bp increments. We then ran the same analysis shifting OAS to the horizon based on our model projections (Figure 15 and Figure 16).

Figure 15: IOS 4.5% 2009 scenario returns Figure 16: IOS 5.0% 2010 scenario returns

-50

-40

-30

-20

-10

0

10

20

30

40

-100 -75 -50 -25 0 25 50 75 100

1Y T

RR

%



-40

-30

-20

-10

0

10

20

30

-100 -75 -50 -25 0 25 50 75 100

1Y T

RR

%



Note: Chart shows 1-year projected total returns for gradual, parallel shifts in the yield curve with reinvestment at 1-month LIBOR. “OAS Const” returns calculated holding OAS constant to the horizon. “Mdl OAS” returns calculated by gradually shifting spot OAS to the horizon by the OAS projected in our model given a set of horizon values fitting the scenario. All levels indicative, as of COB: 5/27/2014. Source: Deutsche Bank, YieldBook

Note: Chart shows 1-year projected total returns for gradual, parallel shifts in the yield curve with reinvestment at 1-month LIBOR. “OAS Const” returns calculated holding OAS constant to the horizon. “Mdl OAS” returns calculated by gradually shifting spot OAS to the horizon by the OAS projected in our model given a set of horizon values fitting the scenario. All levels indicative, as of COB: 5/27/2014. Source: Deutsche Bank, YieldBook

In contrast to the typical conception of IOS returns, the return profiles with the modeled OAS shift display increased negative duration, and reduced negative convexity—and in some cases even positive convexity.

This outcome appears consistent with our theoretical framework and model results. As rates rise and refinance incentive falls, OAS should tighten—enhancing returns on an already negative duration asset. Conversely, as rates fall and the IOS move in-the-money, prepayment model risk should increase

6 June 2014



and OAS along with it—exacerbating losses. This change in return profile is consistently exhibited for the most liquid contracts from the 3.5%s of 2010 through the 5.0%s of 2009. 4

Implications The idea that mortgage assets may perform differently in practice than model predictions is certainly not a foreign concept to experienced market participants. Mortgage traders regularly assign hedge ratios to securities that differ from the model expectation based on their own observed experience and market views. Realized negative convexity may also differ materially from model projections for reasons unrelated to OAS repricing—current benign prepayment levels in theoretically at-the-money coupons being one example.

The reality is that mortgage prepayment models serve as blunt instruments that aid in simplifying evaluation of an otherwise very complex investment decision. Recognition of model limitations and regular adjustment to realized market conditions is a critical function of any active mortgage investment strategy.

In this particular case, one possible implication is that IOS—and perhaps structured IO to some extent—may prove more useful than expected in hedging a traditional mortgage portfolio. The instruments bring longer negative duration and less negative convexity to the position that constant-OAS analysis suggests.

Where a further rally might boost prepayments most

The rally in the 10-year Treasury notes over the past two months lowered primary mortgage rates by nearly 40 bp in our own weekly survey to 4.19% last week raising fears or hopes—depending on your viewpoint—of a renewed burst of refinancing activity. Cohorts with the greatest chances of higher refinance activity share two characteristics (assuming all other characteristics are similar):

Greater concentration of borrowers with higher-note rates relative to the coupon rate; and

Greater concentration of mortgages originated by brokers or correspondents.

Interest rate incentives clearly motivate borrowers to refinance to lower their payment or shorten their mortgage term while brokers and correspondents are motivated to refinance their customers to earn origination fees.

Securities with a 4.0% coupon rate are clearly in the cross-hairs for refinancing activity. Tuesday’s back-up in the Treasury rates took a little wind out of the refinancing sail but 40% of the mortgages backing Freddie Mac 4.0% coupons have nearly 50 bp refinance incentive; an additional 10 bp drop to cross that threshold is quite possible in the short run. That percentage represents nearly

4 It is important to note, however, that our data set included very few observations where refinance incentive approached or was less than negative 100 bp, and none where the IOS was out of the money by more than 108 bp. In the case of IOS 3.5%s of 2010 for the rates up 100 bp scenario the contract is estimated to be 135 bp out of the money. The OAS model results for the scenario are reasonably subject to a high degree of doubt as the model is extrapolating beyond observed parameters.

Doug Bendt Research Analyst (+1) 212 250-5442 [email protected] Jeff Ryu Research Analyst (+1) 212 250-3984 [email protected]

6 June 2014



$80 billion of outstanding volume; the similar portion of Fannie Mae’s outstanding balance likely would bring the total amount outstanding to about $200 billion, enough to make a substantial difference in the supply-demand balance and possibly affect the mortgage basis.

Note rate distribution The distribution of note rates is quite different among coupon/vintage cohorts. In the newest vintages—2013 and 2014—25% to 30% of the borrowers have notes rates above 4.625% compared with only 10% to 20% for other vintages (Figure 17). Older vintages have had greater exposure to lower rates previously so higher-rate borrowers had more chances to refi, so any remaining higher-note rate borrowers are more likely to be experiencing burnout.

Originator channel The concentration of retail borrowers, however, is fairly constant across cohorts with the exception of a much higher concentration of retail borrowers in the 2012 cohort (Figure 18). Note that in 2012 the volume of issuance for 3.5% coupons was four times higher than for 4.0% coupons; the retail share for the 3.5% 2012 cohort was 55%, right in line with the other vintages. Nonetheless, the 4.0% 2012 is likely to prepay slower in the event of any rate dip and as the cohort ages because of this higher concentration of retail-originated loans.

Figure 17: Newer vintages have greater share of higher

note-rate mortgages

Figure 18: 2012 vintages has a greater share of retail-

originated mortgages

Freddie Mac 4.0% 30-year fixed-rate securities with mortgage rates above 4.625%; excludes U6/U9

0%

5%

10%

15%

20%

25%

30%

2009 2010 2011 2012 2013 2014

Freddie Mac 4.0% 30-year fixed-rate securities; excludes U6/U9

Source: Deutsche Bank; Freddie Mac loan-level data via 1010data Source: Deutsche Bank; Freddie Mac loan-level data via 1010data

Conclusion Other factors and mortgage characteristics certainly will affect relative prepayment rates between cohorts. However, rate incentive and origination channel are two of the strongest influences. Thus, a lower prepayment response in percentage terms by the 2012 vintage relative to the 2013/2014 vintages seems likely.

6 June 2014



Europe

Credit HY Strategy

Credit Strategy - HY Credit Still Cheap?

When looking at HY credit spreads in 2014, we have spent a lot of time highlighting the fact that although yields may be at the extremes of history credit spreads are not yet at such extreme levels and given that we continue to operate in an extraordinarily low (and probably artificially created) default environment that HY credit spreads are not too expensive at current levels.

In this article, we look to view HY credit spreads in a slightly different way to arguably highlight that they still provide value to investors at current levels. We look at spreads as a percentage of the overall HY bond yield. We show that we may not be at the highs of history but in the case of EUR HY we are very close and that from a historical perspective we have tended to see spread tightening in the months following spreads as a percentage of yield at higher levels.

Figure 1: EUR Non-Fin HY vs. USD HY Spread as a Percentage of Yield

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2003 2005 2007 2009 2011 2013

EUR USD


Spreads as a percentage of yield – HY credit still cheap?

We start by looking at the USD market as we can track the data back to the late 1980s. In Figure 2 we show the spread as a percentage of overall HY bond yield back to 1988. We can see that with overall yields so low spreads make a large proportion of the overall bond yield however it has dropped by around 20pp over the last couple of years as spreads have continued their grind tighter and underlying Treasury yields have generally been on the rise. In Figure 3 we look at a scatter plot of spread as a percentage of yield against the spread change over the subsequent six months as a percentage of the starting yield. There doesn’t appear to be an undisputable relationship but it does seem that the best of the spread performance is seen when spreads are at a higher percentage of overall yield and perhaps more importantly we see spread widening more often when spread as a percentage of yield is at a lower level. We should point out though that some of the most notable spread widening has also occurred when spread as a percentage of yield was not much lower than current levels. The current level is around 70% and there have been 70 previous occasions when a six month period has started with spread as a percentage of yield around current levels or higher and on 61 of these

Nick Burns, CFA


6 June 2014



occasions (87%) spreads have tightened over the next six months. This compares to the overall number of around 50% based on all observations.

Figure 2: USD HY Spread as a Percentage of Yield

Figure 3: USD HY Spread as % of Yield vs. Subsequent

6M Spread Change as % of Starting Yield

0%10%20%30%40%50%60%70%80%90%

100%

1988 1991 1994 1998 2001 2004 2008 2011

Spread as % Yield

-60%-40%-20%

0%20%40%60%80%

100%120%

20% 40% 60% 80% 100%

Su

bse

qu

ent

6M

Sp

read

C

han

ge

as %

of

Sta

rtin

g Y

ield

Spread as % Yield

Current level


So it does seem that with spreads as a percentage of yield at current levels there’s a good chance (based on historic observations) that spreads will be tighter over the next six months. In Figure 4 we show the results for the same analysis based on BBs and single-Bs with the outcome very similar to the overall HY analysis.

Figure 4: USD HY Spread as a Percentage of Yield and Subsequent 6 Month

Spread Change When at Current Level or Higher

Overall % of Spread Tightening

Current Spread as % of Yield Observations

Tightening over Next 6 Months As a %

All HY 70% 70 61 87% 50%

BB 61% 71 61 86% 53%

B 70% 68 56 82% 52%Source: Deutsche Bank

We now do the same analysis on the EUR market although clearly we have far less history to look at (only back to 2003). In Figure 5 we show spread as a percentage of yield for EUR non-financial HY along with the same for USD HY through time. We can see that the general trend has been very similar. That said the one point worth highlighting is that the current levels for EUR HY are higher than for USD HY this in many ways probably reflects the difference in underlying government bond yield environments. Although government yields for all core markets had fallen to their lows for the year in May, 5 and 10 year Bunds fell to around 15bps above the lows of the past couple of years while in the US 5 and 10 year Treasury yields were still around 100bps above their 2012 lows.

6 June 2014



Figure 5: EUR Non-Fin HY vs. USD HY Spread as a

Percentage of Yield

Figure 6: EUR Non-Fin HY Spread as of % Yield vs.

Subsequent 6M Spread Change as % of Starting Yield

0%10%20%30%40%50%60%70%80%90%

100%

2003 2005 2007 2009 2011 2013

EUR USD

-100%

-50%

0%

50%

100%

150%

20% 40% 60% 80% 100%

Su

bse

qu

ent

6M

Sp

read

C

han

ge

as %

of

Sta

rtin

g Y

ield

Spread as % Yield

Current level


If we now turn our attention to Figure 6 we can once again see a similar dynamic as we saw with the USD market whereby the most significant spread tightening has followed when spreads as a percentage of yield have been at higher levels while at lower levels we have more frequently seen spread widening. Specifically within our analysis there has been 23 months where the spread as a percentage of yield has been at the current level or higher and on 21 (91%) of those occasions spreads have been tighter six months later this compares to a rate of around 63% for all observations.

So again the general trend does seem to be one of an increased likelihood of spread tightening in the months following particularly high levels (>80%) of spread as a percentage of yield. In Figure 7 we provide the results for the same analysis looking at BBs and single-Bs and again we see that the outcome is very similar to the overall HY analysis.

Figure 7: EUR Non-Financial HY Spread as a Percentage of Yield and

Subsequent 6 Month Spread Change When at Current Level or Higher

Overall % of Spread Tightening Current Spread

as % of Yield ObservationsTightening over Next 6 Months As a %

All HY 86% 23 21 91% 63%

BB 83% 20 19 95% 63%

B 90% 16 15 94% 62%Source: Deutsche Bank

With spreads as a percentage of yield at historically high levels, particularly in the EUR market, this does seem to provide a backdrop for potential further spread tightening if the historic relationship is to be any guide of what we might expect going forward. One drawback of this analysis for the EUR market is the lack of history and the fact that most of the occasions where spreads have been at such high percentages of yield have occurred in the last few years when the credit market has benefitted from the support of very accommodative central banks.

On the other hand, it can also be argued that with government yields so low there is greater need to pick-up the extra yield on offer in corporate bonds. That potentially helps to keep demand strong for HY credit and maintain the beneficial technical dynamics that continue to be such a big feature of credit market performance. So if we continue to operate in a world of low government bond yields with low volatility then HY credit should continue to perform. If this changes (i.e. we see a material rise in government bond yields) then the immediate outcome may be less certain and the performance of credit would likely be determined by the cause of any change.

6 June 2014



Europe

Credit Covered Bonds Securitisation

European ABS update

The c.70 bps cover on the Cordusio 4 A2 on what was a postponed BWIC highlights the ongoing strength of the bid for peripheral ABS. The trade in question represents a post-crisis tight for Italian RMBS seniors which last traded at similar levels in January 2008. Part of the recent run up was also in anticipation of some form of ECB easing on Thursday, which did not disappoint with a EUR 400 billion targeted LTRO announced and preparatory work for future ABS purchases confirmed (see our Special Report on the same). Interest in Italian ABS has been sustained this year, with some 7 retained tranches coming to market including the latest CLAAB 2011-1 A (3.6 year WAL class A @ 6mE+125 bps). Applying the “it never rains until it pours” maxim to Italian new issue, prospects remain strong with the next transaction being marketed being a large EUR 1.4 billion consumer ABS (Sunrise 2014) transaction.

While interest in peripheral ABS has been a salient feature, core ABS too has not been left too far behind – in the secondary, wider UK Prime such as GRANM BBBs are higher by 1 ppt in the last 2 weeks. Meanwhile, as Dutch RMBS Storm 2014-II is being marketed in the primary, leveraged loan CLO issuance has been coming through at a steady rate with the class A1 from the recent CGMSE 2014-2 pricing at 135 bps, and four deals (GSO/ Blackstone’s Phoenix Park, 3i’s Harvest IX, Apollo Management’s ALME Loan Funding II, as well as Pramerica’s Dryden 32) expected to amount to EUR 1.6 billion already in the pipeline.

A number of recent sales of assets in the Mozart loan (Total Sales Price: EUR 17.6mn, individual asset sales at 11.2%, 13.8% and 25.9% below valuation) announced over the past few weeks re-iterates to our mind our earlier expressed caution (see "Thinking some more about Talisman 7") about the Classes D and E in Talisman 7. In our view at the Class D level, much of the ultimate recovery will be linked to how much (if any of significance) of the EUR 66mn cash at the Mozart borrower level will be used to redeem principal on the loan. The disbursement of this cash is especially complex, particularly with regard to capital gains tax provisions, and it will be of keen interest to us if further information is disclosed in the Investor Update presentation from Hatfield Philips to take place Friday of next week.

Chart of the week – Target market size of potential ECB purchases

0

50

100

150

200

250

EU

R B

n

Placed Retained

Source: Deutsche Bank estimates, Assuming ECB repo-eligibility criteria will be applicable for purchases–Euro area BBB- rated and above senior bonds

Other articles in the weekly include RMBS/ABS Notice board: - BOE/ECB joint document on securitisation

Conor O'Toole


6 June 2014



Global

Credit Covered Bonds

Covered Bond and Agency Update

With six new issues, the primary market of EUR benchmark covered bonds was very active before the ECB decision. New issues were well bid and tightened further in the secondary market.

Given the pricing of Santander Totta at ms+93bp, Caixa Geral Jan 2019 looked very attractive, providing a pick-up of around 20bp at the time of pricing of new Totta.

While supply picked up this week, bringing ytd issuance to EUR 59bn, with EUR 20bn of EUR benchmark covered bond redemptions in June alone and EUR 90bn for the remainder of the year, the trend of high negative net supply in FY 2014 remains intact.

With the ECB introducing a 4Y targeted LTRO providing unconditional bank funding for 2Y at a 25bp fix rate, covered bond spreads seem very well supported. Due to likely a renewed shortage of overall bank bond supply, the hunt for pick-up is likely to stay, supporting further credit flattening (despite historically low absolute yields already), i.e. the outperformance of lower rated covered bonds is set to continue.

Any significant take-off of SME covered bond issuance seems excluded for the foreseeable future. However, via excluding "loans to households for house purchases" in the TLTRO, the ECB implicitly recognises the effectiveness of covered bonds as a cheap bank funding tool.

We hear arguments that full allotment ECB funding can be treated as stable funding under NSFR. According to the BIS, "available stable funding is defined as the portion of capital and liabilities expected to be reliable over the time horizon considered by the NSFR, which extends to one year". The stated purpose of the NSFR is to promote longer-term resilience by requiring banks to have capital or longer term high-quality funding. In our view, including any crisis-inspired central bank funding as stable funding contradicts with the purpose of NSFR. The incentive to report a non-central bank supported sound NSFR could limit the negative impact of the 4Y TLTRO on public bond issuance by banks.

Fitch upgraded Santander Cedulas Hipotecarias (CH) to AA-*+ and revised downwards the refinancing spreads for Spanish, Irish and Greek mortgage covered pools. Moreover, Santan CH were just the first in the roll-out of the adjusted approach introducing IDR uplifts due to BRRD, i.e. except some specific cases with mainly bank driven downgrade risk, overall rating headlines are likely to be supportive for peripheral covered bonds.

We provide total covered bond redemptions by Eurozone country until Sep 2018, amounting to EUR 690bn in total, suggesting that negative net supply is likely to stay beyond 2014. EUR 207bn of Spanish Cedulas alone mature until Sep 2018, of which EUR 136bn are in publicly issued EUR benchmark format.

Due to ongoing negative net supply, secondary market liquidity is likely to get worse. In our view, while the ECB actions are positive for spreads, they are negative for liquidity.

Bernd Volk


6 June 2014



It is our understanding that commercial mortgage loans are eligible under the TLTRO. The total volume of Spanish CH amounts to around EUR 300bn. With an average share of around 30% non-residential mortgage loans backing CH and taking into account only 25% minimum OC, over EUR 110bn of CH collateral seem eligible for the TLTRO (as these non-residential mortgage loans are not "loans to households for house purchases").

The total volume of German mortgage Pfandbriefe amounted to EUR 200bn as of 31 March 2014, besides others, backed by 42% commercial and 22% multi-family mortgage loans. Hence, only taking into account 2% minimum OC and commercial mortgage loans (i.e. not considering multi-family loans), loans backing mortgage Pfandbriefe that are not "loans to households for house purchases" amount to over EUR 84bn.

S&P placed the A+ rating of BNP Paribas on watch negative due to the penalty to be imposed by the US justice department. S&P expects to lower the rating of BNP by a maximum of one notch. BNP SCF OF have three and BNP SFH OH have two notches of so-called "unused uplift". Overall, the impact on BNP covered bonds should be limited. A downgrade of BNP covered bonds as a consequence of a penalty seems unlikely.

Moody's downgraded Council of Europe Development Bank (COE) to Aa1 stable highlighting COE's high leverage (debt to usable equity of 813.3% in 2013) relative to other highly rated supras. Supporting the Aa1 is the COE's still exceptionally strong liquidity and conservative risk management policies which has ensured good performance of its loan book. In our view, given the specific structure of supras in general and likely strong shareholder support, the downgrade to Aa1 will have no impact on spreads.

Figure 1: Total redemptions of covered bonds by Eurozone country until Sept

2018, amounting to EUR 690bn in total

0.00

10,000.00

20,000.00

30,000.00

40,000.00

50,000.00

60,000.00

70,000.00

Aus

tria

Belg

ium

Cypr

us

Finl

and

Fran

ce

Ger

man

y

Gre

ece

Irel

and

Ital

y

Luxe

mbo

urg

Net

herl

ands

Port

ugal

Spai

n

2014

2015

2016

2017

2018

Source: Dealogic

Please also see our separate publication “Covered Bond and Agency Update” for further details.

6 June 2014



United Kingdom


A new 30Y Gilt

A new 30Y benchmark A new 30Y benchmark bond will be launched via syndication in the week of 23 June. It will be another January series, maturing on 22 Jan 45; we assume a coupon of 3.5% when looking at valuations. The GBP 8.5bn scheduled for syndicated nominal sales this year implies an issue size of around GBP 4.25bn.

When the UKT 44 was launched in Oct 12, the then benchmark had around GBP 26bn outstanding, and the UKT 40 around GBP 24.3bn. Currently the UKT 44 has GBP 26.7 bn outstanding, and the maturity is now slightly shorter than thirty years, which leaves sufficient room for a new benchmark.

Index Extensions We currently see the bond duration at 18.8Y, which is longer than the duration of the 15Y+ index; we estimate its inclusion will therefore lead to an extension of between 0.019Y to 0.021Y of the 15Y+ index, and between 0.035Y to 0.039Y for the full index.

Current outstandings by bond Index Extensions

0

5

10

15

20

25

30Outstanding Amt, £ bn

3.5% coupon Size, bn All 15Y+30Y 4.00 0.035 0.01930Y 4.25 0.037 0.02030Y 4.50 0.039 0.021

Source: Deutsche Bank, DMO Source: Deutsche Bank

Valuations The new bond is a year longer than the current benchmark and its duration is estimated to be 18.8Y, slightly longer than the surrounding bonds. The UKT 44 was issued at a spread of 8bp over the UKT 42, which has narrowed to around 4bp currently. The yield spread between the UKT 44 and UKT 46 is inverted, as it has been since it was issued, though this has disinverted slightly in recent weeks. From a comparator perspective, the bond appears cheap on the curve, but has richened slightly versus surroundings in recent weeks.

UKT 44 vs surrounding bonds Current term structure , 25Y+

4

5

6

7

8

9

10

11

12

13

14

-6

-5

-4

-3

-2

-1

0

3q44 - 4q46 Spread

3q44/4h42/4q46 Fly (rhs)

4h42-3q44 (rhs)

3.3

3.3

3.3

3.3

3.3

3.4

3.4

3.4

3.4

3.4

3.5

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

2039 2042 2045 2048 2051 2054 2057 2060 2063 2066

ASW

Yields, rhs


Soniya Sadeesh

Strategist (+44) 0 207 547 3091 [email protected]

6 June 2014



A simple linear interpolation on a maturity basis between the 44 and 46 implies a yield spread of 0.9bp below the 44, accounting instead for the duration and convexity characteristics of the bond implies a flat spread versus the 44. We also fit z-spreads and ted spreads, which implies a flat or -0.5bp spread respectively. Alternatively, interpolating the real yield curve, and assume a breakeven of around 3.48 suggests an implied spread of -1.5bp versus the 44s. Thus all in all, this would suggest the new bond would have a flat to -1bp spread to UKT 44.

Fitting the new bond Forwards between bonds

3.20

3.25

3.30

3.35

3.40

3.45

3.50

Mar-37 Aug-42 Feb-48 Aug-53 Jan-59 Jul-64 Jan-70

Yield

Implied Yield

3.00%

3.10%

3.20%

3.30%

3.40%

3.50%

3.60%

3.70%

3.80%

Aug-39 Feb-45 Jul-50 Jan-56 Jul-61 Jan-67 Jun-72

Fwds

Fitted

UKT 45 Par

UKT 45 Fwd


Duration and convexity characteristics The new bond offers around 3.3% convexity pickup over the 44s, and a 1.5% increase in duration. Compared to the 10y benchmark, it offers around five times the convexity and double the duration.

Duration vs surrounding bonds Duration and convexity

16.9

18.6 18.4

18.7

12

14

16

18

20

Jul-35 Apr-38 Dec-40 Sep-43 Jun-46 Mar-49 Dec-51

Mod Duration vs maturity

UKT 3.5 1_45

44s vs 46s

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

0.0 5.0 10.0 15.0 20.0 25.0 30.0

Duration (x-axis) vs Convexity (y-axis)

UKT 3.5 1_45


Broader market developments 30y yields have trended lower since the start of the year, currently around 25 bps below the levels seen at the start of the year, and marginally below the average seen over the past year. The 10y30ycurve has remained largely range bound since the start of the year, with 15y15y mostly staying between 3.60%-3.50% and 20y10y between 3.30%-3.50%.

On a cross market basis, the 30Y Gilt Bund spread is approaching multi-year wides; this is also true across the curve, reflecting the diverging economic cycles and monetary policy expectations. UK rates have also underperformed versus US rates, as the 30Y Gilt-Treasury spread has tightened around 30bp since the start of the year.

6 June 2014



Outright and forwards 10Y 30Y Slope 30Y cross market spreads

3.30

3.40

3.50

3.60

3.70

3.80

3.90

4.00

4.10

2.70

2.90

3.10

3.30

3.50

3.70

3.90

Feb-13 May-13 Aug-13 Nov-13 Feb-14 May-14

30Y

15Y15Y (rhs)

Par Yields

40

50

60

70

80

90

100

110

120

1302q23-3q44

2t24-3q44

-60

-40

-20

0

20

40

60

80

100

120

140

11 12 12 13 13 14

Gilt-Bund 30Y

Gilt-UST 30Y

Source: Deutsche Bank Source: Deutsche Bank, Bloomberg Financial LP Source: Deutsche Bank, Bloomberg Financial LP

Swap spreads Swap spreads have performed over the past six months, reflecting the starting point over the turn of the year of relatively cheap levels, as well a limited long end supply calendar. The 10Y-30Y ASW box has been largely stable over the past month, currently in line with the average level since the start of the year. The UKT 44-68 spread is -3.7bps (inverted) which is around 0.5bp richer than average level over the past six months.

Long end ASW ASW Box Spreads

0

5

10

15

20

25

30

35

40

2013 2013 2014

4h42 3q44 Ave 30Y+

0

5

10

15

20

25

30

35

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

Jan-

13

Feb-

13

Mar

-13

Apr

-13

May

-13

Jun-

13

Jul-1

3

Aug

-13

Sep-

13

Oct

-13

Nov

-13

Dec

-13

Jan-

14

Feb-

14

Mar

-14

Apr

-14

May

-14

Jun-

14

3q44-460 Spread 10Y-30Y ASW Box rhs


The curve has been rich relative to the level of rates (and also when including fundamental variables like inflation expectations, volatility and expected deficits) for many months. Flow data (with a lag) suggests this is at least in part due to strong demand from institutional players to manage liabilities.

Curve around syndications ASW around syndications

-40

-30

-20

-10

0

10

20

30

09 10 12 13 14

Residual 30Y+ Syndications

vs short rates, forward breakevens, risk aversion, expected deficits and vol

-10

0

10

20

30

40

50

60

Jul-09 Nov-10 Apr-12 Aug-13 Dec-14

30Y ASW

30Y+ Syndications


6 June 2014



Japan


Japan Strategy

Overview

The aggregate wage increased in April, but scheduled cash earnings per worker continued to fall (– 0.2% YoY). That said, with labor market supply/demand having tightened appreciably, we expect scheduled cash earnings to start rising once the percentage of (lower-paid) part-time workers levels out. The CPI and scheduled cash earnings would then rise in tandem for the first time since 1998, signaling that deflation may have finally been eradicated. A government-declared "victory" over deflation would make additional BOJ easing less likely. Moreover, if we are correct in our belief that the yen will weaken versus the US dollar, then we would expect the market's inflation consensus to move somewhat closer to the BOJ's envisaged trajectory. We therefore see potential for growth in scheduled pay and a depreciation of the yen to trigger a rise in JGB yields in August or September of this year.

Inflation expectations likely to build if wages rise and the yen weakens

Monthly Labour Survey data for April indicate that the per capita wage has yet to enter a clear uptrend. Total cash earnings per worker did rise 0.9% YoY to mark a second straight increase, but that was due solely to increases in non-scheduled and special earnings, with scheduled pay continuing to decline (–0.2% YoY). Scheduled pay rose by 0.1% YoY for full-timer workers and 0.6% YoY for part-timer workers, but the average nevertheless ended up declining as growth in the number of lower-paid part-timers (+2.8% YoY) exceeded that for full-time workers (+0.6% YoY; Figure 1). An increase in average scheduled pay will require (1) faster growth in wages for both full-time and part-time workers and/or (2) faster growth in employee numbers for full-time workers than for part-timers (as was seen back in 2005).

Figure 1: Change in scheduled cash earnings vs. changes in the numbers of

full-time and part-time workers

-6

-4

-2

0

2

4

6

8

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14

Scheduled cash earningsNumber of full-time workers (rhs)Number of part-time workers (rhs)

(year-on-year, %) (year-on-year, %)

Source: Nikkei database, Deutsche Securities

Makoto Yamashita, CMA


6 June 2014



We see potential for (1) and (2) to occur simultaneously. The active job openings-to-applicants ratio and the BOJ Tankan employment conditions DI both point to a tightening of supply/demand in the labor market. That said, as of April we have yet to see base pay hikes arising from this year's spring wage negotiations have a significant impact on average pay levels for all full-time employees or workers at small and midsize firms. Non-scheduled and special earnings are notoriously volatile, but growth in scheduled pay—once it begins—is likely to be relatively stable. It is also encouraging to see that firms are now spending more on personnel, with aggregate employee compensation—as calculated by multiplying the average wage by the total number of workers—rising 2.2% YoY in April. The unemployment rate is now so low that a rise in scheduled pay would not be at all surprising (Figure 2). We expect labor market tightening and base pay hikes to have a greater impact over the coming months, and will therefore be keeping a close eye on scheduled pay for May and June. The core CPI and the per capita wage have not risen in tandem since Japan fell into deflation back in 1997–1998 (Figure 3). Simultaneous positive growth might thus give the government further justification for its claim that Abenomics has been successful in defeating deflation, thereby reducing the likelihood that the BOJ will be called upon to deploy further easing measures.

Figure 2: Change in scheduled cash earnings vs. unemployment rate (1991–)

y = 0.4103x2 - 4.5254x + 11.563R² = 0.8858

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

4.0

5.0

6.0

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 (Unemployment rate, %)

(Change in scheduled cash earnings, %)


Figure 3: Change in scheduled cash earnings vs. core CPI inflation

-6.0

-4.0

-2.0

0.0

2.0

4.0

6.0

92 94 96 98 00 02 04 06 08 10 12 14

Scheduled cash earnings Core CPI(yoy, %)


6 June 2014



Markets are still considerably less bullish than the central bank on inflation: whereas the BOJ maintains that the core CPI inflation rate "is likely to be around 11⁄4 percent for some time (and then) follow a rising trend again from the second half of this fiscal year (4Q 2014 onwards)", the market consensus is for declines to +0.99% in 3Q and +0.92% in 4Q (Figure 4). We therefore see potential for a significant market movement in August or September, by which time it should be clearer just which forecast is more accurate. It is also possible that the expected inflation rate will rise if the yen weakens. History shows a tendency for the core CPI to rise following declines in the yen's nominal effective exchange rate (Figure 5). Market participants will therefore have less reason to predict a deceleration in core CPI inflation if USD/JPY rises in line with our own forecasts. As such, we see potential for growth in scheduled pay and a depreciation of the yen to trigger a rise in JGB yields in August or September of this year.

Figure 4: CPI forecasts: BOJ vs. market consensus

-0.5

0.0

0.5

1.0

1.5

2.0

2012 2013 2014 2015 2016

(%) Market consensus BOJ

Source: Ministry of Internal Affairs and Communications (Statistics Bureau), Japan Center for Economic Research, Deutsche Securities

Figure 5: Core CPI vs. JPY nominal effective exchange rate

-30.00

-20.00

-10.00

0.00

10.00

20.00

30.00 -3

-2

-1

0

1

2

3

95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14

Core CPI (lhs) JPY norminal effective exchange rate (rhs)(year-on-year, %)

Source: Ministry of Internal Affairs and Communications (Statistics Bureau), Bloomberg Finance LP, Deutsche Securities

6 June 2014



BOJ would be better off widening its target range for average residual maturity

On June 2 the BOJ reduced its allocation to the >10y JGB sector from JPY170 billion to JPY150 billion, while its June 4 operation allocated JPY300 billion to the 1y–3y sector (up from JPY250 billion) and JPY200 billion to the 3y–5y sector (down from JPY250 billion). Neither of these changes should have come as a major surprise following the May 29 revision of the BOJ's Outline of Outright Purchases of Japanese Government Bonds. On a delivery-date basis the average residual maturity of the BOJ's May JGB purchases was around 7.9 years, just below the ceiling of its stated target range of 6–8 years. We believe that the June 2 and June 4 adjustments were made with this range in mind, but feel that changes of this nature will do very little to hold down interest rates and thereby encourage portfolio rebalancing. Indeed, uncertainty over supply/demand in the maturity sectors for which the BOJ has reduced its allocations could actually lead to a slight widening of risk premiums. This may not be particularly problematic at the present time given the prevailing expectation that the BOJ will be unable to achieve its +2% "price stability target", but we believe that the central bank's policy objectives would ultimately be better served by widening the target range for average residual maturity beyond 6–8 years, thereby eliminating the need for micro-level tweaking of its operations.

Figure 6: Average residual maturity of BOJ JGB purchases by calendar month

6.00

6.50

7.00

7.50

8.00

8.50

May-13 Jul-13 Sep-13 Nov-13 Jan-14 Mar-14 May-14

(years)

Source: Bank of Japan, Deutsche Securities

6 June 2014



Japan


Japan Relative Value

This article was previously published as a Fixed Income Special Report on 2 June.

We continue to look for higher Japanese yields from late 2014. If this is realised then we think there is a possibility of a bear flattening of the curve as pent-up demand for long duration assets emerges to capture higher yields, given the slowdown in JGB purchases as yields have fallen.

10Y/20Y swap flatteners are expensive trades to hold, with negative carry and rolldown of about 1.3bp over six months. Alternatively, 6M forward 10Y/20Y bear flatteners can be entered at zero cost for about 2bp negative rolldown to spot.

However, despite steep curve levels at present, we think that bear steepening is a material risk especially if global risk appetite accelerates significantly and recent bull flattening of offshore curves is reversed.

Using a butterfly spread to capture a bear flattening of the curve has much less downside risk, in our view. Judicious selection of swap nodes can create a trade with near to flat carry cost (or even slightly positive) that displays good historical correlation with rates levels.

We recommend paying the JPY 5Y*2Y / 5Y*5Y / 10Y*2Y swap butterfly. We think this trade is positioned to capture a JPY rates selloff, including a potential bear flattening. Importantly, carry on the trade is close to flat.

The key risk is excessive cheapening of the 12Y+ part of the swap curve without a commensurate selloff in 10Y rates.

Low JGB yields could be causing a backup in demand Yen rates markets continue to remain wary of extending Q1 economic data trends (which have generally been stronger than expected) into the post-VAT hike period. 10Y JGBs have traded at yields below 0.60% for much of the period since 15 May, whilst 20Y JGBs have rallied through 1.46%.

Domestic market risk barometers initially weakened after the sales tax increase. From early April to 21 May the JPY strengthened from 104 to 101 against the USD whilst the Nikkei declined from 15,000 to 14,000 over the same period. However, both of these metrics have turned around over the past few weeks with the Nikkei now trading above 14,900 whilst the USD/JPY has traded back toward 102 – although JGB yields remain low. We think that downward pressure on yields in offshore markets has played a role in declining yields in Japan.

We noted in the Global Fixed Income Weekly of 23 May that Japan Securities Dealer Association data indicates that low yields may to be discouraging JGB purchases (see chart below). We think that this indicates pent-up demand for JGBs and this could contribute to rates remaining low in any short-term increase in yields. Overall we think that the timing of any substantial move higher in Japanese yields continues to be pushed further towards the end of the year.

Ken Crompton


6 June 2014



Net purchases of JGBs by trust banks and insurers vs. 20y JGB yield

1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4

0.000.250.500.751.001.251.501.752.002.252.50

Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14

Net purchase (lhs) 20y JGB yield (rhs)(trillion yen) (%)

Source: Japan Securities Dealers Association, Deutsche Securities

Demand could re-emerge at higher yields to generate a bear flattening We think that the long end of the yield curve is likely to flatten if rates sell off aggressively. The 10Y/20Y JGB slope is currently at +86bp, slightly above the mid point of the trading range that has held since September 2013 (+81bp to +89bp).

In early 2013, the 10Y/20Y slope was trading above +100bp, propelled in part by curve steepener trades in anticipation of Abenomics. However, we think that if yields begin to move significantly higher in the latter part of the year then some of the pent-up demand for longer duration JGBs (as shown above by the declining purchase volumes) will emerge, and we think this will tend to flatten the yield curve. Notably, the curve was not steeper than current levels in December 2013 when the 10Y JGB was trading at recent yield highs of 0.74%.

10Y JGB rally since 2007 has generally resulted in a

steeper curve; but 2003 was different

Rolldown to spot of forward starting swap flattner

0.4

0.9

1.4

1.9

2.420

30

40

50

60

70

80

90

100

110

120

Jun-98 Jun-01 Jun-04 Jun-07 Jun-10 Jun-13

10Y/20Y JGB slope

10Y JGB (RHS, inv)

-20

-10

0

10

20

30

40

0m 3m 6m 9m 12m 15m 18m 21m 24m 27m 30m

Forward start

Rolldown on forward starting flattener

1Y / 10Y Slope5Y / 30y Slope10Y / 20Y Slope1Y / 5Y Slope2Y / 3Y Slope

Source: Deutsche Bank, Bloomberg Financial LP Source: Deutsche Bank

The steepness of the yield curve through the 10Y sector means that 10Y/20Y flatteners incur negative carry. Executed via the swap market, a 10Y/20Y flattener will cost about 1.3bp in carry and rolldown over six months. Alternatively, ATMF conditional bear flatteners – sell 6M*20Y payer to fund a

6 June 2014



6M*10Y payer – can be entered at zero cost at forward slopes about 2bp flatter than the spot slope. (Over time, the 10Y/20Y swap slope has followed the same broad trends as the 10Y/20Y JGB slope and given greater flexibility for execution in swap rather than bond, we’ll consider the swap market from this point forward).

But we think that bear steepening is a risk – butterflies provide protection for a modest price Despite the evidence that a bear flattening is possible, we think that bear steepening of the curve is still a material risk. As noted in the first chart above in recent times the 10Y/20Y slope has tended to be inversely correlated with rate levels but during the 2003 rates selloff the curve steepened for a substantial period – albeit starting from much flatter levels than present. We think that market dynamics have changed from that time and QQE’s scope and nature is very different than the early BoJ QE programs, but the possibility of a rates selloff led by a global “risk on” move makes us nervous about unprotected flattener trades.

5Y*2Y / 5Y*5Y / 10Y*2Y butterfly is likely to cheapen if yields rise

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

-55.0

-50.0

-45.0

-40.0

-35.0

-30.0

-25.0

-20.0

Jan-11 Jun-11 Nov-11 Apr-12 Sep-12 Feb-13 Jul-13 Dec-13 May-14

5y*2y / 5y*5y / 10Y*2y 'fly

JPY 10Y swap (RHS)

Source: Deutsche Bank, Bloomberg

Instead, we think that butterfly trades are an alternative. Careful selection of forward starting swaps can create a butterfly trade with flat or slightly positive carry that has correlation with the level of rates, and will also perform in a bear flattening environment. One such trade is paying the belly of the 5Y*2Y / 5Y*5Y / 10Y*2Y swap butterfly. As the graph above shows, it has reasonable correlation with the level of the 10Y swap rate and we expect to see the belly of the fly cheapen if rates sell off materially from current levels. The concessions to carry cost mean that exposure to 10Y/20Y flattening is not large (the structure is not exposed to swap beyond the 12Y tenor) but a flattening of this part of the curve would still benefit the position.

The risk to the trade, however, is a richening of the butterfly during a selloff as occurred in late 2012. However, that was before QQE expanded long duration bond purchases and, additionally, the effect we mentioned above of potential emergence of JGB demand at higher yields is likely to cap long end moves.

We recommend paying the belly of the 5Y*2Y / 5Y*5Y / 10Y*2Y JPY swap butterfly. The butterfly has flat carry and rolldown over six months.

6 June 2014



Global


Global Relative Value

We recommend two asymmetrical trades in EUR to position either for further short term rally in the belly of the curve on the back of the ECB package, or for some normalization of the 5y5y sector in the medium run if the market ultimately interprets the ECB package as strong enough to reduce deflation risk.

EUR 1M fwd 2Y-5Y conditional bull flattener: This trade is a cost efficient and carry efficient way to position for further rally in the 5y sector in the near term. Investors who believe the 5y rally has some legs as the unexpectedly long 4Y TLTRO announced by the ECB could delay the policy rate normalization process beyond 2017, should consider the conditional bull flattener. We estimate that the most dovish interpretation of the ECB package could lead to an additional rally of the 5y rate by up to 10-12bp while room for a 2y rally is much more limited. Taking into account the strong directionality of the slope with the 5y rate (stable beta of 55%) we find the 1M expiry 2s5s bull flattener to be 30% cheaper than the equivalent position in the outright 1M5Y payer, highlighting some residual richness in 2y vol vs. 5y vol. More importantly the roll up on the 1M2Y leg adds some positive carry to the trade so that the 1M fwd 2s5s flattener benefits from a 200% carry advantage against an outright position in 1M5Y. Carry more than compensates the cost of the structure so that the trade would start losing for a bull steepening of the curve beyond 0.5bp.

Buy EUR 6M5Y5Y midcurve payer spread +10bp/+50bp or payer ratio +10bp/+50bp/+70bp 1/-3/2: Investors who believe the ECB package will ultimately manage to boost lending to the private sector and remove the deflation tail risk should position for a normalization sell off of the 5y5y rate in the medium run. We find 6M5Y5Y midcurve payer ATMF vol to trade at cheap levels historically while the midcurve payer skew is historically rich. In other words, the combination of historically low 5y5y rate, historically low 6m5y5y implied vol and historically rich 6m5y5y payer skew makes it particularly attractive to implement a 6m5y5y payer spread. With the GDP-weighted Eurozone 5Y5Y yield rich by ~70bp in our adjusted BRP framework which takes into account long term expectations of nominal growth in Eurozone and the SPF inflation skew, we would consider a payer spread 10bp/50bp for a cost of 10bp running and a max payoff of 40bp hence a 4:1 leverage or a payer fly 10/50/70 1:3:2 for a 6bp running premium and leverage ~6.5:1.

#1- EUR 1M fwd 2Y-5Y conditional bull flattener

We recommend a 1M fwd 2s5s bull flattener with ATMF receivers as a cost-efficient and carry-efficient way to position for further rally in the 5y sector in the near term following the larger than expected ECB stimulus package.

While front eonias are already mostly pricing in the 10bp depo/refi corridor cut and the upcoming amount of excess liquidity, far forward eonias beyond the 2y sector still have room to richen according to the Eonia strip. Indeed The extension of the fixed rate full allotment until Dec-2016 is a commitment to

Jerome Saragoussi


6 June 2014



remain fully accommodative at least for the next couple of years, while the 4Y TLTRO potentially expands the forward guidance until 2018. Currently the Eonia strip is implying a normalization of Eonia towards 25bp by the end of 2016 whereas the fixed rate full allotment until Dec-2016 should help anchoring Eonia towards the refi rate of 15bp. Beyond Dec-2016, the market expects an increase of forward eonias via rate hikes towards 50bp by mid 2017. With the end of the 4Y TLTRO in mid 2018, the market could well assume that the normalization of monetary policy won’t take place until the expiry of the TLTRO program. The market could therefore potentially price in a more dovish path of rates with Eonia reaching 50bp in mid-18 instead of mid-17. We show in the chart below the bull flattening that would result from the repricing of Dec16 Eonia from 25bp to 15bp and the one year delay to converge towards 50bp from mid 2017 to mid 2018. In this scenario the 5y rate could rally by an additional 10-12bp and the 2Y-5Y slope could flatten by 6-8bp.

Figure 1: More aggressive repricing of ECB package could lead to significant

rally in 5Y and bull flattening in 2Y-5Y

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Jun-14 Dec-14 Jun-15 Dec-15 Jun-16 Dec-16 Jun-17 Dec-17 Jun-18 Dec-18 Jun-19

Poly. (Series1)Poly. (Series2)Scenario forward Eonia strip

Delaying Eonia normalization to 50bp by one year

Dec16 Eonia at 15bp instead of 25bp


Beyond the above scenario analysis, we find strong empirical evidence that the 2s5s slope has been extremely directional with the level of 5y rate over the past couple of years. With a stable beta (using levels and changes) of 55% and an R-2 above 90% over the past few months (see figure 2 below), the 2s5s flattener can be regarded as an excellent proxy for an outright long 5y, so that the 2s5s bull flattener is simply a proxy for a long 5y receiver. Interestingly a carry and cost analysis of the 1M fwd 2s5s conditional bull flattener against a 1M5Y receiver shows a large superiority of the bull flattener, making it an attractive trade to position for further rally in the belly.

Cost analysis: Taking into account the 55% beta of the 2s5s slope with the 5y rate we find the 1M expiry 2s5s bull flattener to be 30% cheaper than the equivalent position in the outright 1M5Y payer. Indeed the cost of the bull flattener is 2bp running which should be compared to the cost of the beta adjusted equivalent position in 1m5y i.e. 55% x 5.2bp = 2.9bp.

Carry analysis: The roll up on the 1M2Y leg adds some positive carry to the trade. The 1M fwd 2s5s flattener benefits from 2.5bp of positive, which should be compared to the carry on the beta adjusted equivalent long position in 1M5Y i.e. 55% x 1.6bp = 0.9bp. In other words the 1M fwd 2s5s flattener benefits from a ~200% carry advantage against an outright position in 1M5Y. Carry of 2.5bp more than compensates the cost of the structure of 2bp running so that the trade would only start losing money in a rally for a bull steepening of the curve beyond 0.5bp (see figure 3 below for calculation details).

6 June 2014



2 – Strong correlation between 2s5s slope and 5y rate 3- EUR fly remains rich vs. Eonia and liquidity

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Apr-13 Jul-13 Oct-13 Jan-14 Apr-14

EUR 2Y-5Y slope (lhs)

EUR 5Y swap (rhs)

EUR 2s5s flattener vs. 5y receiver

Beta 2s5s slope vs 5y rate 55%

Cost 1M5Y receiver 5.2bp

Positive carry 1M5Y 1.6bp

Total cost 3.6bp

Beta-adjusted cost 2.9bp

Beta-adjusted positive carry 0.9bp

Total cost 2.0bp

Cost 1M2Y vs. 1M5Y conditional bull flattener 2.0bp

Positive carry flattener 2.5bp

Total cost -0.5bp

Discount premium 2s5s vs 5s -30%

Carry advantage 195%


The key risk of the trade would be a bull steepening of 2s5s. In a sell-off the receivers would end up out of the money and the loss would be limited to the small premium spent at inception.

#2 – EUR 6M5Y5Y midcurve payer spread 10bp/50bp 1:1 or EUR 6M5Y5Y midcurve payer ratio 10bp/50bp/70bp 1:-3:2

Investors who believe the ECB package will ultimately manage to boost lending to the private sector and remove the deflation tail risk should position for a normalization sell off of the 5y5y rate by the end of the year. Given the ECB stimulus is significant enough to compress further the belly of the curve and favor carry trades, and given some uncertainty about the conditionality attached to the 4Y TLTRO, the rates direction for 5y5y remains particularly symmetrical in the near term. We therefore favor cheap conditional expressions of the bearish trade in the 5y5y sector via a 6M5Y5Y midcurve payer spread (for instance 10bp/50bp one-to-one) or payer ratio 1:-3:2 10bp/50bp/70bp.

The logic for the trade is threefold (1) 5Y5Y is close to historically rich levels at a time when the ECB is likely to success in lifting inflation expectations and reducing significantly the inflation tail risk (2) 6M5Y5Y midcurve vol is getting close to historically low levels but could be supported going forward (3) 6M5Y5Y payer skew is historically high.

(1) Richness of EUR 5Y5Y rate: The EUR 5y5y rate is trading close to historically rich levels and looks historically dislocated against long term fundamentals i.e. long term growth and inflation expectations in Eurozone. In other words the 5Y5Y bond risk premium is historically negative. Looking at the GDP-weighted 5Y5Y yield in Eurozone relative long term expectations of nominal GDP growth also shows a historically large dislocation. Some the structural decline of the bond risk premia over the past decades has been explained by the decline in high inflation tail risk relative to low inflation tail risk. While the SPF distribution has shown a marginal deterioration of the inflation skew in the past couple of quarters it has been insufficient to justify the magnitude of the 5Y5Y rally in Europe (see figures 1 and 2 below).

6 June 2014



1-Historical richness of EUR 5Y5Y adjusted BRP, taking

into account SPF inflation skew

2- Adjusted Bond risk premium of the EUR GDP-

weighted 5Y5Y yield becoming historically rich too -3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

93 95 97 99 01 03 05 07 09 11 13

Adjusted EUR 5Y5Y swap BRP (lhs)

Subsequent 12M change in 5y5y swap (rhs, inv)

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

98 00 02 04 06 08 10 12 14

EUR GDP-weighted 5Y5Y adjusted BRP

Subsequent 12M change in EU GDP-weighted 5y5y yield (rhs, inv)


Interestingly, the EUR inflation swap and option market has had a much more pessimistic view about the long term inflation outlook in Europe. Unlike the SPF distribution roughly symmetrical around 1.75% expectations for 5Y inflation, the EUR inflation market has been pricing in 5y inflation expectations around 1.20% on average in the past quarter, with a probability of inflation being on average lower than 1% at 40% (see figure 3 below for the difference between market distribution and SPF distribution of 5y inflation). The inflation skew priced in by the market (probably of seeing inflation above 2.25% vs. probability of seeing inflation below 1.25% in 5 years) has been extremely negative reaching about -35%. Historically, this level of inflation skew would help justifying the extremely negative 5Y5Y bond risk premium (see figure 4).

3-Market pricing in extreme deflation risk relative to

economists (survey of professional forecasters by ECB)

4- Market pricing is consistent with large deflation tail

risk, 5y5y should normalize as deflation risk recede

0%

5%

10%

15%

20%

25%

30%

<0 0-0.5 0.5-1 1-1.5 1.5-2 2-2.5 2.5-3 3-3.5 3.5-4 4-4.5 4.5-5 >5

Distribution of 5Y EU inflation implied by ZC HICP cap/floor market

Distribution of 5Y EU inflation from ECB SPF

-50%

-30%

-10%

10%

30%

50%

70%

90%

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

94 96 98 00 02 04 06 08 10 12

5Y5Y basic BRP (lhs)

Probability of high infl vs. low infl from 5Y HICP ditribution from ECB SPF (last data point shows market inflation skew instead of SPF)


The implementation of a large stimulus package by the ECB has already lifted long term inflation expectations with 5Y5Y HICP swap jumping from about 2% a couple of few weeks ago towards 2.12% currently. The probability of inflation being lower than 1% in the next 5 years has already dropped from 40% a week ago towards 35%.If the ECB is successful in boosting the market continues to price out some deflation tail risk, long

6 June 2014



dated forward rates between 5y5y and 10y20y should see an increase in the negative inflation risk premium currently embedded in valuations.

(2) Cheapness of 6M5Y5Y midcurve vol: The history of 6M5Y5Y midcurve implied vol shows some significant cheapening over the past few months thanks to the cheapening of 5Y and 10Y gamma and the stabilization of implied correlations at elevated levels. Interestingly though, 6M5Y5Y midcurve vol is becoming increasingly sticky to the downside because 5Y gamma is cheapening more than 10y gamma. 6M5Y5Y midcurve vol is indeed also a function of the vol slope between 6M5Y and 6M10Y implied vol. With the ECB stimulus compressing the 5y sector, increasing forward guidance and favoring carry trades, 5y gamma is likely to remain under pressure relative to 10y gamma, so that the spread between 6m5y and 6M10Y vol should widen further preventing 6M5Y5Y midcurve vol from cheapening much further going forward.

(3) Richness of 6M5Y5Y payer skew: With payer skew generally historically steep on 5y and 10y payers, we also find 6M5Y5Y midcurve payer skew to be historically high. The combination of historically cheap ATMF vol and historically rich payer skew at a time when the underlying rate is fundamentally rich makes it particularly attractive to position for some medium term partial normalization of the 5Y5Y rate until the end of the year using 6M5Y5Y payer spread (long vol exposure but short skew).

5- Historical cheapness of 6M5Y5Y midcurve vol 6- Historical richness of 6M5Y5Y payer skew

0

20

40

60

80

100

120

140

160

180

200

220

06 07 08 09 10 11 12 13 14

6m 5y6m 10y6m5y5y Midcurve vol

0.85

0.90

0.95

1.00

1.05

1.10

1.15

06 07 08 09 10 11 12 13 14

EUR 6M5Y5Y payer skew OTM+50bp vs. ATMF ratio


We recommend for instance a 6M5Y5Y 10bp/50bp payer spread for a running premium ~10bp at the time of writing for a max payoff of 40bp hence a leverage of 4:1. Pricing the structure historically shows that there has not been a better time to enter this type of trade given the combination of historically cheap cost of the structure and upside risk on EUR 5y5y from historically rich levels. Investors who are ready to sell the tail risk of sell-off beyond 50bp and to get a neutral to marginally short vol exposure can also implement the 6M5Y5Y 10bp/50bp/70bp payer ratio (1:-3:2) for a premium around 6bp running and a leverage around 6.5:1 (see P&L profiles in figure 7 below).

6 June 2014



7- Historical cheapness of the payer spread at a time

when 5Y5Y is historically low

8 – P&L profile of payer spread and payer ratio strategies

at expiry

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

5bp

7bp

9bp

11bp

13bp

15bp

17bp

19bp

06 07 08 09 10 11 12 13 14

Running premium 6M5Y5Y payer spread 10bp/50bp

EUR 5y5y rate (rhs)

-15

-10

-5

0

5

10

15

20

25

30

35

40

2.3% 2.4% 2.5% 2.6% 2.7% 2.8% 2.9% 3.0% 3.1% 3.2% 3.3% 3.4% 3.5%

6M5Y5Y payer ratio 10/50/70, P&L profile at expiry

6M5Y5Y payer spread 10/50 P&L profile at expiry

Level of 5Y5Y rate at expiry in 6M


The risk of the payer spread and payer ratio is limited to the premium paid at inception. The payer spread is at risk of a rally. The payer ratio is at risk of a rally or a large sell-off beyond 70bp but the maximum loss cannot exceed the premium paid at inception.

6 June 2014



Asia


Asia

China: We expect the current round of targeted RRR cuts to have a more meaningful impact on liquidity compared with last time. We forecast about RMB300-350bn liquidity injection in order to provide an additional 1.5-1.8% of new loan growth. We think these cuts indicate that the State Council is reinforcing its targeted credit easing bias in order to stabilize economic growth. Against this backdrop, we prefer to trade the Repo swap curve with a bullish bias via outright longs on 2Y at 3.60% (target 3.35%) and 2Y/5Y steepeners. We continue to expect further policy relaxation, and for assets such as financial bonds and highly rated corporate bonds that offer good carry on risk adjusted basis to outperform.

India: RBI seems willing to turn the dial on rates if the new government in Delhi delivers on fiscal consolidation and better food supply management. The interest rate curve has rallied a fair bit though is arguably still under pricing a possible shift in liquidity outlook, as also broadening out of offshore engagement in debt to long term investors. The immediate risk-reward is perhaps a bit challenging for the rally to extend, given the upcoming Budget. But any back up in yields should be used to add to duration exposure

Korea: We are not chasing the rally at current levels and keep to our existing box trade via KTB futures and swaps. We are past the immediately bullish factors – both domestic and global - for fixed income, as also the main political event of this year in Korea. We also believe that the offshore impact on the rates market is likely to come via KTB futures with a risk of long liquidation, rather than an increase in cash bond investment

Thailand: Linkers appear to us to be far more attractive than the nominals at this juncture in the Thai market, given their compelling valuations, upward trajectory of inflation, a sweet spot to benefit from the potential El Nino problem given its CPI basket composition, a low beta to the nominal yields and most importantly the cushion they provide against rise in yields from the solid carry profile ahead. Linker trading volumes have been picking up as they gain traction following the successive higher inflation prints. PDMO’s plans to include Linkers in global indices in the future should be structurally positive from a demand perspective.

China

Targeted easing reinforces the bullish trend In its regular meeting on 30 May, the State Council decided to step up financial support to the real economy and specifically to expand the scope of targeted RRR cuts to selected financial institutions that have extended new loans above a certain threshold to the agriculture sectors, small and micro firms, etc in order to support domestic structural rebalancing. We expect the PBoC to announce the implementation details in the near future.

This time, the targeted RRR cuts also aim at supporting lending to small and micro firms and other sectors that are important for China’s economic structural rebalancing. We believe this decision is consistent with PBoC’s

Sameer Goel

Strategist (+65 ) 64236973 [email protected]

Linan Liu

Strategist (+852) 2203 8709 [email protected]

Kiyong Seong


Swapnil Kalbande


6 June 2014



forward guidance expressed in the Q1 monetary policy report that (a) it is prepared to relax monetary condition to support growth and (b) it remains committed to credit stimulus to ensure financing support to targeted sectors.

We believe the purpose of targeted easing is to improve monetary transmission in bank loans where credit conditions remain tight for weak credits/less favored borrowers. We believe the breakdown from liquidity transmission to loan growth reflects (a) significant pass-through from banks’ funding costs to bank loan rates to protect NIM, (b) commercial banks are cautious with regard to asset allocation, considering that volatilities of their traditional deposit base are likely to remain high and (c) low credit risk appetite given rising NPLs and risks of more default events in the credit market later this year. By cutting required reserves selectively, commercial banks would have more funds to lend and earn higher margins (commercial banks only earn 1.62% on required reserves), and this hopefully will incentivize banks to lower the loan rates to small and micro firms.

We expect this round of RRR cuts to have a more meaningful impact on liquidity than last time and we forecast about RMB300-350bn liquidity injection in order to provide an additional 1.5-1.8% of new loan growth.

We think the RRR cut indicates the State Council is reinforcing its targeted credit easing bias in order to stabilize economic growth; against this backdrop, we prefer to trade the Repo swap curve with a bullish bias either via outright receiving 2Y Repo NDIRS at 3.60% (target 3.30% and stop loss 3.7%), or by holding 2Y/5Y Repo NDIRS steepeners.

We maintain our expectation of further policy relaxation and outperformance of assets that offer good carry on risk adjusted basis, such as financial bonds and highly rated corporate bonds.

The risks to our call are (a) in the near-term, liquidity squeeze towards the end of June and (b) the scope of the RRR cut is significantly less than our forecast. We expect modest tightening in liquidity towards the last week in June but would take any squeeze as an opportunity to add outright receiving IRS positions.

The rally is likely to continue…. …and CDBs are likely to outperform CGBs

2.50

3.00

3.50

4.00

4.50

5.00

5.50

Dec-11 Jun-12 Dec-12 Jun-13 Dec-13 Jun-14

10Y CGB

2Y NDIRS

1

2

3

4

5

6

7

Jul-08 May-09 Mar-10 Jan-11 Nov-11 Sep-12 Jul-13 May-14

5Y CGB

5Y CDB

Source: Deutsche Bank, CEIC Source: Deutsche Bank, CEIC

6 June 2014



India

RBI willing to turn the dial We were admittedly surprised by the timing of the dovish tone adopted by the Reserve Bank in its policy statement this week, though not necessarily by the direction of its shift. In explicitly noting that ‘if disinflation, adjusting for base effects, is faster than expected, it will provide headroom for an easing of the policy stance’, the central bank has in effect given due nod to the growth objective of the new government. And expressed its willingness to put a certain degree of faith in the ability of the new administration to deliver on the twin goalposts of ‘better fiscal consolidation’ and ‘stronger government action on food supply’. In doing so, it has also put the ball firmly in the government’s court.

The duration play is well underway, and now has the blessings of the central bank. To be sure, a) it will likely take a few months before RBI is convinced enough to pull the trigger on cutting rates; and 2) valuations look fair and domestic positioning somewhat stretched after the post-RBI rally in government bonds.

But equally, we would note that,

The likely expansion of RBI’s external balance sheet, as it intermediates a growing quantum of portfolio inflows into the economy over the next few months, gives the CB another policy lever (liquidity) to effectively ease the term structure of rates gradually without having to resort to rate cuts.

The curve is pricing in close to 50bp of easing over the next 12 months

3

4

5

6

7

8

9

10

11

Jun-10 Jun-11 Jun-12 Jun-13 Jun-14 Jun-15

LAF Corridor, Current vs DB Economics forecastsO/N Mibor: 1m Avg, Current vs Implied from the OIS curveMSF, Current vs DB Economics forecasts

%

Source: Deutsche Bank, Bloomberg Finance LLP

While the OIS curve is already mostly pricing in our expectation of 50bp of rate cuts over the next 12 months, it has not necessarily priced in the possibility of a swing in liquidity to surplus in line with moderation of RBI’s sterilization as and when the CB gets convinced with the sustainability of the disinflationary trend. This could cause the front end of the OIS curve to move towards 7.5%, and likely disinvert.

Second, offshore engagement in the duration play has yet to broaden out. Bulk of the $6bn year to date (including just over 2bn in the last month) increase in FII holdings of government paper have been we believe by bank desks. The part reserved for long term fixed income investors like sovereign wealth funds, insurers, pension funds etc. has seen very little take up. There is still just under $10bn of quotas available for foreign holdings.

6 June 2014



Bulk of the FII buying of government debt YTD has come from bank desks,

not long term real money

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14

open category

reserved for SWFs etc

USD bn

Net cumulative purchases

Source: Deutsche Bank, NSDL

We think the risk-reward on extension of the rally in cash bonds to under 8.5% is a bit challenged ahead of the Budget. But we would use any back up in rates as an opportunity to add to the duration exposure.

Korea

Investor positioning will matter We have argued that the rationale (partly a short squeeze) for the UST rally is not necessarily applicable to KTBs, given the influence of foreign KTB futures movement, which remains at deep net long positions. We believe foreign KTB futures positioning will continue to hold sway, especially on the back of a technical back-up in UST yields. Within the local market, a delayed correction in rates had made local investors nervous and contributed to the recent rally through duration covering to some extent by local long-term investors. The ferry disaster in April had also supported the market bulls. This time, we believe the theme is likely to go the other way round. The confidence on lower yields will likely weaken as time goes on. The concern about a contraction in domestic consumption due to the ferry disaster will likely dissipate gradually (global PMIs are likely to rebound too). Hence, we are not chasing the rally here and will keep to our existing box trade via KTB futures and swaps.

We continue to believe the key difference between Korea and the US is the depth of investors’ positioning on either side (shorted in the US, but mixed in Korea). While domestic long-term investors have been known to restore their duration to some extent, the other pillar of market movement, foreigners, has been relatively quiet in the cash bond side and liquidated their long positions in KTB futures. Their net long positions in the 3Y, 10Y KTB futures decreased to 96 thousand and 20.5 thousand contracts from the recent highs of 108.7 thousand and 22.2 thousand contracts, respectively (see chart following).

6 June 2014



Foreign 3Y, 10Y KTB futures net long positions est.

-10

-5

0

5

10

15

20

25

30

-50

0

50

100

150

200

Mar-13 Jun-13 Sep-13 Dec-13 Mar-14

Offshore 3Y KTB futures net long, thousand, LHS

Offshore 10Y KTB futures net long, thousand, RHS

Source: Deutsche Bank, Bloomberg Finance LP

As of 2 June, foreign bond holdings stand at KRW95.48tr vs. the recent peak of 97.4tr on 16 May. Foreign monthly activities in the secondary market also decreased (see chart following). This is partly due to FX market movement, finding a support for USD/KRW at the 1,020 level. Foreign fixed income players are closely watching to see whether the current support level for USD/KRW will collapse again. At this stage, however, they seem to be cautious about front-loading their bond investment, anticipating Korean won appreciation. While our FX strategists expect USD/KRW to breach the 1,000 level in the future, the pace of such an appreciation would be gradual on the back of smoothing activities by the FX authorities. In this regard, scope for FX gain would not be attractive and imminent enough to induce offshore fast money in our view. In a nutshell, the offshore impact on the rates market is likely to come via KTB futures with a risk of long liquidation, rather than an increase in cash bond holdings, in our view.

Foreigner, pension, insurer monthly KTB net purchase at the secondary

market

-1000

-500

0

500

1000

1500

2000

2500

3000

3500

4000

Insurer Pension Foreigner

Dec-13 Jan-14 Feb-14 Mar-14 Apr-14 May-14

KRWbn

Source: Deutsche Bank, KOFIA, As of 29 May

Lastly, the main political event of this year is now behind us. The local elections on 4 June were very intense amid the public anger over the ferry disaster in April. The disaster had also created a somber mood and hurt domestic consumption. This should ease as head into the World Cup season. Note though that the rates market itself had rallied in all World Cup years since 2002, which was due to trend lower Korea yields. In the past four World Cup years, there was no notable impact of World cup on consumption either (see

6 June 2014



table below). However, the market has already priced in a slowdown in global/domestic growth, including the consumption contraction due to the ferry disaster. Therefore, any sign of a snapback in domestic consumption will likely lift the trading range a bit. If the Korean team performs well in the Cup, that would be also helpful for restoring confidence of the Korean people.

Retail sale (%, yoy) in the years of World Cup and local elections 2014 (Brazil) 2010 (South

Africa) 2006 (Germany) 2002 (Korea-

Japan) World cup year

average

Jan 5.64 6.15 10.65 9.66 8.03

Feb (0.40) 11.74 0.91 18.08 7.58

Mar 2.15 8.73 3.88 15.57 7.58

Apr (0.09) 7.81 4.70 13.40 6.46

May 4.35 5.45 10.53 6.78

Jun 4.33 5.06 8.09 5.83

Jul 9.52 (0.63) 11.23 6.71

Aug 9.32 4.26 13.58 9.05

Sep 5.33 5.31 6.96 5.87

Oct 4.52 4.03 10.98 6.51

Nov 6.69 3.08 6.62 5.46

Dec 3.91 1.78 4.20 3.30 Source: Deutsche Bank, Bloomberg Finance LP

Thailand

Linkers are more attractive than nominals Valuation on the linkers is compelling when compared to the nominals, especially against the back-drop of rising inflation. For example, 15Y Linker (ILB283A) sold off 135bp between mid July to mid-Sep, whereas 15Y nominal bond (LB27DA) sold-off only 50bp during the same period. Falling inflation during this period can explain to some extent this underperformance of Linkers back then.

15Y nominal (LB27DA) has rallied about 65bp since this sell-off and the yield is now trading below the July levels. But, 15Y Linker (ILB283A) yield on the other hand failed to re-visit the July lows, despite the inflation consistently pushing higher since October.

Linkers underperformed the nominals between July to

September on falling inflation…

… But, failed to catch up with nominals despite inflation

picking up since October

3.00

3.25

3.50

3.75

4.00

4.25

4.50

0.50

1.00

1.50

2.00

2.50

3.00

Jun-13 Aug-13 Oct-13 Dec-13 Feb-14 Apr-14 Jun-14

Inflation15Y Linker (ILB283A)15Y LB (LB27DA, RHS)

% %

Linkers underperformed the nominals on back of falling inflaation

3.00

3.25

3.50

3.75

4.00

4.25

4.50

0.50

1.00

1.50

2.00

2.50

3.00

Jun-13 Aug-13 Oct-13 Dec-13 Feb-14 Apr-14 Jun-14

Inflation15Y Linker (ILB283A)15Y LB (LB27DA, RHS)

% %

Nominal yields are well below the July levels now, but, Linker yields are still way above the July Levels despite inflation picking up

Source: Deutsche Bank, Bloomberg Finance LP Source: Deutsche Bank, Bloomberg Finance LP

6 June 2014



This translates into attractive break even. With 15Y nominal yield (LB283A) at 3.82% and real yield (ILB283A) of 2.11%, break-even inflation stands at ~1.71% versus average inflation of 3.15% over the past 10 years, and most recent inflation print of 2.62% YoY.

Inflation has already bottomed out, and is likely to push past 3%. As we have been highlighting inflation has already bottomed out and our economists argue that looking ahead, a likely combination of weakening baht, rising fuel prices, and concerns about El Nino affecting food production would push up inflation past 3% in the next few months. Note that food component has a 33% weighting in Thailand CPI basket, highest compared to other markets in general that offer Linkers. Bloomberg earlier cited Australia’s Bureau of Meteorology that El Nino has at least 70% chance of developing in 2014, with most climate models suggesting it will become established by August. Thai Linkers are thus well placed to benefit the most from these potential El Nino problems, compared to other Linker markets in general given Thailand’s relatively higher weighting of the Food component in the CPI basket.

Break-even inflation implied from the Linker appears too

low, against both the long run avg. and recent print

Much higher weight of food component in Thailand CPI

basket compared to others

Source: Deutsche Bank, Bloomberg Finance LP Source: Deutsche Bank, PDMO

Importantly, carry profile on Linkers is very attractive given rising inflation trajectory. Not surprisingly, the inflation accrual component stands to benefit from rising inflation trajectory, thus translating into an attractive carry profile for Linkers. Based on DB economics forecasts for inflation and assuming real yields remain the same, the linker should provide a return of roughly 1.65% over next 3 months. Assuming a funding cost of 0.55% during this period (2.2% annualised), the net carry return is 1.10% over 3 months. For the 15Y Linker (ILB283A), considering its duration of roughly 12.5, that translates into a cushion of roughly 9bp. Whereas, for the 7Y Linker (ILB217A), considering its duration of roughly 6.75, that translates into a cushion of roughly 16bp, For real money accounts (i.e excluding the repo cost), the cushion would be roughly 12bp and 24bp respectively.

Low Beta of Linkers to the nominals. It is not surprising that changes in Linker yields run a low Beta to the changes in nominal Loan Bond yields. As such, even if we do get some more adjustment in the Thai nominal bond yields as market starts to price in heavy supply, fiscal spending from the government and less policy room for further easing, the low beta of Linker yields to the nominals mean that carry on the linkers should be able to offset any upward adjustment in Linker real yields.

6 June 2014



Inflation is on an upward trajectory, and is likely to push

past 3%

Low beta of Linker real yields to nominal bond yields

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

Dec Jan Feb Mar Apr May Jun Jul Aug Sep

%

CPI : Actual

CPI : DB Economics Forecasts

y = 0.271x + 0.0017R² = 0.12

-0.30

-0.20

-0.10

0.00

0.10

0.20

0.30

-0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20

Yield changes over last two years

Daily yield change in 10Y Loan Bond

Daily yield change in 10Y Linker

Source: Deutsche Bank Source: Deutsche Bank, Bloomberg Finance LP

Linker trading volumes have picked up. Meanwhile, as the inflation is picking up, we are also witnessing a surge in daily trading volumes for Linkers. To wit, daily trading volumes on ILB217A have clearly improved over the past month and rose to YTD record levels following the higher May CPI print earlier this week.

Linker trading volumes rose to YTD record highs

1.50

1.70

1.90

2.10

2.30

2.50

2.70

0

100

200

300

400

500

600

700

800

900

1,000


Daily Trading Value

Inflation (RHS)

THB mn

ILB217A

%

Source: Deutsche Bank, ThaiBMA

Finally, the PDMO’s plans to include Linkers in global indices in future should be structurally positive from a demand perspective.

Plans of Including Thai Linkers in global indices

Source: Deutsche Bank, ThaiBMA

6 June 2014



Pacific Australia New Zealand


Dollar Bloc Strategy

We moved underweight duration in our monthly Australian fixed income

strategy published on 4 June.

The RBNZ publishes its June MPS on 12 June. We expect a 25bp rate hike, but also a modest downward adjustment in the forward path for interest rates as it signals a possible pause in the rate hike cycle. This will close the gap that has opened up between the RBNZ and market pricing somewhat. Still, given this gap we don’t see a lot of scope for the NZ front-end to rally from here.

We think we are at the point where the direction of the US long-end will again become more important for the NZ curve, assuming we are right that the RBNZ is set to pause its tightening cycle after the expected rate hike on Thursday. With this in mind we recommend the 2Y/10Y steepener. The biggest risk to the trade is the RBNZ does not pause in hiking the OCR. A strong rally in the US long-end will also hurt the trade.

The AOFM has outlined its thoughts on possible issuance in 2014/15. There will be a new nominal 2019 bond and it seems likely that the nominal yield curve will be extended beyond the current 2033 maturity. We think the primary surprise for investors will be the indication that the linker curve may not be extended beyond the existing 2035 line in 2014/15, with the AOFM no longer thinking it needs to lead extension of the curve via linkers. We commented on this in the Inflation Monthly published on 5 June where we recommended a break-even steepener. Finally, the AOFM has committed to ongoing issuance of T-notes.

Sharp pullback in 10Y yields After commenting in last week’s publication that 10Y yields had fallen to their lowest point for the year, there has been a subsequent bounce that has taken 10Y yields up by more than 10bp across the $-bloc.

With the 10Y ACGB/UST spread at a level that we think effectively prices in an RBA rate cut it will be difficult for the 10Y ACGB to outperform on either a rally or a sell-off from here, in our view. When we couple this with our medium-term view that yields move higher our preference is to be short. Consistent with this we moved underweight duration in our monthly Australian fixed income strategy published on 4 June.

Will the RBNZ close the gap with market pricing in the upcoming Monetary Policy Statement? The RBNZ releases its June Monetary Policy Statement (MPS) on 12 June. While a 25bp rate hike is widely anticipated and close to fully priced, there is some conjecture that the Bank may signal that a pause in the tightening cycle is likely – though we expect the Bank to keep the option of a July move open. Given the divergence that has opened up between market pricing and the rate track the RBNZ published in its March MPS, the market will be hoping for some indication that the RBNZ has wound back its rate intentions at least somewhat. By the end of 2015 the difference between market pricing and the RBNZ’s March forecast is now a bit more than 50bp.

David Plank

Macro strategist (+61) 2 8258-1475 [email protected]

6 June 2014



90 day bank bill rate: RBNZ vs market pricing

2.25

2.50

2.75

3.00

3.25

3.50

3.75

4.00

4.25

4.50

4.75

5.00

5.25

5.50

Dec-13 Mar-14 Jun-14 Sep-14 Dec-14 Mar-15 Jun-15 Sep-15 Dec-15 Mar-16 Jun-16

Mkt pricing for 3M rate as at 5 June 2014

RBNZ Mar-14 MPS

Source: Deutsche Bank, RBNZ, Bloomberg Financial LP

In his recently published preview of the MPS our Chief Economist for New Zealand expects the RBNZ will project cumulative tightening after June of around 125bp by the end of 2015. This will be down by around one rate hike from that projected in March, but still more than the market is pricing. This suggests that there isn’t a lot of scope for the NZ front-end to rally on the MPS, except at the very front of the curve, unless the market sees more information content in the direction of the RBNZ’s shift (less rate hikes) than in the projected level of hikes.

End 2015 ahead pricing for the 3M rate

4.0

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8


3M rate implied by Dec-15 NZD bill futre

Source: Deutsche Bank, Bloomberg Financial LP

The NZ front-end has been scaling back the expected extent of rate hikes since early April. We think the global rally explains some of this gain, but there are domestic factors as well such as falling dairy prices and evidence that the RBNZ’s macro-prudential policies have slowed the housing market. On the other hand, the GDP data to be published in a couple of weeks seems likely to show growth of well above 1%qoq in the first quarter of 2014.

We think market pricing has reached the point where it will take clear evidence of domestic weakness for the rally to continue. This has us expecting the US long-end to become an even more dominant influence on the long-end of the NZ curve. Given that we think the US long-end will finish the year quite a bit higher than its current level this means we expect a steeper NZ yield curve by year-end.

6 June 2014



2Y/10Y NZD swap curve vs 10Y UST

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

70

90

110

130

150

170

Jan-12 Apr-12 Jul-12 Oct-12 Jan-13 Apr-13 Jul-13 Oct-13 Jan-14 Apr-14

2Y/10Y NZD swap slope, bp (LHS)

10Y UST (RHS)

Source: Deutsche Bank, Reuters

Interestingly, the recent flattening in the NZ curve has run well ahead of what we might have expected given the rally in the 10Y UST. The fattening since mid-February reflects the fact the 2Y NZD swap rate has risen somewhat as the RBNZ has hiked rates, while the 10Y swap has rallied on the back of the global long-end rally. This means the typical directional relationship between the curve and the RBNZ’s official cash rate (OCR) has been maintained, even if the level relationship is different from the past.

OCR vs the NZ swap curve

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0-100

-50

0

50

100

150

200

250

300

Apr-99 Feb-01 Dec-02 Oct-04 Aug-06 Jun-08 Apr-10 Feb-12 Dec-13

2Y/10Y NZD swap curve, bp (LHS)

RBNZ OCR, inverted (RHS)

Source: Deutsche Bank, Bloomberg Financial LP

We think we are at the point where the direction of the US long-end will again become more important for the curve, assuming we are right that the RBNZ is set to pause its tightening cycle after the expected rate hike on Thursday. With this in mind we recommend the 2Y/10Y steepener in NZ swaps. The biggest risk to the trade is the RBNZ does not pause in hiking the OCR. A strong rally in the US long-end will also hurt the trade.

6 June 2014



AOFM signals next curve extension could be via nominal bond rather than a linker On 3 June Rob Nichol, the head of the Australian Office of Financial Management (AOFM), spoke at an Australian Business Economists lunch on the question of whether “the 2014 Budget and fiscal policy regime set a new course for the CGS market”. Mr Nichol took the opportunity provided by the speech to provide some commentary about the possible issuance strategy for the AOFM in 2014/15. We think the most important things to note in this regard are the following:

— While in the recent past the AOFM has led the extension of the yield curve via long dated linker issuance, they no longer feel the need to do so. Indeed, Mr Nichol said “I wouldn’t assume any longer that our longest line will always be inflation-linked” and, further, that maintaining the linker curve beyond 20 years “isn’t a high priority for our portfolio management objectives”.

— Given the focus on maintaining a 20Y nominal benchmark, Mr Nichol indicated the AOFM may “issue slightly longer than 20 years…..over the coming 12 months”.

— A new 2019 nominal bond will be issued in the coming fiscal year.

— After announcing some 18 months ago that there might be times when T-note supply was zero, the AOFM has reviewed the underlying demand for the instrument and “will gradually move back towards a greater reliance on Treasury notes.” This is primarily as a “service component to a core part of our investor base.” The AOFM has not, however, committed to a specific floor on the amount of T-notes outstanding.

We think the primary surprise for investors will be the indication that the linker curve may not be extended beyond the existing 2035 line in 2014/15. We commented on this in the Inflation Monthly published on 5 June where we recommended a break-even steepener.

6 June 2014



Global

Economics Rates Gov. Bonds & Swaps Inflation

Global Inflation Update

Global Breakeven performance was mixed this week, with valuations mostly higher in EUR and GBP and lower in USD (chart 1). TIPS B/Es have proved largely independent of trends in Treasuries over the past few weeks, with real-nominal yield betas close to one. News about inflation drivers were mixed this week, with data trends generally supportive, especially in the US, but commodities—energy, industrial metals and agriculture—lower, which helped B/E curves steepen in GBP and EUR.

The overall macro backdrop remains supportive for B/Es in our view, although in EUR and GBP a weak trend in spot inflation remains a headwind. The trend in forward-looking inflation indicators has however been positive, and points to some upside for B/Es in our models. This is particularly true in the US, where both the momentum in drivers and the level of indicators are a positive. Economic activity was weak in Q1, but data momentum appears to have been picking up into Q2, with business surveys or jobless claims (chart 2) improving. This together with somewhat higher industrial metal prices, signs of a pick-up in spot inflation as well as low risk aversion translates into a positive signal in our USD B/E score models (chart 4).

Relating the level of B/Es to demand, cost and risk proxies also would point to some upside for B/Es (see chart 3 for 10y TIPS as well as ‘EUR’ and ‘GBP’ sections below). ISM price balances rose to the highest since autumn 2012 in May and the trend in labour market (May payrolls report not released at the time of writing), backlogs or new orders indicators has remained positive, pointing to at least 15bp cheapness in 10y TIPS B/Es (chart 3).

Across markets, as discussed last week, GBP/USD B/E spreads had started to look too tight relative to trends in fundamental drivers. In 10y, B/E spreads widened by more than 10bp and we would turn neutral again.

3. USD B/Es cheap against macro backdrop 4. And driver momentum positive

0.5

1.0

1.5

2.0

2.5

3.0

1999 2001 2003 2005 2007 2009 2011 2013

USD BEI 10y

fitted, f(cost, demand & risk indicators)

-3

-2

-1

0

1

2

3

-0.7

-0.5

-0.3

-0.1

0.1

0.3

0.5

0.7

Jan-10 Jul-10 Jan-11 Jul-11 Jan-12 Jul-12 Jan-13 Jul-13 Jan-14

10y B/E, TIPS, 4w chge

Composite Score (rhs)


1. B/Es mixed

-10

-8

-6

-4

-2

0

2

4

6

8

5y 10y 20y 30y

EUR FRF

GBP USD

1w change in BEI on 5-Jun, carry adj., bp

2. USD B/Es v jobless claims -40

-30

-20

-10

0

10

20

30

40

50-250

-200

-150

-100

-50

0

50

100

150

200

1999 2001 2003 2005 2007 2009 2011 2013

10y TIPS B/E 20w change

jobless claims (rhs) 20w change



Markus Heider


6 June 2014



EUR View: Forward-looking inflation indicators are improving, and the ECB is offering more explicit verbal support for CPI expectations, but the economic effect of the policy measures remains uncertain and low spot inflation remains a headwind for B/Es. In that context, the tactical case for long B/Es is not obvious, but given low valuations, we see continue to see medium-term upside. On the BTPei curve, the BTPei16 looks attractive.

Rationale: Spot inflation trends remain a headwind for B/Es, with Eurostat’s May HICP flash estimate this week again surprising to the downside. Downward pressure has recently mostly come from food and while we expect some recovery from H2, further declines in y/y food inflation can be expected in the coming months. Core inflation also fell again in May. In part this was due to monthly volatility in German travel prices (and should be reversed in June), but core goods inflation has recently remained lower than suggested by PPI. Overall evidence remains consistent with the view that core inflation has bottomed around the turn of the year, but momentum so far has remained somewhat lower than we were projecting. With commodity prices falling and CPIs surprising to the downside, front-end B/Es underperformed this week (chart 1). Forward-looking indicators on the other hand improved somewhat, with composite PMI employment, input and output price indices rising in May and downstream PPI inflation recovering in April. On balance, judging by past relationships, current trends in cost and demand indicators would appear consistent with B/Es well above current levels, even taking into account the unusually low level of spot inflation (chart 3). Moreover, while the net economic impact of the measures announced remains uncertain, the inflation market has reacted positively to the ECB’s more explicit support of inflation expectations, despite the currency strengthening slightly on Thursday. B/Es in general (chart 1) and 5y5y forward B/Es in particular (chart 2) rose. At least, so far, this contrasts with the weakness in 5y5y forward B/Es often seen in the aftermath of ECB meetings since last November (chart 2). On balance, the near-term upside for B/Es may be limited by the still weak spot inflation momentum, but we continue to see scope for normalization in the medium-term. In this context, we would have a preference for 10y over 5y B/Es, while LTRO extension to 4y could continue to support 5y real yields.

RV: We find that the BTPei16 offers value on the BTPei curve. It’s about 60bp cheap relative to HICP forecasts, which is more than for the BTPei17 and -18 (chart 4), while on the German and French ILB curves, 2018 issues are cheaper compared to HICP forecasts than shorter maturities. The BTPei16 is also the cheapest BTPei when compared to HICP swaps (around 30bp cheaper; chart 4). The BTPei21 would seem relatively rich (chart 4).

3. 10y B/Es pricing in downside risks 4. BTPei 2016 relatively cheap, -21 relatively rich

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

2003 2005 2007 2009 2011 2013

OATei BEI 10y

fitted, f(cost, demand, spot inflation & risk indicators)

-70

-60

-50

-40

-30

-20

-10

0

BTP

ei1

6

BTP

ei1

7

BTP

ei1

8

BTP

ei1

9

BTP

ei2

1

BTP

ei2

3

BTP

ei2

4

BTP

ei2

6

BTP

ei3

5

BTP

ei4

1

Rich (+)/cheap (-) v HICP forecasts

Rich (+)/cheap (-) v HICP swaps


1. EUR B/Es beyond front-end widen

-20

-15

-10

-5

0

5

10

DB

Rei

16

OB

Lei1

8

DB

Rei

20

DB

Rei

23

DB

Rei

30

OA

Tei1

8

OA

Tei2

0

OA

Tei2

2

OA

Tei2

4

OA

Tei2

7

OA

Tei3

2

OA

Tei4

0

BTP

ei16

BTP

ei19

BTP

ei24

BTP

ei26

BTP

ei41

SPG

ei24

BEI

Real Yld

Nom Yld

1w change on 5-Jun, carry adj., bp

2. 5y5y B/Es rise

1.85

1.95

2.05

2.15

2.25

2.35

2.45

Apr-13 Jun-13 Aug-13 Oct-13 Dec-13 Feb-14 Apr-14 Jun-14

DBRei 5y5y (implied ZC) B/E

HICP swap 5y5y

ECB dates



6 June 2014



GBP View: Macro trends remain supportive for B/Es, but relatively low spot price and wage inflation trends may limit the upside for 5y and 10y B/Es for now. With the B/E curve flat and 30y RPI cheap compared to one-year averages, we would see upside for 30y swaps. In RV, we find 20y cash B/Es rich, both v 30y cash and 20y swaps.

Rationale: B/Es have rebounded from last week’s lows, and real yields have risen this week. Despite some softening from strong levels in some leading indicators of housing market activity, overall economic data appear to have been consolidating at levels consistent with above trend economic growth. This week’s May PMIs, for example, suggest that GDP growth could pick up further in Q2, with employment indices staying at record highs (chart 1). Price and cost indicators remain somewhat more mixed; PMI input price balances rose again, while composite output prices fell. On balance, recent trends in cost and demand indicators would point to some upside for B/Es (chart 3), but signs of rising spot price or wage inflation may be required for shorter-end B/Es to rise significantly from here. We see RPI inflation staying at 2.5% in May and next week’s labour market report will show whether the consistent strength in leading indicators has started to translate into higher official earnings growth in April; a strong base effect will put downward pressure on y/y wage growth however. The B/E curve is flat relative to past averages, and 30y RPI remains well below one-year averages (chart 2). With June being a month which has historically seen relatively strong long-end inflation demand, the macro backdrop supportive and valuations relatively low, we would have a positive bias on 30y swaps, despite still low real yields.

RV: We find 20y cash B/Es expensive on the curve and rich relative to swaps. 20y30y cash B/E spreads for example look relatively tight compared to past averages, and compared to swaps (chart 4). 5y B/Es 15y forward and 10y B/Es 10y forward remain above 4%. A 30y UKTi syndication in July may be a risk, but we would prefer 30y over 20y B/Es.

3. Macro trends supportive for B/Es 4. 20y30y UKTi B/E spread tight

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

2001 2003 2005 2007 2009 2011 2013

GBP BEI 5y

fitted, f(cost, demand, spot inflation & risk indicators)

0

5

10

15

20

25

30

35

40

45

50

-5

0

5

10

15

20

25

Jan-10 Aug-10 Mar-11 Oct-11 May-12 Dec-12 Jul-13 Feb-14

swaps

cash (rhs)

20y30y B/E spread


1. PMI employment still at highs

38

43

48

53

58

-1.0

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

2001 2003 2005 2007 2009 2011 2013 2015

GBP AWE reg pay, priv, % y/y

Emplyt PMI, 9m lead GBP (rhs)

2. RPI curve flat, 30y cheap

-1.2

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

5Y 10Y 20Y 30Y

3m 6M 1Y

GBP RPI swaps: z-scores



6 June 2014



United States


Inflation-Linked

We highlight inflation relative value opportunities. TIPS breakevens have underperformed inflation swaps recently. Investors who have a choice to be long inflation should consider TIPS breakevens.

In the five-year sector, we recommend selling the rich 1.875s of 7/2019 to buy the cheap and more liquid 0.125s of 4/2019. In the ten-year sector, the 0.125s of 1/2023 look attractive to neighboring issues.

In supply, we expect the Treasury to announce a $7 billion reopening of the 1.375s of 2/15/2044 TIPS on Thursday, June 12. The auction will be on Thursday, June 19. Our analysis shows 30-year TIPS breakevens tend to decline ahead of supply.

Inflation RV opportunities abound

Although inflation breakevens have traded in a narrow range over the past couple of months, there has been no lack of relative value opportunities. We highlight a few in this report.

TIPS breakevens have underperformed inflation swaps recently. In the ten-year sector, for example, the spread between TIPS breakevens and inflation swaps was around 26bp in late May, and now trades about 31bp. The underperformance can be better seen in regressions, where both the five- and the ten-year TIPS breakevens appear low relative to inflation swaps on recent data. Therefore, investors who have a choice to be long inflation should consider TIPS breakevens.

10yr TIPS breakevens have underperformed 10yr inflation swaps recently

y = 1.0281x - 0.3588R² = 0.7963

2.10

2.12

2.14

2.16

2.18

2.20

2.22

2.24

2.42 2.44 2.46 2.48 2.50 2.52 2.54

10yr

TIP

S br

eake

vens

10yr inflation swaps

past two months

6/5/2014

Source: Bloomberg FinanceeLP and Deutsche Bank

Alex Li


6 June 2014



5yr TIPS breakevens have underperformed 5yr inflation swaps recently

y = 1.2033x - 0.6513R² = 0.8222

1.80

1.85

1.90

1.95

2.00

2.05

2.05 2.10 2.15 2.20 2.25

5yr

TIPS

bre

akev

ens

5yr inflation swaps

past two months

6/5/2014

Source: Bloomberg Finance LP and Deutsche Bank

In the five-year sector, we recommend selling the rich 1.875s of 7/2019 to buy the cheap and more liquid 0.125s of 4/2019. This relative value opportunity is evident in our spline model. The 1.875s are a seasoned ten-year TIPS with $15 billion outstanding, whereas the 0.125s are the on-the-run five-year with a larger $18 billion size. The latter offers better deflation protection, although we don’t expect cumulative deflation over the next five years under dovish Fed policies.

Sell the rich 1.875s of 7/2019 TIPS to buy cheap and more liquid 0.125s of

4/2019 TIPS


In the ten-year sector, the 0.125s of 1/2023 look cheap. The 1/2022-1/2023-1/2024 spread is in the top end of the range (from about 2bp to 5.5bp) for the past few months. We’d have an overweight on that cheap issue.

In supply, we expect the Treasury to announce a $7 billion reopening of the 1.375s of 2/15/2044 TIPS on Thursday, June 12. The auction will be on Thursday, June 19. Our analysis shows 30-year TIPS breakevens tend to decline ahead of supply. On average, the 30-year TIPS breakevens drops about 5bp in the two-week period leading to an auction. The October 2012 auction cycle was an exception.

6 June 2014



1/22-1/23-1/24 real yield spread

1

2

3

4

5

6

7

1/25/14 2/25/14 3/25/14 4/25/14 5/25/14

Spread 50%-50% weighted

Source: Bloomberg Finance LP and Deutsche Bank

30yr TIPS breakevens tend to decline ahead of supply


6 June 2014



Contacts Name Title Telephone Email

EUROPE

Francis Yared Head of European Rates Research 44 20 7545 4017 [email protected]

Alexander Düring Euroland & Japan RV 44 207 545 5568 [email protected]

Markus Heider Global Inflation Strategy 44 20 754 52167 [email protected]

Bernd Volk Covered Bonds/SSA 41 44 227 3710 [email protected]

Jerome Saragoussi Global RV & Rates Vol 33 1 44 95 64 08 [email protected]

Abhishek Singhania Euroland Strategy/ EUR Govt. bonds 44 20 754 74458 [email protected]

Soniya Sadeesh UK Strategy & Money Markets 44 20 7547 3091 [email protected]

Nick Burns Credit Strategy 44 20 7547 1970 [email protected]

Stephen Stakhiv Credit Strategy 44 20 7545 2063 [email protected]

Sebastian Barker Credit Strategy 44 20 754 71344 [email protected]

Conon O’Toole ABS Strategy 44 20 7545 9652 [email protected]

Paul Heaton ABS Strategy 44 20 7547 0119 [email protected]

Rachit Prasad ABS Strategy 44 20 7547 0328 [email protected]

US

Dominic Konstam Global Head of Rates Research 1 212 250 9753 [email protected]

Steven Abrahams Head of MBS & Securitization Research 1-212-250-3125 [email protected]

Aleksandar Kocic US Rates & Credit Strategy 1 212 250 0376 [email protected]

Alex Li US Rates & Credit Strategy 1 212 250 5483 [email protected]

Richard Salditt US Rates & Credit Strategy 1 212 250 3950 [email protected]

Stuart Sparks US Rates & Credit Strategy 1 212 250 0332 [email protected]

Daniel Sorid US Rates & Credit Strategy 1 212 250 1407 [email protected]

Steven Zeng US Rates & Credit Strategy 1 212 250 9373 [email protected]

ASIA PACIFIC

David Plank Head of APAC Rates Research 61 2 8258 1475 [email protected]

Makoto Yamashita Japan Strategy 81 3 5156 6622 [email protected]

Kenneth Crompton $ bloc RV 61 2 8258 1361 [email protected]

Sameer Goel Head of Asia Rates & FX Research 65 6423 6973 [email protected]

Linan Liu Asia Strategy 852 2203 8709 [email protected]

Arjun Shetty Asia Strategy 65 6423 5925 [email protected]

Kiyong Seong Asia Strategy 852 2203 5932 [email protected]

CROSS-MARKETS

George Saravelos Head of European FX and cross markets strategy

44 20 754 79118 [email protected]


6 June 2014



Appendix 1

Important Disclosures Additional information available upon request For disclosures pertaining to recommendations or estimates made on securities other than the primary subject of this research, please see the most recently published company report or visit our global disclosure look-up page on our website at http://gm.db.com/ger/disclosure/DisclosureDirectory.eqsr 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. Francis Yared/Dominic Konstam

6 June 2014



Regulatory Disclosures

1. Important Additional Conflict Disclosures Aside from within this report, important conflict disclosures can also be found at https://gm.db.com/equities under the "Disclosures Lookup" and "Legal" tabs. Investors are strongly encouraged to review this information before investing.

2. Short-Term Trade Ideas

Deutsche Bank equity research analysts sometimes have shorter-term trade ideas (known as SOLAR ideas) that are consistent or inconsistent with Deutsche Bank's existing longer term ratings. These trade ideas can be found at the SOLAR link at http://gm.db.com.

3. Country-Specific Disclosures

Australia and New Zealand: This research, and any access to it, is intended only for "wholesale clients" within the meaning of the Australian Corporations Act and New Zealand Financial Advisors Act respectively. Brazil: The views expressed above accurately reflect personal views of the authors about the subject company(ies) and its(their) securities, including in relation to Deutsche Bank. The compensation of the equity research analyst(s) is indirectly affected by revenues deriving from the business and financial transactions of Deutsche Bank. In cases where at least one Brazil based analyst (identified by a phone number starting with +55 country code) has taken part in the preparation of this research report, the Brazil based analyst whose name appears first assumes primary responsibility for its content from a Brazilian regulatory perspective and for its compliance with CVM Instruction # 483. EU countries: Disclosures relating to our obligations under MiFiD can be found at http://www.globalmarkets.db.com/riskdisclosures. Japan: Disclosures under the Financial Instruments and Exchange Law: Company name - Deutsche Securities Inc. Registration number - Registered as a financial instruments dealer by the Head of the Kanto Local Finance Bureau (Kinsho) No. 117. Member of associations: JSDA, Type II Financial Instruments Firms Association, The Financial Futures Association of Japan, Japan Investment Advisers Association. This report is not meant to solicit the purchase of specific financial instruments or related services. We may charge commissions and fees for certain categories of investment advice, products and services. Recommended investment strategies, products and services carry the risk of losses to principal and other losses as a result of changes in market and/or economic trends, and/or fluctuations in market value. Before deciding on the purchase of financial products and/or services, customers should carefully read the relevant disclosures, prospectuses and other documentation. "Moody's", "Standard & Poor's", and "Fitch" mentioned in this report are not registered credit rating agencies in Japan unless "Japan" or "Nippon" is specifically designated in the name of the entity. Malaysia: Deutsche Bank AG and/or its affiliate(s) may maintain positions in the securities referred to herein and may from time to time offer those securities for purchase or may have an interest to purchase such securities. Deutsche Bank may engage in transactions in a manner inconsistent with the views discussed herein. Qatar: Deutsche Bank AG in the Qatar Financial Centre (registered no. 00032) is regulated by the Qatar Financial Centre Regulatory Authority. Deutsche Bank AG - QFC Branch may only undertake the financial services activities that fall within the scope of its existing QFCRA license. Principal place of business in the QFC: Qatar Financial Centre, Tower, West Bay, Level 5, PO Box 14928, Doha, Qatar. This information has been distributed by Deutsche Bank AG. Related financial products or services are only available to Business Customers, as defined by the Qatar Financial Centre Regulatory Authority. Russia: This information, interpretation and opinions submitted herein are not in the context of, and do not constitute, any appraisal or evaluation activity requiring a license in the Russian Federation. Kingdom of Saudi Arabia: Deutsche Securities Saudi Arabia LLC Company, (registered no. 07073-37) is regulated by the Capital Market Authority. Deutsche Securities Saudi Arabia may only undertake the financial services activities that fall within the scope of its existing CMA license. Principal place of business in Saudi Arabia: King Fahad Road, Al Olaya District, P.O. Box 301809, Faisaliah Tower - 17th Floor, 11372 Riyadh, Saudi Arabia. United Arab Emirates: Deutsche Bank AG in the Dubai International Financial Centre (registered no. 00045) is regulated by the Dubai Financial Services Authority. Deutsche Bank AG - DIFC Branch may only undertake the financial services activities that fall within the scope of its existing DFSA license. Principal place of business in the DIFC: Dubai International Financial Centre, The Gate Village, Building 5, PO Box 504902, Dubai, U.A.E. This information has been distributed by Deutsche Bank AG. Related financial products or services are only available to Professional Clients, as defined by the Dubai Financial Services Authority.

6 June 2014



Risks to Fixed Income Positions Macroeconomic fluctuations often account for most of the risks associated with exposures to instruments that promise to pay fixed or variable interest rates. For an investor that is long fixed rate instruments (thus receiving these cash flows), increases in interest rates naturally lift the discount factors applied to the expected cash flows and thus cause a loss. The longer the maturity of a certain cash flow and the higher the move in the discount factor, the higher will be the loss. Upside surprises in inflation, fiscal funding needs, and FX depreciation rates are among the most common adverse macroeconomic shocks to receivers. But counterparty exposure, issuer creditworthiness, client segmentation, regulation (including changes in assets holding limits for different types of investors), changes in tax policies, currency convertibility (which may constrain currency conversion, repatriation of profits and/or the liquidation of positions), and settlement issues related to local clearing houses are also important risk factors to be considered. The sensitivity of fixedincome instruments to macroeconomic shocks may be mitigated by indexing the contracted cash flows to inflation, to FX depreciation, or to specified interest rates - these are common in emerging markets. It is important to note that the index fixings may -- by construction -- lag or mis-measure the actual move in the underlying variables they are intended to track. The choice of the proper fixing (or metric) is particularly important in swaps markets, where floating coupon rates (i.e., coupons indexed to a typically short-dated interest rate reference index) are exchanged for fixed coupons. It is also important to acknowledge that funding in a currency that differs from the currency in which the coupons to be received are denominated carries FX risk. Naturally, options on swaps (swaptions) also bear the risks typical to options in addition to the risks related to rates movements.

GRCM2014PROD032228

David Folkerts-Landau

Group Chief Economist Member of the Group Executive Committee

Guy Ashton

Global Chief Operating Officer Research

Marcel Cassard Global Head

FICC Research & Global Macro Economics

Richard Smith and Steve Pollard Co-Global Heads Equity Research

Michael Spencer Regional Head

Asia Pacific Research

Ralf Hoffmann Regional Head

Deutsche Bank Research, Germany

Andreas Neubauer Regional Head

Equity Research, Germany

Steve Pollard Regional Head

Americas Research

International Locations

Deutsche Bank AG Deutsche Bank Place Level 16 Corner of Hunter & Phillip Streets Sydney, NSW 2000 Australia Tel: (61) 2 8258 1234

Deutsche Bank AG Große Gallusstraße 10-14 60272 Frankfurt am Main Germany Tel: (49) 69 910 00

Deutsche Bank AG Filiale Hongkong International Commerce Centre, 1 Austin Road West,Kowloon, Hong Kong Tel: (852) 2203 8888

Deutsche Securities Inc. 2-11-1 Nagatacho Sanno Park Tower Chiyoda-ku, Tokyo 100-6171 Japan Tel: (81) 3 5156 6770

Deutsche Bank AG London 1 Great Winchester Street London EC2N 2EQ United Kingdom Tel: (44) 20 7545 8000

Deutsche Bank Securities Inc. 60 Wall Street New York, NY 10005 United States of America Tel: (1) 212 250 2500

Global Disclaimer The information and opinions in this report were prepared by Deutsche Bank AG or one of its affiliates (collectively "Deutsche Bank"). The information herein is believed to be reliable and has been obtained from public sources believed to be reliable. Deutsche Bank makes no representation as to the accuracy or completeness of such information.

Deutsche Bank may engage in securities transactions, on a proprietary basis or otherwise, in a manner inconsistent with the view taken in this research report. In addition, others within Deutsche Bank, including strategists and sales staff, may take a view that is inconsistent with that taken in this research report.

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 recipient thereof in the event that any opinion, forecast or estimate set forth herein, changes or subsequently becomes inaccurate. Prices and availability of financial instruments are subject to change without notice. This report is provided for informational purposes only. It is not an offer or a solicitation of an offer to buy or sell any financial instruments or to participate in any particular trading strategy. Target prices are inherently imprecise and a product of the analyst judgement. The financial instruments discussed in this report may not be suitable for all investors and investors must make their own informed investment decisions. Stock transactions can lead to losses as a result of price fluctuations and other factors. If a financial instrument is denominated in a currency other than an investor's currency, a change in exchange rates may adversely affect the investment. Past performance is not necessarily indicative of future results. Deutsche Bank may with respect to securities covered by this report, sell to or buy from customers on a principal basis, and consider this report in deciding to trade on a proprietary basis.

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 theinvestors' 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.theocc.com/components/docs/riskstoc.pdf . 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.

The risk of loss in futures trading and options, foreign or domestic, can be substantial. As a result of the high degree of leverage obtainable in futures and options trading losses may be incurred that are greater than the amount of funds initially deposited.

Unless governing law provides otherwise, all transactions should be executed through the Deutsche Bank entity in the investor's home jurisdiction. In the U.S. this report is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this report is approved and/or communicated by Deutsche Bank AG Frankfurt authorized by the BaFin. In the United Kingdom this report is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange and regulated by the Financial Conduct Authority for the conduct of investment business in the UK and authorized by the BaFin. This report is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. This report is distributed in Singapore by Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch (One Raffles Quay #18-00 South Tower Singapore 048583, +65 6423 8001), and recipients in Singapore of this report are to contact Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch in respect of any matters arising from, or in connection with, this report. Where this report is issued or promulgated in Singapore to a person who is not an accredited investor, expert investor or institutional investor (as defined in the applicable Singapore laws and regulations), Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch accepts legal responsibility to such person for the contents of this report. 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 © 2014 Deutsche Bank AG

global fixed income weekly - dws...6 june 2014 global fixed income weekly deutsche bank ag/london...

Documents