Deutsche Bank Markets Research Global

Rates Credit

Date 4 April 2014 Francis Yared

Global Fixed Income Weekly

Strategist (+44) 020 754-54017 [email protected] Dominic Konstam Research Analyst (+1) 212 250-9753 [email protected]

„

The ECB raised the stakes with a fairly aggressive verbal intervention. However, a more dovish ECB remains unevenly priced

„

On some metrics, 10Y real rates are already pricing an ECB QE somewhere in between QE2 and QE infinity. This makes the outright duration call less obvious, especially that inflation breakevens have scope to widen

„

We favour trades that are attractive from a valuation perspective and would benefit from a more aggressive ECB: 1Y forward EUR2s30s steepeners, EUR5Y breakeven wideners, 10Y BTP ASW tighteners and short euro

„

Tactically, US rates are close to faire value. Strategically, there is scope for further repricing of the pace of normalization of monetary policy with a medium term target of 2% on June 16 Fed funds

„

The risk is that markets stop challenging guidance based on the Fed’s forecasts, and instead start challenging the Fed’s forecasts given the guidance.

„

Such a dynamic would be consistent with our “slow and low” market scenario whereby longer forwards, say 5y5y or longer, rally to reflect the perception that potential growth is something closer to 2.0% rather than 3%. In the short run this should allow 10s to fall to 2.5% and 5y5y to 3.5% or lower.

„

„

Future ECB easing measures depend in part on the economic picture in the US. A Fed exit could allow the ECB to passively target the currency simply by retaining dovish guidance and full allotment. A weaker US growth picture could reduce the rate differential and necessitate ECB asset purchases. We continue to find US/Europe cross market trades compelling because of this tacit linkage – falling US forwards could be seen to increase the probability of rising EUR forwards!

Table of contents Bond Market Strategy

Page 02

US Overview

Page 08

Treasuries

Page 12

Derivatives

Page 19

Agencies

Page 23

Mortgages

Page 25

US Credit Strategy

Page 43

Credit Strategy: The Callable Short Life

Page 47

European ABS update

Page 50

Covered Bond and Agency Update

Page 51

UK Strategy

Page 53

Japan Strategy

Page 56

Global Relative Value

Page 60

Asia

Page 66

Dollar Bloc Strategy

Page 71

Global Inflation Update

Page 82

Inflation Linked

Page 86

________________________________________________________________________________________________________________ Deutsche Bank AG/London DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1. MICA(P) 054/04/2013.

4 April 2014 Global Fixed Income Weekly

Global

Rates Gov. Bonds & Swaps Rates Volatility

Bond Market Strategy „

The ECB raised the stakes with a fairly aggressive verbal intervention. However, a more dovish ECB remains unevenly priced

„

On some metrics, 10Y real rates are already pricing an ECB QE somewhere in between QE2 and QE infinity. This makes the outright duration call less obvious, especially that inflation breakevens have scope to widen

„

We favour trades that are attractive from a valuation perspective and would benefit from a more aggressive ECB: 1Y forward EUR2s30s steepeners, EUR5Y breakeven wideners, 10Y BTP ASW tighteners and short euro

„

Francis Yared Strategist (+44) 020 754-54017 [email protected] Soniya Sadeesh Strategist (+44) 0 207 547 3091 [email protected] Jerome Saragoussi Strategist (+33) 1 4495-6408 [email protected] George Saravelos Strategist (+44) 20 754-79118 [email protected]

Tactically, US rates are close to faire value. Strategically, there is scope for further repricing of the pace of normalization of monetary policy with a medium term target of 2% on June 16 Fed funds

Raising the stakes The view: Tactically US rates are close to fair value, but we do maintain a strategic bearish bias in the front-end of the US curve. In Europe, we continue to favour trades that are attractive from a valuation perspective and would benefit from a more aggressive ECB: steepeners in core rates, tighter peripheral speads, wider breakevens and a weaker euro. The rationale: The ECB has raised the stakes with a fairly aggressive verbal intervention. More specifically, Draghi stepped up the dovish rhetoric with: (a) an explicit reference to the FX as a downside risk to inflation and (b) a stated unanimous agreement to adopt non-conventional measures (including QE) if inflation stayed low for longer than currently expected. This has lowered the bar for further intervention, but the practical hurdles have not changed. As Draghi himself recognized, there are technical difficulties in implementing private sector asset purchases and a US style QE may not be as effective in Europe. Ultimately, the key implicit target is the euro: it was formally included in the statement and Draghi reiterated that imported inflation played a significant role in the decline of inflation. Thus, the combination of verbal intervention by the ECB and better data in the US may turn out to be sufficient to help bring down the currency and achieve the ECB’s objectives. In any event, a more dovish ECB remains unevenly priced. On some metrics, 10Y real rates are already pricing a program somewhere in between QE2 and QE infinity. At the other end of the spectrum, inflation breakevens remain depressed relative to fundamentals, the euro too high relative to the interest rate differentials, and the 2s30s curve too flat. Peripheral spreads have benefited from the ECB’s stance, but Italy and Spain remain in the 150-200bp range which we had penciled in for this year even without explicit ECB support. From a risk reward perspective, we would continue to focus on trades that are attractive from a valuation standpoint and benefit if the ECB delivers more aggressive easing: steepeners in core rates (which could trade bearish), peripheral ASW tighteners, breakeven wideners and short euro. The rich valuations of core real rates and the scope for wider breakevens make the outright duration call less obvious for now. A better QE trade would be to receive conditionally 3M forward 5s10s30s which is more efficient from a beta adjusted carry perspective. Page 2

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

In the US, the market is now close to “short-term fair value” (ahead of NFP which was not released at the time of writing). The front-end is almost pricing a reasonably conservative interpretation of the FOMC’s guidance, with June 16 Fed funds close to 1.5%. Strategically, there is scope for a sell-off towards a longer term target of 2% for June 16 fed funds. However, such repricing is likely to require a confirmation of further normalization of wage inflation which we expect later this year. Finally, the USD5Y5Y which was too low last week, is now also close to current fair value of 4.2%.

Local disequilibrium in Europe We expected the ECB to talk the talk but not walk the walk. On Thursday, the ECB did not announce any new policy measure. However, Draghi shouted the talk with notably: (a) an explicit reference to the FX as a downside risk to inflation and (b) a stated unanimous agreement to adopt non-conventional measures (including QE) if inflation stayed low for longer than currently expected. The Governing Council has effectively endorsed and institutionalized the various statements made by ECB members in recent weeks (ranging from Draghi’s explicit reference to the FX in his Vienna speech, to Weidmann conceptual endorsement of QE). The fact that persistently low inflation is deemed to be enough to trigger a more aggressive response has lowered the bar for unconventional policies. However, the issues surrounding the unconventional measures have not disappeared. The ECB’s preference for private sector purchases was clear, but Draghi also recognized the technical limitation of such program. On the other hand, Draghi emphasized the institutional differences with the US, suggesting that a US style QE may not work as well as in the US. Also, the FX was designated as the chief culprit for the disinflation risks. Thus, a repricing of monetary policy expectations in the US and a verbal intervention in Europe (or even further strengthening of the forward guidance), may in fact be sufficient to meet the ECB’s objectives. Thus, even if the ECB’s willingness to do more has been verbally strengthened, it is not obvious that it will lead to a large scale government bond purchases. However, it is clear that the ECB will be increasingly aggressive (including asset purchases) if the euro appreciates or inflation expectations decline noticeably. From a market perspective, a more aggressive ECB is unevenly priced. At one end of the spectrum, core rates are largely pricing further easing. To illustrate this point, we have adapted our US QE models to the Eurozone. A US style QE would primarily impact 5Y5Y real rates and lead to a significant richening of the 10Y on the curve. To measure how much QE is priced in, we compare the 10Y real rates vs a model that incorporates data surprises and risk aversion. The deviation away from the model would be symptomatic of QE expectations. Currently the model indicates 100bp of richness of real rates. This compares with 10Y Tips which were 70bp too rich ahead of QE2 and 140bp too rich ahead of QE infinity (see graphs below).

Deutsche Bank AG/London

Page 3

4 April 2014 Global Fixed Income Weekly

The richness of 10Y German real yields is halfway between the richness of 10Y Tips ahead of QE2 and QE infinity 2.0 Inf luence of Fed QE inf inity and tapering

Inf luence of Fed QE2

1.5

ECB QE expectations building up

1.0

0.5

1.0

0.0

0.5

-0.5

0.0

-1.0

-0.5 -1.0 Jul-09

Germany 10y real y ield

Jan-11

Oct-11

QE2

-1.5

Model f (EU eco surprises, risk av ersion) Apr-10

Tapering

Jul-12

Apr-13

Jan-14

QE Dislocation of German 10y real y ield inf inity Dislocation US 10Y real y ield -2.0 Sep-09 Mar-10 Sep-10 Mar-11 Sep-11 Mar-12 Sep-12 Mar-13 Sep-13 Mar-14

Source: Deutsche Bank, Bloomberg Finance LP

The 5s10s30s fly is also a good proxy for US-style QE expectations. As can be seen in the graph below, the fly is at levels similar to the one observed in the US ahead of QE2, but a bit further from the richness observed ahead of QE infinity. The EUR5s1030s fly is close to levels observed in the US ahead of QE2

Source: Deutsche Bank

Peripheral bond markets have also clearly benefited from the ECB support. However, the current spread level is within the 150-200bp range that we had penciled in for 2014 even without explicit ECB backstop. However, the euro and breakevens are not yet reflecting a more aggressive ECB. Indeed, the euro remains too high vs. the real interest rate differential. Similarly, breakevens remain too low vs. the data, risk aversion and commodity prices (See graphs below). Finally, the curve models presented last week continue to indicate that the slope in Europe is too flat, which is consistent with the depressed breakevens currently priced in.

Page 4

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

But this is not reflected in breakevens nor the euro fitted based on US, EUR, JP & China busines surveys, core inflation and VIX

2.8 2.6

5y OATei implied ZC BEI

2.4

1.40 1.39

DB HICP fcst assuming oil futures

2.2 2.0 1.8 1.6

1.0

2Y real rate spread (rhs)

0.9

1.37

0.8

1.36

0.7

1.35

1.2

1.34

0.8 Jan-05

EURUSD currency

1.38

1.4

1.0

1.1

1.41

0.6 0.5

1.33 Jul-06

Jan-08

Jul-09

Jan-11

Jul-12

Jan-14

1.32 Nov -13

0.4 Dec-13

Jan-14

Feb-14

Mar-14

Apr-14

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

Given current pricing, we would continue to favour trades which are attractive from a valuation perspective and would benefit from a more aggressive ECB: short euro, long inflation breakevens, 1Y forward EUR2s30s steepeners and long peripheral spreads. The latter should be implemented on ASW to avoid the potential impact of a GDP-weighted US style QE which would disproportionately benefit the low debt-to-gdp countries such as Germany at the expense of the high debt-to-gdp countries such as Italy. These trades happen to coincide with the implicit or explicit objectives of the ECB: lower euro, higher inflation and tighter peripheral spreads. As for core rates, the risk reward for the outright duration view is not obvious. As discussed above, our models suggest that real rates are pricing something in between QE2 and QE infinity. These models are more art than science and disentangling what is exactly driving the richness of rates is difficult (forward guidance? QE expectations? deflation risks? portfolio shifts?). But the overall message is that long-term rates are already low. There is potential for real rates to rally more if the ECB delivers a very aggressive US-style QE, but this could be largely compensated by the widening of breakevens which are too low. In fact, this is what happened in the US and Japan between the preannouncement of the QE programs and their actual implementation (see table below). A better trade would be to receive conditionally 3M forward 5s10s30s which has scope to adjust and is more efficient from a beta adjusted carry perspective (see Global Relative Value for more details). Market reaction from QE pre-announcement to implementation 5Y real yield (bp) 5Y inflation breakeven (bp) 5Y nominal yield (bp)

US QE2

Japan

-49

-55

58

51

9

-5

Trade weighetd currency (%)

-6%

-12%

Domestic equity market (%)

14%

30%

-20

-59

Domestic IG credit spreads (bp)

(1) US: changes from 2010 Jackson Hole to QE2 announcement (2) Japan: changes from 2012 election to QE announcement Source: Deutsche Bank

Deutsche Bank AG/London

Page 5

4 April 2014 Global Fixed Income Weekly

Local equilibrium in the US As we await the release of NFP (not known at the time of writing), the US rates market is close to “short-term fair value” both in the front-end and 5Y5Y. From a strategic perspective, there is scope for more sell-off in the front-end of the curve, especially around Dec-16. But more evidence of wage inflation may be necessary for the market to challenge the Fed and/or the Fed to adapt its guidance. We define the “short-term fair value” of the front end of the curve as the market pricing 1%/2% for Dec-15 and Dec-16 Fed funds respectively. These targets are set to reflect a mildly conservative interpretation of the FOMC’s intention. The median of the FOMC’s dots were at 1% and 2.25%, but the center of the committee is probably 25bp lower. To account for a potential trade-off between the timing of the first hike and the pace of the hikes, we focus on June 16FF which is currently only 8bp below the implicit 1.5% target (see table below). This target is an intermediate one, which could hold until the market challenges the Fed and/or the Fed loosens its guidance. For now, the centre of the committee continues to argue that the slack in the labour market may be larger than what the unemployment rate suggests (e.g. this week’s speech by Yellen). Ultimately, wage inflation should settle the debate and confirmation of further normalization (which we expect later this year), should lead to a more material repricing of the timing and/or pace of policy normalisation. To quantify the extent of such potential repricing, we estimate a pre-crisis Taylor rule adjusting upward the unemployment rate forecast (based on a simple extrapolation of the current trend) for the drop in the participation rate which is deemed to be cyclical (about 1%). The logic of the exercise is to assume that the Fed is deviating from the pre-crisis Taylor rule mostly because of the implicit assumption that the unemployment rate underestimates the slack in the labour market. We also adjust the pre-crisis Taylor rule parameters to reflect a lower neutral real rate to 1.75% instead of 2% assuming the FOMC will continue to gradually nudge down its neutral rate estimate. Finally, we assume NAIRU at 5.4% (the mid-point of the FOMC’s central tendency), and the FOMC’s forecasts for core PCE. This exercise suggests a Fed Funds around 1% end of 2015 and 3% end of 2016, i.e. June 16 Fed funds at 2%. Note also that this analysis suggests that our focus on conditional bear flattener in the front-end of the curve was premature. USD 5Y5Y is now back at fair value

Short term and long term target in the front end

Market pricing

Time of first hike

Median of median FOMC dots of last FOMC forecasts

Median of Mar-2014 FOMC dots

Long term fair value based on pre crisis Taylor Rule

-3.5

3.0

Adjusted USD 5y 5y BRP

2.5

Subsequent 12-month change in US 5Y 5Y y ield (rhs)

2.0

-3.0 -2.5 -2.0

1.5

-1.5

1.0

-1.0

0.5

-0.5

0.0

0.0

Aug-15

Jun-15

Jun-15

Jun-15

-0.5

0.5

115bp/year

100bp/year

125bp/year

200bp/year

-1.0

1.0

FF in June 2016

1.42%

1.50%

1.63%

2.00%

-1.5

USD 1Y1Y rate

0.87%

0.98%

0.98%

1.05%

-2.0

USD 2Y1Y rate

2.02%

2.04%

2.24%

2.82%

-2.5

Pace of hikes

Source: Deutsche Bank, Bloomberg Finance LP, Haver

Page 6

1.5 2.0 2.5 90

92

94

96

98

00

02

04

06

08

10

12

14

Source: Deutsche Bank, Bloomberg Finance LP, Haver

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

To assess the long end valuation, we rely on our 5Y5Y adjusted bond risk premium model. Last week we highlighted the fact that for the first time since July 2013 it indicated that the 5Y5Y rate was too low relative to a fair value of 4.15-4.2%. The bear steepening that has occurred since has led the 5Y5Y close to fair value (see graph above). This would justify a more neutral stance further out the curve.

Deutsche Bank AG/London

Page 7

4 April 2014 Global Fixed Income Weekly

United States

Rates Gov. Bonds & Swaps Rates Volatility

US Overview „

„

„

„

The risk is that markets stop challenging guidance based on the Fed’s forecasts, and instead start challenging the Fed’s forecasts given the guidance. Such a dynamic would be consistent with our “slow and low” market scenario whereby longer forwards, say 5y5y or longer, rally to reflect the perception that potential growth is something closer to 2.0% rather than 3%. In the short run this should allow 10s to fall to 2.5% and 5y5y to 3.5% or lower. Future ECB easing measures depend in part on the economic picture in the US. A Fed exit could allow the ECB to passively target the currency simply by retaining dovish guidance and full allotment. A weaker US growth picture could reduce the rate differential and necessitate ECB asset purchases. We continue to find US/Europe cross market trades compelling because of this tacit linkage – falling US forwards could be seen to increase the probability of rising EUR forwards!

Dominic Konstam Research Analyst (+1) 212 250-9753 [email protected] Aleksandar Kocic Research Analyst (+1) 212 250-0376 [email protected] Alex Li Research Analyst (+1) 212 250-5483 [email protected] Stuart Sparks Research Analyst (+1) 212 250-0332 [email protected] Daniel Sorid Research Analyst (+1) 212 250-1407 [email protected] Steven Zeng, CFA Research Analyst (+1) 212 250-9373 [email protected]

Growth uncertainty should weigh on longer forwards The March employment number clearly disappointed the more aggressive forecasts and whisper numbers. The point however is not that weather payback didn’t materialize, but rather is that it remains unclear whether it did, didn’t, or ever will, leaving the market to question the vigor of a number that was to be fair reasonably near consensus. The risk for the current pricing paradigm of “slow and high” with respect to the timing of the beginning of rate hikes and the terminal rate for the cycle is that the market stops challenging guidance given the forecasts and starts challenging the forecasts given the guidance. If forecasts themselves come to be seen as a policy tool for optimistic signaling then the market might conclude it is waiting for Godot in a low potential growth, low inflation, low productivity, low rate equilibrium that the data suggest is very much intact.

Page 8

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 1: Payroll income growth remains well within the post crisis range 10%

5%

0%

-5% Payroll Income

-10% Retail Sales ex gas 3m/3m ann

-15% 2007-Nov

2009-Nov

2011-Nov

2013-Nov

Source: Deutsche Bank

The point of uncertainty in the market is whether the current curve environment resolves to flattening in a “fast and low” scenario where inflation perks up due to capacity constraints and the Fed must increase rates relatively soon, though to a lower terminal level, or if instead it resolves to a “slow and low” bullish flattening whereby low potential growth begins to pull the long forwards lower toward a lower equilibrium rate and the Fed has the luxury to remain accommodative in the absence of pressure from inflation. Either way the curve flattens, the former case is bearish the 5y and the latter is bullish 5y5y. As we have pointed out in recent weeks, it is plausible that NAIRU could actually be lower, not higher, even though potential growth is lower. This could be because of the demographics, with an older workforce and hence on a weighted basis a lower equilibrium unemployment rate, or perhaps due to skill deprecation and subsequent falling participation in an increasingly technology-oriented and service-driven economy. The effect at any rate is that the output gap could be bigger rather than smaller, which would go some way to fit the empirical facts on unemployment and observed inflation. Clearly this sort of scenario would be suggestive of a “slow and low” resolution to the currently still-steep curve. Our forecasts in the short run lean toward the slow and low resolution. We see the 10y yield falling to 2.5% by mid-year. This should allow the 5y5y to fall to 3.5% or even below. The emerging markets adjustment looks implausibly muted or more likely incomplete. China has slowed. Industrial metals prices and global trade volumes are not suggestive of a global acceleration. For the time being it remains unclear whether the US can be the engine to restart global growth, and it looks unlikely that other parts of the world can serve that role either. The system requires final demand to restart the engines.

Deutsche Bank AG/London

Page 9

4 April 2014 Global Fixed Income Weekly

Figure 3: IMF Global exports volume

10

100

40

8

80

30

60

4

40

2

20

10 y/y %

6

IMF Global exports y/y

20

%

%

Figure 2: Industrial metals and global GDP

0 -10

-30 Jan-13

Sep-11

Jan-09

May-10

Sep-07

Jan-05

May-06

Sep-03

Jan-01

May-02

Sep-13

Jul-12

Feb-13

Dec-11

Oct-10

May-11

Mar-10

Jan-09

Aug-09

Jun-08

Apr-07

Nov-07

Sep-06

Jul-05

Feb-06

Dec-04

Oct-03

May-04

Mar-03

Source: Haver Analytics, Bloomberg Finance LP

Sep-99

-60

-6

Jan-97

-40 May-98

-40

Sep-95

GSCI Ind Metals y/y (rhs)

Jan-93

Global GDP y/y

-4

-20

May-94

-20

Sep-91

-2

Jan-89

0

May-90

0

Source: Deutsche Bank

This becomes interesting in the context of Europe and the ECB. Our European economists reason that the ECB is unlikely to resort first to further unconventional measures such as purchases of private sector or public sector debt, but rather are likely to first effectively target the currency by retaining full allotment refunding and dovish guidance as the Fed begins its exit. If the Fed’s forecasts are realized and the front end follows the forwards then rate differentials can work increasingly to weaken the euro and provide a tail wind to capital goods exports from the European core. The risk to that strategy is that the Fed’s exit is longer in coming. While we think the taper will be completed, progressing to the point where SOMA reinvestment is halted and ultimately excess liquidity is drained and rates put higher will as Yellen re-assured us last week depend on more broad based improvements that thus far remain elusive. The credit impulse in Europe remains quite anemic, in fact though it is no longer the case that public sector credit appears to be crowding out private sector credit, clearly there remains a hesitancy to provide credit to Euro area households and businesses. There remains the problem that banks are expected to de-lever and de-risk, yet also increase credit provision. While the AQR could encourage banks to purge bad legacy assets, it does little to encourage new use of balance sheet. In this sense QE is more likely intended to ease financial conditions via lower borrowing costs to extremely attractive levels than it is to directly target higher inflation. Figure 5: Credit growth and nominal GDP

20 15 10 5 0

6.0%

6.0%

4.0%

4.0%

2.0%

2.0%

0.0%

0.0%

-2.0%

-2.0%

-4.0% Nominal Growth

-5

-4.0%

Jan-13

Aug-13

Jun-12

Apr-11

Nov-11

Sep-10

Jul-09

Feb-10

Dec-08

Oct-07

May-08

Mar-07

Jan-06

Aug-06

Jun-05

Apr-04

Source: Deutsche Bank

Nov-04

Sep-03

Feb-03

-10

Page 10

8.0%

General Gov't Noiminal Growth

Y/y credit outstanding, %

Private Sector

-6.0%

Chg Total Credit Growth (rhs)

-6.0%

Change in Total Credit Growth

Figure 4: EUR Credit growth to private and public sectors

-8.0%

Source: Deutsche Bank

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

From there the narrative is not dissimilar to that in the US. Increased investment should help induce a virtuous circle of higher productivity, wages, and consumption, and inflation should converge to target as activity accelerates.

7 6 5 4 3 2 1 0 -1 -2 -3 -4

US

Nov-13

Jul-12

Mar-13

Nov-11

Jul-10

Mar-11

Nov-09

Jul-08

Mar-09

Nov-07

Mar-07

Jul-06

Nov-05

Mar-05

Jul-04

Nov-03

Mar-03

Jul-02

Nov-01

Europe

Mar-01

Productivity y/y, %

Figure 6: Productivity growth in US and Europe

Source: Deutsche Bank

This leaves the ECB inactive for roughly a quarter or so with markets likely to fret that a policy error is in train and the ECB remains far too tight. This in turn is negative to global growth and on the margin is another headwind for the US. This, we think, is the elegance to US/Europe spread tighteners – falling US forwards should actually increase the probability that the ECB resorts to asset purchases, which in turn should steepen the EUR curve, bolster breakevens and help push EUR rates higher. And the longer the ECB and the rest of the market must wait for the US to accelerate, the more likely it is that the US forwards will fall. This class of trades works on both sides! Within this class we continue to favor receiving the spread outright in 1y10y or 5y5y, and harnessing the implied volatility differential between US and EUR via conditional bearish spread tighteners.

Deutsche Bank AG/London

Page 11

4 April 2014 Global Fixed Income Weekly

United States

Rates Gov. Bonds & Swaps

Alex Li

Steven Zeng, CFA

Research Analyst Research Analyst (+1) 212 250-5483 (+1) 212 250-9373 [email protected] [email protected]

Treasuries „

We highlight relative value among off-the-run 10s. The 2022 Treasuries have outperformed in the past two months, and now look rich. We recommend an underweight in that sector.

„

Fed custody holdings of Treasuries increased $13.2 billion this week. The data include the settlement of March month end coupon Treasury auctions. The change appears more normal than those in March and suggests foreign official demand for Treasuries.

„

The market rally after the non-farm payroll data on Friday was consistent with short positions in the market ahead of the data. Measures from COT, put/call ratio and SMR all pointed to short positions.

Sell the rich 2022 Treasuries We highlight relative value among off-the-run 10s. The 2022 Treasuries have outperformed in the past two months, and now look rich against neighboring issues. We recommend an underweight in that sector. The 1.625s of 11/2022 have a spread of -1.9bp through our spline curve. The 11/2021, 11/2022, and 11/2023 spread has moved from about +11bp in early February to the current +7bp. The spread appears too tight when regressed on the rate levels as well as on the curve slope. The spread has a higher correlation with the curve slope (R-squared at 81% on the data from mid November 2013 to the present); the belly tends to richen on flatteners. 2021-2022-2023 spread is too tight

Source: Deutsche Bank

Page 12

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

2021-2022-2023 spread versus rate level

2021-2022-2023 spread versus curve

Source: Deutsche Bank

Source: Deutsche Bank

Custody data return to normalcy? Fed custody holdings of Treasuries increased $13.2 billion this week. The data include the settlement of March month end coupon Treasury auctions. The change appears more normal than those in March and suggests foreign official demand for Treasuries. During the week ended March 12, 2014, Fed custody holdings of Treasuries lost $104 billion. The decrease coincided with the geopolitical tensions in Ukraine. Foreign investors have been a major buyer of Treasuries. In 2013, their demand accounted for 35% of the supply. Fed custody holdings of Treasuries increased $13.2 billion this week 60 40

$ Billions

20 0 -20 -40 -60 -80 -100 -120 1/1/09

Treasuries: $2,957.1B, last change $13.2B

1/1/10

1/1/11

1/1/12

1/1/13

1/1/14

Source: Fed and Deutsche Bank

Deutsche Bank AG/London

Page 13

4 April 2014 Global Fixed Income Weekly

Supply and demand for Treasuries during QE3 in 2013

QE3

4Q2012

4Q2013

Household sector

$

962 $

944

Foreign investors

$

5,574 $

5,842

Fed

$

1,666 $

2,209

Banks

$

243 $

217

Insurance

$

270 $

273

Pension

$

672 $

734

Money market funds

$

458 $

488

Mutual funds

$

580 $

641

Brokers and dealers

$

247 $

136

Other

$

899 $

843

Treasury Issues

$

11,569 $

12,328

$ $ $ $ $ $ $ $ $ $ $

Change % supply -2% (18) 35% 268 71% 543 -3% (26) 0% 3 8% 63 4% 30 8% 62 -15% (110) -7% (56) 100% 759

Source: Fed and Deutsche Bank

Positions were short The market rally after the non-farm payroll data on Friday was consistent with short positions in the market ahead of the data. Measures from CFTC Commitment of Traders (COT) report, put/call ratio and SMR all pointed to short positions. In last week’s COT data, spec investors were near a record short in FV, and set a new record short in Eurodollar futures. The put/call ratio in Treasury options was at 1.54, about one standard deviation above its long term average. The SMR weighted index dropped to 97.8 this week, its lowest reading in two months. Put/call ratio suggests short positions 2.25 2.00

Put/call ratio

Average

1.75 1.50 1.25 1.00 0.75 0.50 0.25 3/1/07

3/1/08

3/1/09

3/1/10

3/1/11

3/1/12

3/1/13

3/1/14

Source: CME Group and Deutsche Bank

Page 14

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Spec investors are near a record short in FV

Spec positions in all Treasury futures

500,000 400,000

Total net specs in Treasury futures in TY

400,000

Net spec in FV

300,000

200,000

200,000 0

100,000 0

-200,000

-100,000 -400,000

-200,000 -300,000 1/1/00

1/1/03

1/1/06

1/1/09

1/1/12

Source: Bloomberg Finance LP and Deutsche Bank

-600,000 1/1/08

1/1/09

1/1/10

1/1/11

1/1/12

1/1/13

1/1/14

Source: Bloomberg Finance LP and Deutsche Bank

The SMR weighted index dropped to 97.8 this week, its lowest reading in two months 101.0 100.5 100.0 99.5 99.0 98.5 98.0 97.5 97.0 SMR Weighted Index

96.5 96.0 Apr-11

Oct-11

Apr-12

Oct-12

Apr-13

Oct-13

Apr-14

Source: SMR and Deutsche Bank

Auction preview: 3s, 10s, and 30s Treasury will raise $64bn of notional worth $58.6bn in ten-year equivalents through three- and ten-year notes and 30-year bond auctions next week. These auctions will settle on Tuesday, April 15 against $50.5bn of coupon securities (including inflation adjusted TIPS) maturing on the same day. Last set of these auctions recorded lowest customer participation in about eight months as 3s and 30s were softly bid. But the demand for ten-year notes was the fourth largest at 70.9% (avg. 62%) since at least May 2006. 3-year note Customers combined for the lowest takedown of 45.4% (avg. 49.3%) in the last auction since June while their participation in the previous seven auctions was strong. Direct and indirect bidders both were below their respective oneyear averages and bid-to-cover ratio of 3.25 was also relatively soft. Combined allotment share to fund and foreign investors was right about the average at 38.4% as slack in latter’s share was picked up by the former, but it was far below their 49.9% share in February.

Deutsche Bank AG/London

Page 15

4 April 2014 Global Fixed Income Weekly

Dealers’ net inventory in 3- to 6-year Treasuries was about $3.9bn around last auction but jumped by $11bn in the week after FOMC meet to $9.6bn on March 26. 3-year note auction statistics Size ($bn) $30.9

1yr Avg

Primary Dealers

Direct Bidders

Indirect Bidders

Cover Ratio

50.7%

16.0%

33.2%

3.28

Stop-out 1PM WI Yield Bid

BP Tail -0.2

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

May-13 $ 32.0

54.7%

14.6%

30.7%

3.38

0.354

0.354

0.0

Apr-13

64.9%

16.2%

19.0%

3.24

0.342

0.338

0.4

$ 32.0

Source: US Treasury and Deutsche Bank

10-year note (re-opening) Thanks to solid directs last auction recorded strongest customer participation of 70.9% (avg. 62%) in a year. Directs bidders beat their average 18.9% participation with 27.5% takedown for the first time since October while indirects were about average 43.4%. Fund investors were allotted 47.2% (avg. 37.3%) of the supply while foreign investors got 20.1% (avg. 18.8%) and their combined allotment share of 67.4% was a record in past twelve months. Bidto-cover ratio of 2.92 (avg. 2.66) was also the highest of the year and the auction came through by a large 1.4 basis points. As of March 26, dealers’ net short position in 7-11 year Treasuries was $5.4bn, which is slightly below the $6.2bn around last auction but well below the $9.9bn level in prior week. 10-year note auction statistics Size ($bn) 1yr Avg $ 22.0

Primary Dealers 37.9%

Direct Bidders 18.9%

Indirect Bidders 43.3%

Cover Ratio

Stop-out 1PM WI Yield Bid

2.66

BP Tail -0.2

Mar-14 $

21.0

29.1%

27.5%

43.4%

2.92

2.729

2.742

-1.3

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

Jun-13 $

21.0

36.6%

11.7%

51.7%

2.53

2.209

2.208

0.1

May-13 $

24.0

49.2%

16.9%

33.9%

2.70

1.810

1.799

1.1

Apr-13 $

21.0

33.6%

29.1%

37.3%

2.79

1.795

1.791

0.4

Source: US Treasury and Deutsche Bank

Page 16

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

30-year bond (re-opening) Last auction recorded the lowest customer participation of 51.4% (avg. 56.2%) in nine months but each of the previous three auctions was solidly bid. Both the direct and indirect bidders were below their respective averages and the auction generated a large tail of 1.5 basis points. Allotment share to foreign investors was weak at 9.7% versus the average 11.8% while fund investors were about average at 38.6%. Bid-to-cover ratio of 2.35 was better than prior auction’s 2.27 but was still soft as compared the average 3.38 level. Dealers’ inventory in long-term Treasuries rose to $7.1bn from $4.3bn around the March auction. 30-year bond auction statistics Size ($bn) 1yr Avg

$14.0

Primary Dealers

Direct Bidders

Indirect Bidders

Cover Ratio

43.8%

16.2%

40.0%

2.38

Stop-out 1PM WI Yield Bid

BP Tail 0.1

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

Jun-13

$ 13.0

51.3%

8.5%

40.2%

2.47

3.355

3.324

3.1

May-13 $ 16.0

45.7%

15.5%

38.8%

2.53

2.980

2.990

-1.0

Apr-13

49.3%

19.2%

31.4%

2.49

2.998

2.990

0.8

$ 13.0

Source: US Treasury and Deutsche Bank

April Fed buyback schedule Fed plans to remove about $30bn of notional worth $33.4bn in ten-year equivalents though eighteen purchase operations in April. The frequency and weights of the purchases remain same across maturity sectors as shown in the table below but the duration takeout has fallen from $35bn in previous month in accordance with the last FOMC statement. TIPS purchases of the month will be carried out on Tuesday, April 8 whereas no operation is scheduled on 30th day of the month when the Fed will release its next policy statement. Net duration takeout of next week’s purchases is about $8.5bn in both notional and ten-year equivalents.

Deutsche Bank AG/London

Page 17

4 April 2014 Global Fixed Income Weekly

Fed Treasury buyback schedule for April 2014 Date

Operation type

Maturity Range

1-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

2-Apr

Treasury

5/15/21

2/15/24

2.25

7.2

7.94

2.09

3-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

7-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

8-Apr

TIPS

4/15/18

2/15/44

0.88

11.4

9.09

0.93

9-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

1.90

10-Apr

Treasury

4/30/18

12/31/18

3.13

4.14

4.54

1.66

11-Apr

Treasury

5/15/21

2/15/24

2.25

7.2

7.94

2.09

14-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

15-Apr

Treasury

1/31/20

3/31/21

2.00

5.8

6.11

1.43

16-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

21-Apr

Treasury

1/31/19

12/31/19

3.63

4.9

4.99

2.12

22-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

23-Apr

Treasury

5/15/21

2/15/24

2.25

7.2

7.94

2.09

24-Apr

Treasury

11/15/24

2/15/31

0.58

9.7

13.08

0.88

25-Apr

Treasury

2/15/36

2/15/44

1.03

16.7

17.52

2.10

28-Apr

Treasury

1/31/20

3/31/21

2.00

5.8

6.11

1.43

29-Apr

Treasury

5/15/21

2/15/24

2.25

7.2

7.94

2.09

29.400

9.21

9.77

33.41

Total

Par Amt. Average Avg ($b) Duration DV01

10y equ. ($b)

Source: Deutsche Bank

Distribution of monthly Treasury purchases by frequency and weight Nominal Coupon Securities by Maturity Range

4 - 4¾ 1 11%

4¾ - 5¾

5¾ - 7

7 - 10

10 - 20

1 12%

2 16%

4 29%

1 2%

TIPS

20 - 30 8 27%

4 - 30

years

1 frequency 3% weight

Source: NY Fed, Deutsche Bank

Page 18

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

United States

Rates Gov. Bonds & Swaps Rates Volatility

Aleksandar Kocic Research Analyst (+1) 212 250-0376 [email protected]

Derivatives „

The reversal of monetary policy should take place in three acts: Unwind of correlations, return of volatility, and actual rate hikes. The first stage, which started about a year ago with taper talk, has been more or less completed. The second act, return of volatility, is experiencing a difficulty in taking off. It seems like any Fed’s attempt of injecting it back into the market is encountering rejection.

„

We believe that in the light of continued pressure on long rates and a possibility of lower growth potential, volatility might not return even when Fed starts hiking. Risk premia are well bid, hikes could be limited and are likely already priced in by the forwards. The pressure currently experienced by gamma would propagate to intermediate expiries, sector that has so far seen continuous support.

„

In our view, in the short term the dialogue between the market and the Fed will continue, along the lines of too-much vs. not-enough, and will be visible especially in the Fed sensitive sector of the curve. This creates good entry levels for calendar spreads. We recommend:

„

Buy $100mn 6M3Y ATMF payers vs. sell $100mn 2Y3Y ATMF payers at 100c takeout.

Vol as an enabling obstacle and why are markets reluctant to take it back The reversal of monetary policy should take place in three acts: Unwind of correlations, return of volatility, and actual rate hikes. The first stage, which started about a year ago with taper talk, has been more or less completed, though not without consequences (and still waiting for a final verdict). Correlations have definitely repriced relative to QE days and are unlikely to go back to where they were 2 or 3 years ago. Fig 1 shows the residuals (the amount unexplained by breakevens) across different assets – after several years of range-bounded regime, their dispersion begins in the mid 2013. The second act, return of volatility, is experiencing a difficulty in taking off. It seems like any Fed’s attempt of injecting it back into the market is encountering rejection although the curve has been right on schedule in repricing the risk premia in a way that is consistent with Fed’s communications. Fig 2 shows the recent history of 6M10Y vol and 5Y swaps rate. While rates rose in mid-2013, and remained at elevated levels ever since, gamma gradually reverted back to pre-taper levels after a short-lived spike. This state of affairs, although not immediately raising red flags, is a cause of some concern as low volatility and risk premia are likely to cause misallocation of capital and be a catalyst to asset bubbles which could backfire in the long run.

Deutsche Bank AG/London

Page 19

4 April 2014 Global Fixed Income Weekly

Figure 1: Risky assets: the amount not explained with breakevens

Figure 2: Gamma cannot sustain its bid

Residuals to BKE 6

S&P

USD

CAD

EEM

Copper

2.250

IG Credit

5s 2.000

4

6M10Y vol

1.750 2

1.500 0

1.250 -2

1.000 0.750

-4

0.500

-6

Jan-11

Jul-11

Jan-12

Jul-12

Jan-13

Jul-13

Source: Deutsche Bank

Jan-14

J

F

M

A

M

J

J

A

S

O

N

D

J

F

M

Source: Deutsche Bank

Although the post-FOMC weeks remained ambiguous with the market struggling to attach a definitive meaning to the Fed’s message, none of those efforts stuck and the curve continued to bounce between bear flatteners and bull steepeners. What seems to be puzzling during that time is that, despite withdrawal of clarity, gamma remained under pressure and long gamma position, which held so much promise right after the meeting, had a disappointing performance with time decay eating up practically all of the P&L. The reality after the Friday’s payroll seems to be unchanged as far as vol is concerned. It is back to bull steepeners full force and further pressure on vol.

Is there hope for vol? The inability of vol to take off in a material way in the last 12 months seems to be a recurrent theme. Our understanding of this is that on one side there does not seem to be enough negative convexity 1 in the market to generate a sustained realized vol in rates, while, on the other, the macro shocks lack the power to produce sufficient escape velocity. So far, the market’s response to the policy unwind innuendos has been along the lines of bid for forward vol – the low vol regime will persist as long as there are no hikes, but once they start, volatility will return. We believe that in the light of continued pressure on long rates and a possibility of lower growth potential, volatility might not return even when Fed starts hiking. Risk premia are well bid -- both short term intermediates and steepens of the forwards are near historical highs. Hikes could be limited and are likely already priced in by the forwards. Given the low vol environment and the run risky assets had in the last year, they appear vulnerable and the adverse reaction should reside away from rates. The pressure currently experienced by gamma would propagate to intermediate expiries, sector that has so far seen continuous support on the back of believe that once we approach rate hikes, the dynamics will undergo a qualitative change.

1

The absence of negative convexity is a result of general change in rates landscape. Compared to pre-2008 period, convexity market is much smaller both in terms of supply as well as demand. With the bulk of current coupon convexity on Fed balance sheet, the transmission mechanism between rates and mortgages has been severed, while the options market has morphed into an insurance market with most of the exposure being far OTM and, therefore, low gamma. In addition, money managers have emerged as sellers of strangle, in size, so that dealers, who are now long those strikes generate substantial resistive force at the boundaries with their delta hedging. In such markets, macro or event shocks are short lived and gamma position remains unattractive.

Page 20

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 3: Calendar spreads remain at historical extremes 60 2Y3Y - 6M3Y

40 20 0 -20 -40 -60

00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 Source: Deutsche Bank

In our view, in the short term the dialogue between the market and the Fed will continue, along the lines of too-much vs. not-enough, and will be visible especially in the Fed sensitive sector of the curve. It is likely to see high levels of sensitivity to data in the Green-Gold sector of the curve with highly directional vol. This creates good entry levels for calendar spreads. We are sellers of intermediate vol as a way to finance a short-dated gamma position in 3Y tenors where both curve and vol roll-down appear optimal for such trades. This can be structured in different ways as we trade off the comfort of wide breakevens vs. discomfort of MTM exposure on long gamma and short market position. To be specific, we are sellers of 2Y3Y payers vs. buyers of 6M3Y payers. With respective vols at 105bp and 70bp, a premium neutral ratio is 3:1 (a $100mn notional on 2Y3Y buys $300mn of 6M3Y). This is roughly a vega neutral ratio as well. Here, we fix the ratio 1:1 and examine the structure across different strikes on the financing leg. Our first trade is: „

Buy $100mn 6M3Y ATMF payers vs. sell $100mn 2Y3Y ATMF payers at 100c takeout.

The numbers for different variants of the trade are summarized in Table, Fig 4. All positions are short vega and long gamma of various degree. Figure 4: Buy $100mn 6M3Y ATMF payers vs. sell $100mn 2Y3Y XXbp OTM payers Structure

Strikes

Takeout

6M P&L: roll to spot

6M P&L: roll to fwds

AT/AT

1.40 / 2.70

100c

16c

-41c

AT/50

1.40 / 3.20

46.5c

-3c

-42c

AT/100

1.40 / 3.70

9.5c

-19c

-49c

Source: Deutsche Bank

First row: For both legs ATMF, there is net takeout – we are selling “too much” of vega. The position is net long the market with initial takeout of 100c. 3Y spot is at 1%, 6M3Y at 1.40% and 2Y3Y at 2.70%. So here, the BKE is at 3.05%. If we roll to spot in 6 months (40bp below 6M3Y forwards), MTM is +16c (because the trade is net long the market). Roll to forwards in 6M renders the long position ATM and the net exposure is short OTM 18M3Y which is -41c.

Deutsche Bank AG/London

Page 21

4 April 2014 Global Fixed Income Weekly

Second row: use 50bp OTM 2Y3Y payers as a financing leg (1:1 weighting). The takeout is less generous at 46.5, but the breakeven moves to 3.36%. This is a net short position with roughly flat P&L over the 6M horizon if we roll to spot and roughly the same negative P&L (-42c) if we roll to forwards. Third row: selling 100bp OTM 2Y3Y payers to finance a gamma position reduces takeout to 9.5c while moving the BKE to 3.73%. This is a more extreme short position with more gamma. The P&L over the 6M horizon is negative in both mild rally or sell off. All trades are vulnerable to long term sell off beyond the breakevens with theoretically unlimited downside.

Page 22

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

United States

Rates Gov. Bonds & Swaps Credit Sovereigns

Steven Zeng, CFA Research Analyst (+1) 212 250-9373 [email protected]

Agencies The agency asset swap curve has flattened considerably since beginning of the year. On our spline model, 2s5s asset swap is just 16bp, down from a peak of 24bp in February, and it has dropped below the long-term average for the first time in almost a year. We believe the flattening of the agency asset swap curve was driven by two primary factors. The first is the steepening of the yield curve in both cash and swaps as improving data steadily pulled forward the timing of the Fed’s tightening cycle. As of Thursday’s close, 2s5s swap was just 3bp off its multi-year peak. The other reason is that intermediate and long agencies have outperformed during March, thanks to the GSE reform headlines providing a bullish sentiment for agency bonds. Positioning wise, investors also appear to be less bearish on fixed income and have extended their overall portfolio duration since the end of 2013, according to market surveys. Agency asset swap curve flattened with 2s5s asw

Unlike last year, the recent curve steepening reflects

dropping below its long-term average

more of a changing expectation in path of rates than investors reducing duration

bp 160

bp 30 2s5s US swap curve

Agency 5y asw less 2y asw (rhs)

140

25

120 100

20

80

15

SMR money managers survey. Shown is the actual portfolio duration as a percentage of target.

100 99.5

Less concerned about duration

99 98.5 98 97.5

60 10

40 20 Jan-13

101 100.5

97 96.5

Apr-13

Jul-13

Source: Deutsche Bank

Oct-13

Jan-14

5 Apr-14

Market bearish on rates and wary of duration

96 Jan-13

Apr-13

Jul-13

Oct-13

Jan-14

Apr-14

Source: Stone & McCarthy Research and Deutsche Bank

In Q1, US agencies returned 1.21% in total return and 0.27% in excess return over same duration Treasuries. For RV investors, the chart below shows the agency excess return per unit of duration by different maturity buckets. Clearly, it was the intermediates (5-7y) sector that had the best risk/reward profile during this period. Also, not surprisingly, the largest outperformance against Treasuries came in March, when the front-end of the Treasury curve became unhinged after the FOMC meeting and the GSE reform headlines combined a one-two punch for agency spreads. However, as we’ve pointed out for some time that spreads for 5y and in are already at historical tights, it was really the intermediates and long-end that were able to benefit from the bullish factors.

Deutsche Bank AG/London

Page 23

4 April 2014 Global Fixed Income Weekly

Agency bonds generated the most excess return over Treasuries in March this year, with the 5-7y sector having the best risk/reward profile Percent 0.14

Agency excess return per unit of duration Mar-14

0.12

Feb-14

0.10

Jan-14

0.08 0.06 0.04 0.02 0.00 (0.02) (0.04) 1-3y

3-5y

5-7y

7-10y

10y+

All

Source: Deutsche Bank Index Quant

Redemptions of callable agencies picked up in Q1 following two quarters of low activities. As bond yields more or less settled into a range, issuers redeemed $71bn of callables between January and March, 50% higher than in Q4 2013 and more than three times the amount in Q3 2013. In April, $120bn agency securities will be eligible for call. We estimate about $29bn of those are currently in the money.

Billions

Redemption of callable agencies announced during the quarter

200 180

18

160 140

51

FFCB 13 18 14

55

41

120 100 80

FHLMC

32

54

60 40 20

FNMA

54 62

77

78

0

FHLB

13 19 10 24 34 32

17

6

10 12 33 29

1 7 6 8

4 17 16 11

24 24 17

Source: Bloomberg Finance LP and Deutsche Bank

Page 24

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

United States

Credit Securitization

Steve Abrahams Research Analyst (+1) 212 250-3125 [email protected]

Mortgages Originally published on April 2 in The Outlook in MBS and Securitized Products.

The importance of partial hedging, the continuing bounce in housing The forward market usually prices in a rise in rates and a flattening of the yield curve, and the forward market usually has it wrong. But this time the forwards likely have it right. Soft US economic data hasn’t changed consensus projections of 3% GDP growth, which implies gains in nonfarm payrolls of more than 200,000 monthly for the rest of this year. The market clearly isn’t priced for that. The front end of the yield curve could easily move a lot faster than the back. And that continues to matter for pricing and hedging MBS. Stable MBS prepayments, with good prospects of more to come, have added to the importance of marginal shifts in rates. Cash flow discounting has become the new king. MBS investors that switch from vanilla duration to partial durations—from assumptions of parallel moves in rates to assumptions of nonparallel—should end up better off, at least as measured by expected return and variability of return. MBS portfolios with the wherewithal should be able to find positions that have adequate carry after hedging out partials and, taking any residual risks into account, leverage them.

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]

Nowhere in MBS is the importance of hedging partials greater than MBS derivatives, and that market continues to show good carry and manageable expected market risk after hedging with partials. Later in The Outlook, “A partial view of inverse IO” finds a few representative MBS derivatives that show manageable mark-to-market risk under most plausible scenarios and that generate estimated carry net of partial hedges of around $0-07 a month. That carry almost certainly reflects net spread and liquidity risk in the MBS derivatives, and each portfolio will have to judge whether that’s good enough given views on both risks. Expectations of wider spreads would work for most MBS derivatives, expectations of limited liquidity would not. But the principle is that even in the part of the market with the greatest exposure to likely reshaping in US rates, the tools are there to manage it and still have something left to show for the effort. So you got that goin’ for you, which is nice. *** A continuing bounce in housing The housing market still has room to run even though the discussion about it has become complicated in recent months by some voices focused on home prices and others on new construction and all worried about the impact of rising interest rates. It’s important to distinguish between home prices and the elements of residential investment because although both can describe the health of the housing market, they have different sensitivities to interest rates. Rising interest rates should have little impact on prices, a meaningful impact on construction and residential investment and a net impact that still stands to add 0.8% to GDP this year. That’s the conclusion of “US housing: The rebound continues,” published as a special report in this edition of The Outlook. And that implies continuing support for agency MBS supply, private MBS credit and the broader US economy. Deutsche Bank AG/London

Page 25

4 April 2014 Global Fixed Income Weekly

Yield (%)

Figure 1: Forwards predict a faster rise in the front end of the yield curve 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0

5

10

15

20

25

30

35

Treasury maturity (years) 31-Mar-14

1Y Fwd

3Y Fwd

5Y Fwd

Source: Bloomberg Finance LP

The price of the average US home should rise in 2014 by 7.8% and over the next three years by 17.5%, based on Deutsche Bank analysis. By 2017, nominal national home prices should reach their previous peak, restoring the housing equity lost in the financial crisis. For the economy, the expected rise in the most broadly held consumer asset should offset the drag of higher interest rates on new and existing home sales and other elements of residential investment. By lifting the net value of most US households and encouraging more consumer spending, housing should add 0.45% to US GDP growth. New construction should add another 0.37% for a total impact of around 0.8%. For the mortgage-backed securities markets, the price rebound should continue reducing loan defaults and losses and lift the supply of outstanding agency and private MBS. Lower defaults should accelerate home prices even further in a virtuous cycle. But the rate of house price appreciation should eventually slow to an annual pace of less than 4% or lower as the supply of distressed property falls to equilibrium levels. Since the supply of these properties varies dramatically across local markets, the prospects for housing across these markets should vary, too. Nevertheless, at equilibrium, if broad terms of financing remain stable, the price of the average home should rise in line with inflation. The updated outlook for US housing lays out details in a few key areas: „

The decelerating path of average US home prices in the next few years, and the dispersion across local markets

„

The impact of home prices on GDP both directly through new and existing home sales and indirectly through their influence on consumer wealth

„

The continuing powerful relationship between local serious delinquencies and the ability of local markets to recover

„

The local markets likely to draw the next wave of attention from investors in distressed property ***

Page 26

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 2: US home prices should keep rising, but the pace should slow 20 15

HPA (%)

10 5 0 -5 -10 -15 -20

Actual

Projected

Source: Deutsche Bank

The view in rates The risk in rates remains concentrated in the short end of the yield curve. Fed funds futures, although illiquid at long horizons, show levels of around 0.80% for December 2015 and 1.90% for December 2016—below the FOMC’s median forecasts of 1.0% and 2.0%, respectively, for those dates. That could be a concession for illiquidity, but it also could be a view that the economy will restrain the Fed. The latter looks like a poor assumption. My colleague Peter Hooper and his economics team expect the labor market to bounce back quickly into April and May. The market already expects a jump in nonfarm payrolls of 200,000 for March, but payrolls should keep hitting or exceeding that mark for the balance of the year. That should keep the Fed taper on track and their guidance for Fed funds in place. The short end of the curve will likely have to catch up. The view in spread markets MBS have generally trended sideways since last week, with TBA 30-year 3.0%s through 5.0%s within 1/32 to 3/32s of their hedges. Core positions remain unchanged since last week: „

Underweight the MBS-Treasury basis

„

Overweight 30-year 4.0%s and higher, underweight 30-year 3.5%s and lower

„

Overweight 15-year pass-throughs against 30-year

„

Underweight Ginnie Mae/conventional

„

Overweight LLB 3.0%s and 3.5%s

„

Overweight CQ/U6 3.0%s and 3.5%s

The view in mortgage credit Although stable-to-higher mortgage rates stand to put a drag on new and existing home sales, home prices should keep rising. Seasonally adjusted home prices have gone up in 23 of the last 24 months, according to the Federal Housing Finance Agency index. A detailed outlook on housing appears as a special report in this issue of The Outlook.

Deutsche Bank AG/London

Page 27

4 April 2014 Global Fixed Income Weekly

A partial view of inverse IO With a yield curve that looks likely to both rise and flatten in response to the wind down of Fed accommodative policy, agency MBS face a unique set of issues. Total duration should become less important while key rate durations become increasingly significant—and in no place should this have a greater impact than in the inverse IO market. It is instructive to see how investors in that market can manage this risk. The conclusion is that hedging out IIO partials with swaps can still leave investors with positive carry today and protect them against a wide range of curve moves in the future.

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

Inverse IO characteristics and curve exposure It is useful to look at two broad types of IIO when considering exposure to a curve move up and flatter—those backed by lower- and higher-coupon collateral (Figure 3). Figure 3: Descriptive characteristics on inverse IO backed by low- and high-coupon collateral Issue

Collat

Px

Strike

WAC

WAL

OAD

OAC

OAS

FHS 311 S1

FGLMC 3.5%

22-00

5.95%

3.97%

8.1

3.33%

YTF

9.2

-18.2

98

1M carry 0-11 6

1M px speed 4 CPR

FNR 13-130 SB

FNCL 5.0%

17-16

6.05%

5.35%

4.2

-1.93%

-1.1

-14.3

-60

0-07 6

14 CPR

Note: All levels indicative, as of COB: 3/26/14. Source: Deutsche Bank, Bloomberg Finance LP

The average life differences between the two creates unique exposures to forward short- and long-term rates—influencing the present value of future cash flows through projected UPB, coupon rate and the rate at which those flows will be discounted. As a result of the different cash flow intervals, these instruments also have very different exposures to different parts of the curve (Figure 4). Specifically, IIO backed by the collateral with the longer average life has much more exposure to the curve 5-years and in, and much less exposure to falling 10-year rates than an IIO backed by collateral that could easily become refinanceable. Figure 4: Key rate exposures differ significantly with collateral coupon Issue

KRDur 2Y

KRDur 5Y

KRDur 10Y

KRDur 20Y

FHS 311 S1

6.19

6.98

-2.33

-1.48

FNR 13-130 SB

6.93

3.12

-9.32

-1.89

Note: KRDs calculated as of 3/26/14. Source: Deutsche Bank.

Figure 5: Net carry after hedging partials still significantly positive Issue

Px

OAD

10yr HR

20yr HR

B/E 1M OAS chg

FHS 311 S1

22-00

7.28

2yr HR 5yr HR 79%

33%

-6%

-3%

5.9

1M carry unhedged 1M carry hedged 0-11 6

0-07 3

1M carry chg -0-04 3

FNR 13-130 SB 17-16

-0.75

70%

12%

-21%

-3%

6.5

0-07 6

0-06 7

-0-01 0

Note: 1M carry shows IIO carry net of pay or received fix swap positions in the ratios displayed above, a positive HR implies pay fixed, negative is receive fixed. B/E 1M OAS chg based on option-adjusted spread duration and equals the OAS widening necessary in one month to generate a mark-to-market loss equal to the hedge adjusted carry. All levels indicative, as of COB: 3/26/14. Source: Deutsche Bank

Hedging IIO partials with swaps Simply hedging the duration of IIO with one point on the curve is unlikely to properly protect current market value, and embeds implicit views on the future shape of the curve. Hedging the key rate durations with the corresponding swaps should more effectively protect market value against curve exposure. Although operationally a bit more complex, this approach still leaves investors with significant hedge-adjusted carry—which, in addition to the obvious Page 28

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

income benefit, also provides a cushion against mark-to-market losses on spread widening (Figure 5). Just as important as the cost of the hedge is its efficacy in actually hedging the market value of the IIO against changes in rates. Against instantaneous parallel curve shifts up, the analysis shows that the hedged performance is quite similar. Yet for large parallel shifts down the IIO hedged with partials performs substantially better than when hedged only with 10-year swaps (Figure 6 and Figure 7). Hedging with the partials actually adds some positive convexity in these cases through receiving on the 10- and 20-year points—though that kind of rate shift looks rather unlikely in the current environment. However, rates are unlikely to shift in parallel, and the true effectiveness of hedging the partials can be better established against scenarios where the curve is flatter with nominally higher rate levels across maturities. Indeed, if we shift the curve to the flatter and higher yield levels currently implied by the 1and 5-year forward curves, hedging the partials provides substantially better market value protection than hedging duration with only 10-year swaps (Figure 8 and Figure 9). The un-hedged IIO backed by 3.5% collateral loses over 7.7% in market value in an instantaneous shift to the 5-year forward curve, 5.9% hedged with 10-year swaps, and only 1.0% when fully hedged at the 2-, 5-, 10- and 20-year key rates. Additional risk factors Hedging the partials with swaps can effectively reduce curve and rate exposure, but does not provide protection against changes in the mortgage basis. A more encompassing hedging strategy for IIO first hedges the basis risk with a current coupon TBA position, and then hedges the new net key rate durations of the combined portfolio with swaps. However, this approach should reduce hedge-adjusted carry further, and adds an additional layer of operational complexity. Figure 7: IIO off 5.0% collateral 10Y vs. KRD hedges

0.0

0.0

-0.5

-0.2

-1.0

-0.4

Hedged MTM %

Hedged MTM %

Figure 6: IIO off 3.5% collateral 10Y vs. KRD hedges

-1.5 -2.0 -2.5 -3.0

-0.6 -0.8 -1.0 -1.2 -1.4

-3.5

-1.6 -100

-75

-50

-25

0

25

50

75

100

-100

Instantaneous parallel shifts in yield curve (bp) FHS 311 S1 KRDs

-75

-50

-25

0

25

50

75

100

Instantaneous parallel shifts in yield curve (bp)

FHS 311 S1 10s

FNR 13-130 SB KRDs

FNR 13-130 SB 10s

Note: Chart displays the hedged mark-to-market performance of the IIO over parallel, instantaneous shifts in the yield curve when hedged at the 2-, 5-, 10- and 20-year key rates with the corresponding swaps, and when total dv01 is hedged with 10-year swap. All levels indicative, as of COB: 3/26/14.

Note: Chart displays the hedged mark-to-market performance of the IIO over parallel, instantaneous shifts in the yield curve when hedged at the 2-, 5-, 10- and 20-year key rates with the corresponding swaps, and when total dv01 is hedged with 10-year swap. All levels indicative, as of COB: 3/26/14.

Source: Deutsche Bank.

Source: Deutsche Bank.

Deutsche Bank AG/London

Page 29

4 April 2014 Global Fixed Income Weekly

Figure 9: IIO off 5.0% collateral 10Y vs. KRD hedges at curve shifts to forwards

0.00

0.00

-1.00

-1.00

-2.00

Hedged MTM %

Hedged MTM %

Figure 8: IIO off 3.5% collateral 10Y vs. KRD hedges at curve shifts to forwards

-3.00 -4.00 -5.00

-2.00 -3.00 -4.00 -5.00

-6.00 -7.00

-6.00 Base

FHS 311 S1 Hedge KRDs

1Y Fwds

5Y Fwds

FHS 311 S1 Hedge 10s

Base

FNR 13-130 SB Hedge KRDs

1Y Fwds

5Y Fwds

FNR 13-130 SB Hedge 10s

Note: Chart displays the hedged mark-to-market performance of the IIO over instantaneous shifts in the yield curve to the current 1- and 5-year forward curves when hedged at the 2-, 5-, 10- and 20-year key rates with the corresponding swaps, and when total dv01 is hedged with the 10-year swap. All levels indicative, as of COB: 3/26/14.

Note: Chart displays the hedged mark-to-market performance of the IIO over instantaneous shifts in the yield curve to the current 1- and 5-year forward curves when hedged at the 2-, 5-, 10- and 20-year key rates with the corresponding swaps, and when total dv01 is hedged with the 10-year swap. All levels indicative, as of COB: 3/26/14.

Source: Deutsche Bank

Source: Deutsche Bank

Indeed, even just hedging the partials with swaps requires a more dynamic and involved approach to hedging than a simple holistic approach to duration. This approach may incur marginally higher transactional and operational costs. Finally, investors moving from MBS pass-throughs or rates into MBS derivatives likely give some liquidity—which may exacerbate mark-to-market volatility beyond that which modeled scenario analysis would suggest. Conclusion The benefit of decreased mark-to-market volatility should outweigh the marginal increase in costs and operational complexity. Indeed, in the wake of the FOMC and the subsequent curve move, the past few weeks have seen IIO bid list activity running at a 4x multiple of the recent average volume—with longer WAL lower coupons widening out as much as one point in dollar price. However, for a well hedged portfolio with capital to deploy this can present an opportunity to add on weakness—and we have in fact seen clearing levels stabilize over the past several days on an uptick in demand. If this reaction to the mere mention of policy tightening is prologue, then markto-market protection should become even more valuable as we inch closer to eventual Fed policy tightening. A more dynamic partial duration hedging approach against a higher and flatter curve should benefit IIO investors, and may also be worth consideration for investors in other MBS derivatives, passthroughs and structure with evolving key rate exposure.

Page 30

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Uneven impact of home price appreciation on turnover

„

CA has led the way but other areas stand ready to maintain the trend

Conventional wisdom maintains that the sharp climb in housing prices over the past 1-2 years should have an impact on housing sales and purchases that should be reflected in higher turnover rates in MBS. In 2013 alone, housing prices increased more than 10% in many big states such as CA, FL and TX, along with hard-hit states such as AZ, NV, MI and GA (Figure 10). Consistent with these home price increases, the side chart shows that existing home sales and mortgage applications for the purpose of buying homes with a conventional mortgage were steadily increasing through the first half of 2013, until rising mortgage rates took their toll. Meanwhile, increases in FHA premiums kept some downward pressure on mortgage applications for purchases of homes using government financing. Given the disparity in housing price increases across states, it’s not surprising that the impact on prepayments varies by state. Before mortgage rates started rising, there was a reasonably direct relationship between larger changes in housing prices and faster prepayment speeds for conventional mortgages (Figure 11). The relationship broke down somewhat for HPI increases above 10%. In the more recent period, the relationship broke down completely— higher home price appreciation was not associated with higher speeds when interest rates rose.

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

Mixed measures of home purchase activity 300

6

250

5

200

4

150

3

100

2

50

1

0

0

Milllions

Home price appreciation drives FHA speeds faster relative to conventional speeds

201101 201104 201107 201110 201201 201204 201207 201210 201301 201304 201307 201310 201401

„

MBA Conv. Purchase Index MBA Govt. Purchase Index NAR existing home sales (RHS) Note: all data seasonally adjusted Source: Deutsche Bank; National Association of Realtors; Mortgage Bankers’ Association

Despite the weakness in applications for purchases using government loans, home prices boost prepayments stronger among FHA borrowers (Figure 12). As home prices have risen rapidly in CA, many FHA borrowers have been able to refinance into conventional mortgages by eliminating paying the annual premiums, thus boosting prepayment speeds. As home prices continue to rise in areas other than CA, additional FHA borrowers will be able to refinance into conventional mortgages. This trend will keep upward pressure on FHA speeds relative to conventional speeds, even for coupons apparently not refinancable.

Figure 10: 2013 home prices increase in almost every state

25% 20% 15% 10% 5%

-5%

NV CA OR MI GA AZ FL WA HI ID TX UT WY IL MA SC MN CO MO NY AL TN DE NC NJ MD LA ME ND SD VA NH MT DC OH RI IN WV NE KS CT PA AK VT OK WI IA NM KY AR MS

0%

Note: Single-family only, all transaction types. Source: Deutsche Bank; CoreLogic

Deutsche Bank AG/London

Page 31

4 April 2014 Global Fixed Income Weekly

Figure 11: Conventional prepayment speeds related to HPI increases only when mortgage rates not rising Prepayment speed from 9/13–2/14 25

20

20

15

15 CPR

CPR

Prepayment speed from 3/13–8/13 25

10 5 0 -10%

10 5 0

0%

10%

20%

30%

-5%

0%

1-year change in home prices

5% 10% 15% 1-year change in home prices

20%

25%

Note: Mortgage note rates 3.75% to 4.25% for 2012 originations. Source: Deutsche Bank; CoreLogic HPI and Prime Servicing database

Figure 12: FHA prepayment speeds related to HPI increases much faster than conventional speeds Prepayment speed from 3/13–8/13 25

Prepayment speed from 9/13–2/14 25

CA

15

15

NV

10 5 0 -10%

CA

CPR

20

CPR

20

NV 10 5 0

0%

10%

20%

30%

-5%

0%

1-year change in home prices

5%

10%

15%

20%

25%

1-year change in home prices

Note: Mortgage note rates 3.75% to 4.25% for 2012 originations. Source: Deutsche Bank; CoreLogic HPI and Prime Servicing database

Despite high appreciation in Nevada, FHA prepayment speeds are substantially lower than would be expected. There are three possible explanations: „

Despite rapid underwater;

„

Nevada has relatively more investors than other areas, who may have different motivations than homeowners; and

„

Nevada homeowners have higher delinquencies that may prevent refinancing into conventional mortgages.

Page 32

appreciation,

many

Nevada

homeowners

are

still

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Special Report: US housing: The rebound continues Rising prices for US homes over the next few years should continue supporting both the broader economy and the mortgage-backed securities markets, although the pace of appreciation is set to slow. The price of the average US home should rise in 2014 by 7.8% and over the next three years by 17.5%. By 2017, nominal national home prices should reach their previous peak, restoring the housing equity lost in the financial crisis. For the economy, the expected rally in the most broadly held consumer asset should offset the drag of higher interest rates on new and existing home sales. By lifting the net value of most US households and encouraging more consumer spending, housing should add 0.45% to US GDP growth. New construction should add another 0.37%. For the mortgage-backed securities markets, the price rebound should continue reducing loan defaults and losses and lift the supply of outstanding agency and private MBS. Lower defaults should accelerate home prices even further in a virtuous cycle.

Steven Abrahams Research Analyst (+1) 212 250-3125 [email protected] Richard Mele Research Analyst (+1) 212 250-0031 [email protected] Ying Shen Research Analyst (+1) 212 250-1158 [email protected] Doug Bendt Research Analyst (+1) 212 250-5442 [email protected]

But the rate of house price appreciation should eventually slow to an annual pace of less than 4% or lower as the supply of distressed property falls to equilibrium levels. Since the supply of these properties varies dramatically across local markets, the prospects for housing across these markets should vary, too. Nevertheless, at equilibrium, if broad terms of financing remain stable, the price of the average home should rise in line with inflation. This updated outlook for US housing lays out details in a few key areas: „

The path of average US home prices in the next few years, and the dispersion across local markets

„

The impact of home prices on GDP both directly through new and existing home sales and indirectly through their influence on consumer wealth

„

The continuing powerful relationship between local serious delinquencies and the ability of local markets to recover

„

The local markets likely to draw the next wave of attention from investors in distressed property

Forecasts Following a gain of 11.0% in 2013 on the CoreLogic US home price index, home prices should rise 7.8% in 2014, 5.2% in 2015 and 3.6% in 2016 (Figure 13). These forecasts follow largely from the expectation that the supply of loans delinquent by 60 days or more will fall from 6.9% of outstanding mortgage principal at the end of 2013 to 4.3% at the end of 2016. Rising home prices accelerate the drop in this distressed supply, which, in turn, lifts home prices further in a reinforcing cycle.

Deutsche Bank AG/London

Page 33

4 April 2014 Global Fixed Income Weekly

Figure 13: National historical and projected house price annual returns 20 15

HPA (%)

10 5 0 -5 -10 -15 -20

Actual

Projected

Source: Deutsche Bank, CoreLogic

Local outcomes over the next one to two years should vary significantly as each area works through varying levels of distressed properties at various speeds. We saw that last year. Among the top 25 markets, returns in 2013 varied from 25.1% in Las Vegas to only 2.6% on Long Island (Figure 14). However, forecast returns three years out are converging toward a 3% to 4% annual pace similar to the national forecast. Areas in the West are expected to outperform over the near term (Figure 15): „

Las Vegas, where prices dropped 59.8% from their peak, should gain 7.4% over the next year and 17.2% over the next three years.

„

Oakland, which dropped 45.1%, is expected to gain 15.6% over one year and 27.9% over three years.

„

Santa Ana, which dropped 37.2%, is expected to gain 16.8% over one year and 36.8% over three years.

Figure 14: Local returns (historical and projected) vary widely among the top 25 markets

30 25

HPA (%)

20 15 10 5 0

2013 Actual

2014

2015

2016

Source: Deutsche Bank, CoreLogic

Page 34

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

18

40

16

30

14

20

12

10 HPA (%)

HPA (%)

Figure 15: Areas in the West have fallen the most, rebounded the most, and are expected to outperform next year

10 8

0 -10

6

-20

4

-30

2

-40

0

-50

National

West

2013 Actual

South 2014

2015

Northeast 2016

Midwest

2017

National

West Peak to Trough

South

Northeast

Midwest

Trough to Current

Source: Deutsche Bank, CoreLogic

Other markets are expected to lag: „

Long Island only fell 26.0% from its peak. However the recovery has been slow with prices rising only 5.5% since they bottomed out. We expect prices to rise 0.6% over one year and 6.6% over three years.

The differences across markets largely reflect differences in distressed supply. Serious delinquencies in the West at the end of 2013 stand at 5.3% of outstanding mortgage principal compared to 9.9% in the Northeast, and those delinquencies in the West are clearing the market faster than anywhere else. This is reflected in our forecast. Areas in the West have significantly higher returns over the next two years while returns in the Northeast areas are more moderate and do not drop off much after two years. Differences in distressed supply and the pace at which it clears the market often trace back to the legal system that manages that supply. There is only one judicial state in the West while there are 21 judicial states across the South, Northeast and Midwest. The stock of seriously delinquent properties in nonjudicial states has dropped by more than 50% from its peak while the stock in judicial states has dropped by less than 25% (Figure 16). Much of the regional historical and projected return patterns also can be traced back to judicial status (Figure 17). House prices in judicial and non-judicial states fell by nearly the same amount but judicial states have recovered half as much as nonjudicial states, and that pattern is expected to continue in 2014. Figure 16: Serious delinquency back log in judicial states

Serious Delinquency (%)

16 14 12 10 8 6 4 2 0

Judicial States

Non Judicial States

Source: Deutsche Bank, CoreLogic

Deutsche Bank AG/London

Page 35

4 April 2014 Global Fixed Income Weekly

14

30

12

20

10

10 HPA (%)

HPA (%)

Figure 17: Judicial and non-judicial states fell by nearly the same amount but judicial states have recovered half as much as non-judicial states and that pattern is expected to continue in 2014

8 6

0 -10

4

-20

2

-30 -40

0 National 2013 Actual

Non-Judicial States 2014

2015

Judicial States 2016

National

Non-Judicial States

Peak to Trough

Judicial States

Trough to Current

Source: Deutsche Bank, CoreLogic

Other key influences on home prices show promising trends: „

Census data shows that median household income fell during 2009 and 2010 but has been rising since then.

„

Unemployment should continue to decline based on forecasts from the Fed and the CBO.

„

Mortgage rates should rise 2.25% over the next three years based on the forecast for 10Y Treasury yields published in the Deutsche Bank US Fixed Income Weekly, but higher rates should have little influence in light of other factors.

„

Household formation is expected to continue to grow.

Detailed forecasts and related data for the top 25 markets are listed in Figure 18 and Figure 19.

GDP impact House price appreciation has a significant effect on GDP through new construction, wealth effects and other indirect effects. The cumulative impact through these channels should add 0.8% to 2014 GDP. In 2013 household residential investment grew by 16.8% to $414 billion contributing $59.5 billion to GDP (0.37% growth). We forecast residential investment by modeling the impact of house price appreciation, the change in delinquencies and the change in interest rates. Figure 20 shows the change in residential investment versus the change in interest rates. The recent rise in interest rates is expected to weigh on residential investment in 2014. Figure 21 shows residential investment versus HPA and Figure 22 shows residential investment versus the change in delinquencies. Although both higher HPA and lower delinquencies are expected to support residential investment, the impact of rising interest rates will offset much of it. For every 1% rise in interest rates we expect residential investment to fall by 2.9%. Still our model forecasts that residential investment will rise by 14.9% in 2014 adding another year of 0.37% growth to GDP in 2014.

Page 36

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 18: HPA forecasts and related data (sorted by balance)

Area Los Angeles New York Washington Chicago Atlanta Santa Ana Oakland San Diego Riverside Long Island San Francisco Phoenix Seattle Houston San Jose Minneapolis Baltimore Dallas Philadelphia Denver Sacramento Portland, OR Tampa Miami Las Vegas

Balance ($BB) 447 383 296 263 199 190 176 171 162 155 148 140 138 138 137 120 120 117 113 105 98 88 80 75 68

Level 6.0 11.0 5.3 9.3 5.9 4.4 4.5 4.6 7.3 13.3 2.8 3.8 5.9 3.8 3.4 3.8 7.9 4.1 5.8 2.9 4.8 5.6 15.2 23.0 13.6

Serious Delinquency Actual 1Y Resolution Resolution Change Pace Pace (%) -2.7 1.1 17.9 -1.5 1.6 14.4 -1.4 0.5 9.6 -3.1 1.2 13.1 -2.2 0.9 15.9 -2.4 0.7 16.3 -2.7 0.7 15.8 -2.3 0.7 15.6 -3.8 1.2 16.4 -1.3 1.5 11.2 -1.4 0.3 10.6 -2.8 0.5 12.2 -2.7 0.7 12.1 -0.7 0.3 8.7 -1.9 0.4 12.6 -1.6 0.3 7.7 -1.4 0.9 11.3 -0.6 0.3 8.6 -0.5 0.7 12.0 -1.1 0.2 5.3 -3.0 0.7 15.3 -1.3 0.8 13.6 -4.4 1.8 11.8 -6.8 2.3 10.1 -4.7 1.9 14.1

REO HPA Actual 1Y Level 2013 Actual 1Y Projection 3Y Projection Change 0.2 -0.3 19.1 13.5 31.3 0.1 0.0 7.3 6.2 17.3 0.3 0.0 7.7 7.3 13.5 1.2 0.2 12.5 7.2 18.8 10.9 21.9 0.4 -0.2 15.0 0.1 -0.3 19.7 16.8 36.8 0.2 -0.3 24.1 15.6 27.9 0.2 -0.3 18.5 13.7 26.4 12.6 23.0 0.4 -0.4 22.0 0.1 0.0 2.7 0.6 6.6 0.1 -0.2 18.9 12.7 19.1 0.3 -0.3 13.9 12.5 21.5 0.4 0.1 13.6 8.4 20.2 0.2 -0.1 10.7 7.7 14.1 0.1 -0.2 18.5 12.7 21.2 0.6 -0.2 7.5 7.6 17.1 0.5 0.2 5.0 3.8 11.6 0.2 -0.1 9.8 9.0 16.5 0.4 0.1 7.7 10.6 19.6 0.2 -0.1 10.0 9.9 14.4 0.3 -0.3 19.9 13.6 23.8 15.0 8.2 16.1 0.3 0.0 1.2 0.4 10.2 4.0 13.1 2.1 0.7 10.6 7.3 25.6 0.6 0.1 25.1 7.4 17.2

Source: Deutsche Bank, CoreLogic

Figure 19: HPA forecasts and related data (sorted by 3Y projection)

Area Santa Ana Los Angeles Oakland San Diego Miami Sacramento Riverside Atlanta Phoenix San Jose Seattle Philadelphia San Francisco Chicago New York Las Vegas Minneapolis Dallas Portland, OR Denver Houston Washington Tampa Baltimore Long Island

Balance ($BB) 190 447 176 171 75 98 162 199 140 137 138 113 148 263 383 68 120 117 88 105 138 296 80 120 155

Level 4.4 6.0 4.5 4.6 23.0 4.8 7.3 5.9 3.8 3.4 5.9 5.8 2.8 9.3 11.0 13.6 3.8 4.1 5.6 2.9 3.8 5.3 15.2 7.9 13.3

Serious Delinquency Actual 1Y Resolution Resolution Change Pace Pace (%) -2.4 0.7 16.3 -2.7 1.1 17.9 -2.7 0.7 15.8 -2.3 0.7 15.6 -6.8 2.3 10.1 -3.0 0.7 15.3 -3.8 1.2 16.4 -2.2 0.9 15.9 -2.8 0.5 12.2 -1.9 0.4 12.6 -2.7 0.7 12.1 -0.5 0.7 12.0 -1.4 0.3 10.6 -3.1 1.2 13.1 -1.5 1.6 14.4 -4.7 1.9 14.1 -1.6 0.3 7.7 -0.6 0.3 8.6 -1.3 0.8 13.6 -1.1 0.2 5.3 -0.7 0.3 8.7 -1.4 0.5 9.6 -4.4 1.8 11.8 -1.4 0.9 11.3 -1.3 1.5 11.2

REO HPA Actual 1Y Level 2013 Actual 1Y Projection 3Y Projection Change 0.1 -0.3 19.7 16.8 36.8 0.2 -0.3 19.1 13.5 31.3 0.2 -0.3 24.1 15.6 27.9 0.2 -0.3 18.5 13.7 26.4 2.1 0.7 10.6 7.3 25.6 0.3 -0.3 19.9 13.6 23.8 0.4 -0.4 22.0 12.6 23.0 0.4 -0.2 15.0 10.9 21.9 0.3 -0.3 13.9 12.5 21.5 0.1 -0.2 18.5 12.7 21.2 0.4 0.1 13.6 8.4 20.2 0.4 0.1 7.7 10.6 19.6 0.1 -0.2 18.9 12.7 19.1 1.2 0.2 12.5 7.2 18.8 0.1 0.0 7.3 6.2 17.3 0.6 0.1 25.1 7.4 17.2 0.6 -0.2 7.5 7.6 17.1 0.2 -0.1 9.8 9.0 16.5 15.0 8.2 16.1 0.3 0.0 0.2 -0.1 10.0 9.9 14.4 0.2 -0.1 10.7 7.7 14.1 0.3 0.0 7.7 7.3 13.5 1.2 0.4 10.2 4.0 13.1 0.5 0.2 5.0 3.8 11.6 0.1 0.0 2.7 0.6 6.6

Source: Deutsche Bank, CoreLogic

Deutsche Bank AG/London

Page 37

4 April 2014 Global Fixed Income Weekly

30

2

20

1.5

10

1 0.5

0

0

-10

-0.5

-20

Change (%)

Change (%)

Figure 20: Residential investment will be negatively impacted by interest rates

-1

-30

-1.5

-40

-2

Residential Investment Change next 12 months (%) Negative of Interest Rates Change over 12 months (rhs,%) Source: Deutsche Bank, Bloomberg Finance LP, BEA

New construction is a direct contribution to GDP but housing also has an indirect effect through the wealth created by home price appreciation. Houses are a large component of the assets on the consumer’s balance sheet. As of the end of 2013 the Federal Reserve reports that household real estate represents $19.4 trillion out of $94.4 trillion in household (and non-profit) assets. In 2013 alone household real estate grew by $2.0 trillion, which is an 11.5% increase. This compares well with the CoreLogic national home price index, which grew by 11.0% in 2013. National home prices should rise 7.8% in 2014 and 17.5% in three years. This would imply an additional $1.5 trillion in wealth on the consumer’s balance sheet in one year and $3.4 trillion over three years. This additional wealth contributes to GDP through the wealth effect, which is well documented in academic literature.2,3 While there are a range of estimates for the consumer’s marginal propensity to consume (MPC) with respect to housing wealth, we use a conservative estimate of 5%. That is, for every dollar Figure 21: Positive HPA will add to residential investment

Figure 22: Decline in delinquencies will also add to 30

20

20

10

10

Change (%)

Change (%)

residential investment 30

0 -10 -20

0 -10 -20

-30

-30

-40

-40

House Price Change over 12 months (%) Residential Investment Change over 12 months (%) Source: Deutsche Bank, CoreLogic, BEA

Negative of Delinquency Change over 12 months (%) Residential Investment Change over 12 months (%) Source: Deutsche Bank, MBA, BEA

2 Calomiris, Longhofer and Miles (2012). “The Housing Wealth Effect: The Crucial Roles of Demographics, Wealth Distribution and Wealth Shares,” NBER Working Paper 17740. 3 Case, Quigley and Shiller (2011). “Wealth Effects Revisited,” Cowles Foundation Discussion Paper 1784.

Page 38

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 23: HPA tracked serious delinquencies for top 25 markets in 2013 30

HPA in 2013 (%)

25 20 15 10 5 0 -50.0

-40.0

-30.0

-20.0

-10.0

0.0

Change in Serious Delinquencies in 2013 (%) Source: Deutsche Bank, CoreLogic

increase in housing wealth the consumer should spend an additional 5 cents. This wealth effect estimation follows a similar analysis published in the Deutsche Bank US Economics Weekly.4 Applying this to our forecast means that increasing housing wealth will spur consumers to spend an additional $75.7 billion in one year and $170 billion in three years. This could add 0.45% to GDP (currently $17 trillion) in one year. If the MPC is closer to 10% then housing wealth could add 0.9% to GDP. This is in addition to any direct contribution from construction.

Serious delinquencies House price returns continue to closely track the change in serious delinquencies (Figure 23). In 2013 serious delinquencies fell 21.5% (from 8.8% to 6.9%) nationally while house prices rose 11.0%. Locally Santa Ana serious delinquencies fell 35.5% and prices rose 19.7%. Meanwhile, Philadelphia and Long Island serious delinquencies fell only about 9% while prices only rose 7.7% and 2.7%, respectively. The strong relationship between serious delinquencies and HPA also holds in the past. Furthermore, serious delinquencies can be used to forecast HPA. Figure 24 and Figure 25 show the negative of the change in serious delinquency versus HPA that follows in the next 12 months. Since 2000 the change in serious delinquency has been a good predictor of the direction and relative magnitude of future HPA. We define serious delinquency as loans that are 60+ days delinquent, in foreclosure or held as REO (Figure 26). From 2000 to 2006, serious delinquencies were relatively stable around 2% of loan balances. However, during the house price run up just prior to 2006 serious delinquencies were on a slight downtrend. In 2006 the housing market peaked and starting dropping rapidly. At the same time serious delinquencies made a rapid rise reaching a peak of 12.9% in 2010. Since then serious delinquencies have been on a gradual path downward while the trajectory of house prices turned up.

4

“Don’t underestimate the lift from housing,” US Economics Weekly, January 18, 2013.

Deutsche Bank AG/London

Page 39

4 April 2014 Global Fixed Income Weekly

Figure 25: Oakland

20 15 10 5 0 -5 -10 -15 -20 -25 -30 -35

Change (%)

Change (%)

Figure 24: Long Island

Negative SD Change

30 20 10 0 -10 -20 -30 -40 -50 -60 -70

Negative SD Change

HPA next 12 months

Source: Deutsche Bank, CoreLogic

HPA next 12 months

Source: Deutsche Bank, CoreLogic

We expect serious delinquencies to continue to decline at its current pace, driving our expectation for home prices up for the next few years. We also expect the decline in serious delinquencies to eventually moderate as the stock of distressed properties diminishes. Once serious delinquencies stabilize we expect that income, population growth and other local factors will become the main drivers of house price appreciation. But for now serious delinquencies will continue to be important in forecasting HPA. Thus we require a serious delinquency forecast. We start by creating a transition matrix for every locality and mortgage product type. The transition matrix summarizes the probabilities (or roll rates) that a current loan will be delinquent, a delinquent loan enters foreclosure, a foreclosed loan becomes REO, an REO property liquidates and other possible transitions. We consider seven different loan statuses: current, 30 days delinquent, 60 days delinquent, 90+ days delinquent, in foreclosure, REO and terminated. With the exception of terminated loans, each month a loan can move from any of the statuses to any of the other statuses with 42 (7x6) different roll rates. The roll rates are estimated using historical data for each locality (456 of them) and each product type (Alt-A, prime, subprime, option ARM, GSE, GNMA, bank loan). Figure 26: Serious delinquencies and sub-components 14

Delinquency (%)

12 10 8 6 4 2 0

DQ60

DQ90

FCL

REO

Source: Deutsche Bank, CoreLogic

Page 40

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 27: Forecasting serious delinquencies in Riverside

Serious Delinquency (%)

30 25 20 15 10 5 0

SD Actual

SD Roll Matrix without HPA

SD Model with HPA

Source: Deutsche Bank, CoreLogic

Roll rates will also vary depending on income, unemployment and in particular, prior HPA. Figure 27 shows the twelve month forecasted serious delinquency compared to the actual, with and without using HPA. Including HPA improves the fit dramatically. This also introduces two-way causality in our model, between serious delinquencies and HPA.

Investor activity Finally, investors have played an important role in the housing rebound by bringing in equity to replace the substantial amount lost by homeowners in the housing crash. RealtyTrac has reported that, as of January 2014, institutional investors who purchase at least ten properties in a year account for 5.2% of all sales, down from 8.2% a year ago. Distressed sales have been on a downtrend since 2011 (Figure 28). But with serious delinquencies at 6.9% nationally there is still a substantial inventory of houses that investors may be able to buy at bargain prices. We take a look at potential income from buying-to-rent across large markets. While the monthly rent is highest in San Jose, Los Angeles and Oakland, those areas have higher priced homes that would cut into investor returns (Figure 29). Figure 28:Distressed sales declining since 2011

Percent of All Sales (%)

35 30 25 20 15 10 5 0

Short Sales

REO Sales

Source: Deutsche Bank, CoreLogic

Deutsche Bank AG/London

Page 41

4 April 2014 Global Fixed Income Weekly

Figure 30: Gross rental yield in large markets

3,000

14

2,500

12 10

2,000

Yield (%)

Monthly Rent

Figure 29: Monthly rent in large markets ($)

1,500 1,000

8 6 4

500

2

0

0

2009 Source: Deutsche Bank, RentRange

2013

2009

2013

Source: Deutsche Bank, RentRange, CoreLogic

Several areas show a large increase in rent since 2009 but much of the increase goes away when looking at price per square foot, with the exception of San Jose. Figure 30 takes purchase price into account by dividing the annual rent by the local median sales price. This shows that on a cash flow basis Miami, Tampa and Philadelphia present the best opportunities for investors as of the end of 2013. The opportunities available in Las Vegas, Dallas, Phoenix and Riverside have declined since 2009. In reality the returns should be adjusted to reflect renovation, taxes, insurance, maintenance and leverage. Of course investors should also consider house price appreciation in their total return assessment. In this case Oakland looks like a good opportunity and Miami looks like an exceptional opportunity over one year and three year horizons.

Page 42

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

United States

Credit HY Strategy IG Strategy

US Credit Strategy Credit Briefs Invitation: U.S. Credit Strategy Outlook Conference Call Deutsche Bank clients are invited to join us for a discussion on the outlook for U.S. corporate credit on Tuesday, April 8, 2014 at 11:00am EDT. Dial-in: + 1 800-309-8606 Intl: +1 706-679-0645 Access Code: 19270958. 1-week replay: +1 866 247-4222 Intl.: +1 631 510-7499 Access Code: 19270958

Oleg Melentyev, CFA Strategist (+1) 212 250-6779 [email protected] Daniel Sorid Strategist (+1) 212 250-1407 [email protected]

Figure 1: Credit markets in March 5yr Trsy 10yr Trsy IG HY Loans S&P 500

A look at market trends and developments In this publication, we intend to provide our readers with a concise summary of recent performance in US credit markets, touching on its most important attributes, such as changes in spreads and underlying Treasuries, returns, issuance and positioning trends, as well as other recent developments in related markets. The report will provide a quick rundown of recent events and their impact on our previously published views and recommendations.

AAs As BBBs BBs Bs CCCs -1.00

-0.80

-0.60

-0.40

MTD Total Return

-0.20

0.00

0.20

0.40

0.60

0.80

MTD Excess Return (Trsy)

HY Duration Segments OAD 0..1 yrs OAD 1..2 yrs OAD 2..3 yrs

No shelter in short duration In an overall strong month for credit, short-duration IG paper lagged in March as rising fed funds expectations bear-flattened the Treasury curve nearly 25 bps between the 5Y and 30Y maturity points. IG duration buckets between one and six years posted negative total returns in March, while the long end of the IG curve returned an impressive 80 bps. Strong spread performance across IG drove positive excess returns over Treasuries across most duration buckets. Nevertheless, the yield curve flattening proved to be a pain trade for the mutual fund money that has been crowding into short-duration IG paper for the past six months. HY also strongly outperformed Treasuries in all but the longest-duration category.

OAD 3..4 yrs OAD 4..5 yrs OAD 5..6 yrs OAD 6..7 yrs

OAD 7..9 yrs OAD 9..14 yrs -0.20

0.00

0.20

MTD Total Return

0.40

0.60

0.80

1.00

1.20

MTD Excess Return (Trsy)

IG Duration Segments OAD 0..1 yrs OAD 1..2 yrs OAD 2..3 yrs OAD 3..4 yrs OAD 4..5 yrs OAD 5..6 yrs

Money creeps back into longer-duration IG With 10-year yields unable to push through 3 percent, we are seeing some early evidence of mutual fund investors capitulating on their short-duration overweight. Cumulative fund outflows from long-duration IG paper peaked around 10 percent at the end of last year, but now stand at around 6 percent as money has begun to creep back into longer-duration assets in Q1. For both HY and IG credits, relative value appears to reside in the 9-14 year duration segments.

OAD 6..7 yrs

OAD 7..9 yrs OAD 9..14 yrs -0.80

-0.60

-0.40

-0.20

MTD Total Return

0.00

0.20

0.40

0.60

0.80

1.00

MTD Excess Return (Trsy)

Source: Deutsche Bank

Financials lagged amid heavy issuance Non-domestic financials accounted for an unusually high share of issuance in Q1, providing perhaps some headwind to further financial outperformance. The spread differential between financials and non-financials remained at double-digit wides to non-financials in most quality and durations segments. We introduce a credit/duration heatmap for the financial sector, and a comparison of spreads relative to historic tights would favor single-A financials relative to equivalently rated non-financials.

Deutsche Bank AG/London

Page 43

4 April 2014 Global Fixed Income Weekly

Credit Briefs

300

30

250

25

200

20

150

15

100

10

50

5

0

0

-50

-5

-100

-10

-150

-15

-200 Mar 13

-20 May 13

Jul 13

Sep 13

Nov 13

Jan 14

Change in IG Spread

Change in 5yr Yield

Cumulative LT Fund Flows

Cumulative ST Fund Flows

Source: Deutsche Bank

Figure 3: HY Pure Sector Returns ex Quality Factors

Figure 4: IG Pure Sector Returns ex Quality Factors

Transports Capital Goods Media Utilities Commrcl Svcs Real Estate Consmr Prodts Food Retail Technology Financials Energy Materials Telecoms Automotive Health Care Gaming

Real Estate Technology Media Commrcl Svcs Telecoms Energy Transports Gaming Capital Goods Consmr Prodts Food Financials Materials Retail Automotive Health Care Utilities

-0.80

-0.60

-0.40

-0.20

0.00

0.20

0.40

0.60

-0.20

-0.15

-0.10

MTD Excess Return (Rating)

0.00

0.05

0.10

0.15

0.20

0.25

Source: Deutsche Bank

Financials still offering a modest spread pick-up Financial issuers’ 300 bp excess spread to non-financial issuers has been ground away over the past 2.5 years, and now rests around 25 bps for singleA, 4-5 year duration issuers. The financial-nonfinancial spread differential has been bouncing around these levels since February. If the market rallies further, and confidence continues to rise in the financial regulatory regime, financials could be closer to parity with non-financials. This wouldn’t be a first. Even after accounting for credit quality and duration, financials traded as tight as 10 bps through non-financials in 2006. Though the upside is clearly capped, we’re inclined to position for modest financial sector outperformance consistent with our constructive 12-month credit view. As the figure below shows, the excess financial spread for single-A credits ranges between 10 and 40 bps, with the biggest spread differential in lowerquality single-A of mid to longer duration. Mid-BBB financials offer the greatest spread differential, although financials in many duration buckets within the lowest quality BBB paper are already through non-financials.

Page 44

-0.05

MTD Excess Return (Rating)

4-5Y Duration Fins ex-NonFins OAS 550 450 350 OAS

Source: Deutsche Bank

Cumulative IG Fund Flow, Pct of Assets

Figure 2: IG cumulative fund flows Cumulative Change IG Spread/5yr Trsy, bps

Sector performance update Financials stood out as the worst performer in IG on a total return basis in March, but the negative performance was primarily the result of the sector’s shorter-duration profile. Relative to Treasuries, financials, like all other sectors in IG, outperformed. On a pure sector basis, that is, as outlined in an earlier publication, taking into account the sector’s duration and credit quality. Financials were more or less flat. The next largest sector by size – energy, one we favor -- turned in a middle of the pack performance. The media sector proved a top performer on a pure sector basis –– in both HY and IG. The utility sector posted the weakest performance on a pure sector basis in IG, but interestingly was among the top performers in HY. Utilities in the IG space tend to be regulated players which oftentimes don’t participate in spread tightening environments, while in HY the players are largely unregulated and thus can tend to move with a higher market beta. Sector selection remains an important driver of returns in these tight-spread markets, with monthly puresector returns fluctuating by more than 1 percentage point difference between the best and worst sector in HY, and about more than 30 bps in IG.

250 150

50 -50 2006

2007

2008

2009

2010

2011

2012

2013

2014

Fins Ex NonFins (4-5Y OAD)

Source: Deutsche Bank

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 5: Financials ex-Nonfincials OAS, by Duration

Figure 6: Financials ex-Nonfincials OAS, by Rating 70

60

60

50

50

40

40

30

30

OAS (bps)

70

20 10 0 -10

20 10 0

1..2

2..3

3..4

4..5

5..6

6..7

7..9

9..14

-20

-10

A1

A2

A3

BBB1

BBB2

BBB3

-20

-30

-30

A1

A2

A3

BBB1

BBB2

BBB3

1..2

Source: Deutsche Bank

2..3

3..4

4..5

5..6

6..7

7..9

9..14

Source: Deutsche Bank

Map to proper duration positioning In February, we introduced a framework for identifying value across duration and quality segments in IG and HY credit. The heat map we introduced applies a percentile ranking of current spread levels against their historical ranges (since 2006). Here, we update the heat map and introduce a sector-specific map for both financial and non-financial IG credits. (Where pairs of rating and duration don’t provide a sufficient number of historical data points, we withhold a score.) For the overall IG market, the heat map demonstrates that the front-end of the IG market has taken the lead as the overall market has been grinding toward historic tights. Since our initial report we have seen many more red squares emerge, particularly in non-financials, where percentile scores that are at or below 5% are becoming the norm for duration segments out to the 7-year duration point, in credit qualities as low as high BBB. The highest-quality single-A names now trade at or near tights through most duration ranges, while mid-BBB provide perhaps the best overall credit quality segment.

Figure 7: IG OAS percentiles ALL IG OAD A1 A2 A3 BBB1 BBB2 BBB3

1..2 3% 12% 18% 19% 18% 2%

2..3 0% 0% 0% 0% 15% 16%

3..4 1% 0% 6% 5% 15% 11%

OAD 4..5 5..6 0% 1% 2% 18% 15% 15% 14% 16% 17% 18% 23% 18%

6..7 0% 3% 18% 8% 19% 19%

7..9 14% 12% 23% 17% 31% 19%

9..14 26% 12% 31% 30% 36% 32%

Source: Deutsche Bank

The value signals we had inferred from IG credit in the 7-9 year bucket in February are now less compelling, although on a relative value basis we would still argue that the 7-14 year duration segments should offer the biggest upside if spreads continue to grind toward tights. Our non-financial heatmap paints an even more vibrant picture of front-end richness. Our financials heatmap suggests a sector where value – as measured in spread percentiles – is more broadly based in the single-A segment. Low single-A financials in the 7-9 year duration bucket have the highest reading on a percentile basis. Figure 8: Valuation heat map for non-financial IG NONFINS OAD OAD 1..2 2..3 3..4 4..5 5..6 6..7 7..9 A1 2% 1% 0% 34% A2 5% 2% 0% 1% 17% 0% 4% A3 3% 0% 0% 0% 3% 6% 21% BBB1 4% 0% 1% 6% 9% 10% 11% BBB2 16% 3% 12% 17% 16% 17% 23% BBB3 4% 18% 17% 21% 18% 22% 15% Source: Deutsche Bank

Deutsche Bank AG/London

9..14 48% 15% 37% 35% 37% 38%

Figure 9: Valuation heatmap for financial sector IG FINS OAD OAD 1..2 2..3 3..4 4..5 5..6 6..7 7..9 9..14 A1 23% 0% 3% 2% 20% 9% 25% 18% A2 17% 0% 14% 18% 13% 15% 22% 22% A3 17% 1% 18% 9% 20% 33% 28% BBB1 16% BBB2 26% BBB3 Source: Deutsche Bank

Page 45

4 April 2014 Global Fixed Income Weekly

The way we read the high-yield charts, we would conclude that relative value continues to reside in longest-duration segments, mainly in the 7..9 and 9..14 years. Many regions in the middle – 5 to 6 year durations in particular – appear to be trading quite tight here. The shortest duration segment – those bonds under 1yr duration – also appears to be on the wide side, although this remains largely a function of still-distorted short rates as opposed to any real “value” in the shortest HY bonds.

Figure 10: HY OAS percentiles OAD BB1 BB2 BB3 B1 B2 B3 CCC

0..1 16% 34% 27% 30% 29% 30% 30%

1..2 18% 14% 19% 20% 18% 25% 14%

2..3 19% 13% 11% 13% 21% 20% 24%

3..4 19% 13% 8% 18% 18% 22% 21%

4..5 18% 14% 12% 18% 18% 19% 14%

5..6 18% 17% 1% 14% 6% 9% 3%

6..7 21% 17% 18% 16% 8% 13%

7..9 9..14 16% 48% 21% 43% 12% 28% 39% 76% 23% 25%

Source: Deutsche Bank

Figure 11: HY spread curves by duration buckets

Figure 12: IG spread curves by quality buckets 250

600 500

200

400 150

300 100

200 50

100

0

0 BB1

BB2 0..1

1..2

BB3 2..3

3..4

B1 4..5

5..6

B2 6..7

B3

A1 1..2

7..9

Source: Deutsche Bank

A2 2..3

A3 3..4

4..5

BBB1 5..6

6..7

BBB2 7..9

9..14

BBB3 14..18

Source: Deutsche Bank

Issuance update In a recent update, we noted the very slow pace of activity in HY issuance and notably above-average activity in loans. Here, we update the trend in IG issuance, which crossed through $300 bn in the first quarter, about 15% more than the very strong Q1 of 2013. (When normalized to market size, IG issuance is about on par with last year.) Equally notable is the surge in issuance by nonU.S. financial institutions, which by our count jumped close to 50% from the prior year. Heavy financial sector issuance may be providing some headwinds to further financial sector tightening against non-financials, and furthers a trend of financials growing as a share of total IG market issuance. Financials accounted for roughly half of IG issuance in the first quarter, versus a share of about 35% seen toward the end of last year. The uptick may be a temporary one, driven by a return of risk appetite for European financials. (Italian 10-year government bond yields recently made a new record low, and peripheral sovereign spreads to Bunds are nearly 4 percentage points tighter than the wides of late 2011.) Also, in a low spread environment investors are growing enthusiastic about European bank subordinated paper that qualifies as capital under Basel III regulations. Figure 13: First quarter IG issuance, $ bn

Figure 14: First quarter IG issuance, % of IG market 30%

400 Q1 Issuance, % of IG market

Q1 Issuance, $ bn

350 300 250 200 150 100 50

20% 15% 10% 5% 0%

00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 YANKEE FIN Source: Deutsche Bank; Dealogic

Page 46

25%

US FIN

YANKEE CORP

US CORP

00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 YANKEE FIN

US FIN

YANKEE CORP

US CORP

Source: Deutsche Bank; Dealogic

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Europe

Credit HY Strategy

Stephen Stakhiv Strategist (+44) 20 754-52063 [email protected]

Credit Strategy: The Callable Short Life There are two trends that we have been watching closely as of late: average bond life and callability. In general, much of the European HY market is callable and we wanted to explore when and how this may affect the market going forward.

Nick Burns, CFA Strategist (+44) 20 754-71970 [email protected]

We first look at Average Years to Maturity in Figure 1, which breaks down HY bond issuance by year for both Europe and the US. The general trend since our data begins is a slow decrease in average issue life. This tends to hold especially true since 2006 where we have seen the average issued life go from roughly 9 to 6 years. Although the US tends to be a longer duration market, we can see the same type of activity occurring with life decreasing from 9 years to 7 years since 2007. To tie this into DB’s current rates view we look to our World Outlook 5 (published 28 March 2014). DB’s current view is for an upward trend in rates across both Europe and the US with Treasuries roughly 50bps wider (to 3.25%) by Q1 2015 and an 80bps move wider in the EUR 10Y (to 2.35%). By 2016 the forecast shows Treasuries leveling off at 3.50% in the US with the EUR 10Y advancing to 3.25% (a total increase of about 170bps in Europe from current levels). With DB’s current house view in mind, we cannot help but wonder if this is a turning point for average maturity life. With the new issue HY markets being so frothy, we could start to see issuers locking in the current low rates over a longer time frame while still maintaining demand from the investor base. Figure 1: Average Issued Years to Maturity in Europe (left) and the US (right) Fixed Rate Issued (RHS, €Bn) Average Issued Years to Maturity (LHS)

€60

12

€50

11

€40

10

€30

9

€20

8

€10

7

€0

6

Fixed Rate Issued (RHS, $Bn) Average Issued Years to Maturity (LHS)

$400 $350 $300 $250 $200 $150 $100 $50 $0

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999

12 11 10 9 8 7 6 5 4 3

Source: Deutsche Bank, Dealogic, Bloomberg Finance LP

5

http://pull.db-gmresearch.com/p/10680-19B2/38704136/DB_WorldOutlook_2014-03 28_0900b8c087fe28f2.pdf

Deutsche Bank AG/London

Page 47

4 April 2014 Global Fixed Income Weekly

In Figure 2 we again look at the average bond life but this time compare the Average Life to Maturity to the Average Life to Worst/Call for Single-B’s. As we have stated in previous reports, we concentrate on the Single-B cohort as we feel it is the most representative of the overall HY market. Figure 2: EUR Single-B Non-Financials Life to Maturity/Worst 9

Life to Maturity

Life to Worst

8 7 6 5 4 3 2 1 0 2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Source: Deutsche Bank, Bloomberg Finance LP

It’s interesting to note from this chart that not only are we seeing the shortest maturity issuance in history but we are also seeing the EUR HY Single-B market trading at a mere 2.2 years Life to Worst. A lot of this shortening in maturity and call life came from an increased investor demand for shorter duration issues during the crisis. With economic conditions unstable post 2007/2008, investors did not want to take on additional duration risk. As economic fundamentals have begun to look more favourable, investors may be more apt to handle a longer Life to Worst/Maturity going forward. This begs the question as to what proportion of the current HY market is indeed callable and if credits are trading above their respective call price. As one might guess, the majority of the market (52% of the iBoxx EUR HY Non-Fin Index) is both callable and trading above call. Figure 3 illustrates this point for the EUR Non-Fin market. This year roughly 75% of bonds are trading above their call price, with this percentage rising to 90% for those bonds with their next call in 2015 or 2016. Figure 3: EUR Non-Financials Callable/Non-Callable Split at Issuance (left) and Current Callability Trends (right) €60

Callable (€Bn)

€40

Non-Callable (€Bn)

Above Call (€Bn)

Below Call (€Bn)

€35

€50

€30

€40

€25

€30

€20

€20

€15 €10

€10

€5

€0

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

€0 2014 2015

2016 2017

2018 2019

2020 2021

Source: Deutsche Bank Note: Reflects the bonds next call date

Page 48

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

As we can see in Figure 4, bond redemption has been somewhat sporadic and can depend heavily on the micro situation of a Company. That said, last year was a standout year compared to the previous four for redemptions. We think the combination of a more robust issuance environment as well as improved economic sentiment has contributed heavily to this increase. Looking to 2014, we would estimate the same if not better environment for companies to redeem. Figure 4: Total Potential First Calls Redeemed First Call Redeemed (€Bn)

€7

60%

% of Total Potential First Calls

€6

50%

€5

40%

€4 30% €3 20%

€2

10%

€1 €0

0% 2009

2010

2011

2012

2013

Source: Deutsche Bank, Markit

Market Refinancing: Trading Non-Calls and CDS So what does this all mean? Three points standout: (1) We have an expected rising rate environment, (2) current issued life is at its shortest level in history with Single-B’s pointing us to an Average Life to Call of 2.2 years and (3) the majority of the HY market is both callable and trading above call price. These three points lead us to believe that average maturity may have hit a turning point and will now start to rise. We view this potential event as something to be more cognizant of rather than as a catalyst to trade on. In talking with the market, many seem to be conscious that this is happening but are still focusing on new issues rather than re-aligning an entire portfolio for any type of shift in the market although two potential trades do present a case. (1) Non-callable bonds that currently give a strong yield may start to trade at a premium once the HY market begins to refinance and (2) selling callable bonds and 5 year protection via CDS. In Figure 5 we highlight our HY Analysts current Non-Callable bond recommendations and in Figure 6 we highlight our CDS Buy recommendations. We of course encourage further diligence but found the topic interesting amidst what may be a turn in the HY issuance market. Figure 5: Current Non-Callable Buy Recommendations Company

Size

Coupon

Maturity

Moody's

S&P

Current Rec

Analyst

Ask Price

Cable & Wireless

£ 147

8.625%

Mar-19

B1

B+

BUY

Khanna/Velimoukhametova

116.875

Source: Deutsche Bank, Bloomberg Finance LP

Figure 6: Current CDS Buy Recommendations Company

Spread

Current Rec

Analyst

Ardagh

365

BUY

Phelan/O'Neal

Lafarge

197

BUY

O'Neal/Phelan

Cable & Wireless

250

BUY

Khanna/Velimoukhametova

Source: Deutsche Bank Note: Spread levels reflect mids

Deutsche Bank AG/London

Page 49

4 April 2014 Global Fixed Income Weekly

Europe

Credit Covered Bonds Securitisation

Conor O'Toole Research Analyst (+44) 20 754-59652 [email protected]

European ABS update Excerpt from European Asset Backed Barometer at https://gm.db.com/absEurope

On the back of compressing long dated peripheral sovereign yields, senior peripheral RMBS, despite trading up by as much as 50c on the week, continue to look attractive. While the DOURM tender (1,2, and 3) has brought Portugal back into focus this week, Spanish paper in particular looks cheap with spreads in the 5-7 year part of the curve still offering some 60-125 bps pick up. Indeed absolute levels still remain one of the most compelling “safe spread” trades in European ABS (see chart of the week). The RMBS/sovereign spread pick-up is not clear cut across all peripheral jurisdictions however. While like Spain, Irish and Italian RMBS trade 15-75 bps and 25-50 bps wider than the sovereign, Portuguese and Greek senior RMBS trade on top of or just below. We highlight this basis in spread and yield terms through a series of charts in our noticeboard article on page 2 of the European Asset Backed Barometer.

Other articles in European Asset Backed Barometer include – RMBS/ABS Noticeboard - Peripheral RMBS versus sovereign relative value

The unabated bid for spread product can be observed in both recent new issue trading up on the break and strong legacy secondary pricing. The price rally in Quadrivio RMBS 2011-1 A1- some 50c above placement just last week, is an example of the former. The maintenance of a stable AAA CLO 2.0 bid in the face of not insubstantial – EUR 2.5 billion YTD - supply is another case in point. The recent GBP 200 million+ (original face) UK BTL AIREM BWIC trading at 80 bps area, post-crisis record tights, serves as an example of the latter. With no change expected to the ongoing overall ABS market supply constraints - new issue is down 18% y-o-y for Q1 - and sentiment for the current benign macro liquidity backdrop improving daily (think last week’s ECB QE commentary) we expect further spread compression to feature prominently. While Q1 drew to a close with no floating rate European CMBS issuance, Chinese markets sprang to life in mid March with what we understand to be the first cross border issuance since 2006. Backed by 9 retail assets, China Real Estate Asset Mortgages (Bloomberg ticker: CNRAM 1), priced at +200bps for the front pay USD tranche (rated AA3 by Moody's) - noticeably inside the + 225bps pricing for the front pay of the Italian Retail transaction Gallerie in late 2013. In secondary CMBS, the release of the valuation of the Brunel shopping centre (down to GBP 63mn vs GBP 87.2mn in October 2011 presumably driven by the valuer getting a firmer handle on non recoverable costs) in DECO 6 sent the A2 note to what we consider to be more rational indicative pricing (mid 70s). This represents a sharp correction from the mid-high 80s that existed in early March and on which we expressed concern, and is consistent (after cash trapping) with the lower boundary of the GBP 70-90mn that we forecast in our earlier publication "UK CMBS - A Secondary Renaissance". We still do not rule out the final outcome on the loan being more towards the upper end of our range, but this will be contingent on an upturn in UK secondary retail markets and/or a way being found to fund the opening of the centre's extension. Chart of the week: Spanish yield curves – RMBS versus sovereign Spanish RMBS

5

IMPAS 4 A

4

Yield %

Spanish sovereign BCJAM 4 A2

TDA 27 A2

TDAC 5 A

UCI 6 A

3

BFTH 8 A

TDAC 4 A

2

UCI 16 A2

1 0 0

2

4

6 WAL (yrs)

8

10

12

Source: Deutsche Bank trading for prices, recent repayment rates used to determine CPRs, zero default assumed, Bloomberg Finance LP for modeling. Levels are purely indicative and offer-side

Page 50

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Global

Credit Covered Bonds

Bernd Volk Strategist (+41) 44 227-3710 [email protected]

Covered Bond and Agency Update „

Q1 public covered bond issuance by Eurozone banks accounted for 25% of total bond issuance, a historical Eurozone low (compared to 29% in FY 2013 and 32% in FY 2012). With 14%, the share of subordinated bond issuance by Eurozone banks was at a Eurozone high, even higher than the 12% in FY 2007. Q1 issuance of unsecured bonds by Eurozone banks accounted for 57% compared to 59% in FY 2013 and 61% in FY 2012. A historically low share of covered bond issuance and historically high share of subordinate bond issuance by Eurozone banks in Q1 confirms the strongly improved market perception and also increased the fundamental buffer of Eurozone covered bonds.

„

New 10Y EUR 500m Unicredit mortgage Pfandbriefe (Aa1/-/AA), priced at ms+12bp, had a 74% share of bank investors and 10Y EUR 500m Baylaba public Pfandbriefe (Aaa/-/AAA), priced at ms+8bp, had a 47% share of bank investors, showing that banks also buy long dated covered bonds, likely due to the low yield environment.

„

The ongoing bullish sentiment was confirmed by Bankia outperforming Santander Cedulas and Monte Dei Paschi bonds outperforming Unicredit covered bonds. Besides general strong demand for higher yielding bonds, the latter was supported by MPS Foundation selling a 6.5% stake in Monte Dei Paschi (4.5% to Fintech and 2% to BTG Pactual), making Blackrock the largest shareholder with 5.7% (compared to 5.5% of MPS). We expect MPS covered bond spreads to tighten somewhat further versus Unicredit covered bonds. MPS's recent EUR 500m 5Y unsecured bond being an outperformer this week and tighter by more than 20bp from its ms+275bp reoffer supports this view.

„

Fitch is rolling-out the amended covered bond criteria, mainly driven by BRRD. Hence, the increased potential uplift for covered bonds will, over the medium term, be balanced with lower expected support for banks. While most covered bonds have been affirmed by Fitch at the current rating with a stable outlook, UBI (A+), Unicredit (A+), Santander (A), Caja Laboral (A-) and NBG (B+) have been put on positive outlook. However, the negative outlooks (for example, Carige BBB+, MPS A, BP BBB+, Bankia BBB+) clearly outnumber the positive outlooks. Nevertheless, as rating downgrades are anything but certain and a medium term issue anyway, we do not expect a significant spread impact at this stage.

„

With new 10Y German federal states pricing in the primary wider than new 10Y Baylaba public Pfandbriefe (ms+8bp), we highlight our strong view regarding states. More than 90% of revenues are from shared taxes and transfers and federal states have only limited tax rate-setting rights. Moreover, besides the strong revenue equalisation system, there is a constitutional Federal Solidarity Principle, according to which all federal states must show mutual solidarity in the event that one of them or the Bund faces severe hardship. According to the debt-brake, after 2019, the 16 states are no longer allowed to have structural fiscal deficits, supported by joint supervision. Strong 2013 budget data of German states, an aggregate deficit of only EUR 500m, confirm the positive trend.

Deutsche Bank AG/London

Page 51

4 April 2014 Global Fixed Income Weekly

„

On Tuesday, the vdp announced to cease to publish Pfandbrief curves citing burdensome supervisory requirements. Given that there is a lack of such data and most market participants will probably use secondary market index data as indication now, Pfandbrief curves were definitely useful. According to Börsen-Zeitung, participating banks are concerned regarding potential legal claims. The vdp and also BaFin are quoted in the article that the supervisory requirements in question do not directly apply to rates generating the Pfandbrief curve. However, given that Pfandbrief curves have been used by market participants to draw conclusions regarding funding costs of banks and also regarding loan pricing it seems vdp member banks obviously decided to play it safe.

Page 52

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

United Kingdom

Rates Gov. Bonds & Swaps Inflation Rates Volatility

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

UK Strategy View: The PMI surveys for March all posted declines though remain above long term averages, indicating slowing but still comfortably strong growth rates. The decline in price balances reflect a global phenomenon, with surveys also pointing to ongoing strength in the labour market, and tighter capacity utilization (the agents survey for example). The move higher in core rates has partially corrected the richness in 5Y5Y, but it has been too early to see a flattening of front end, even while the OIS curve continues to reprice the likely scenario of rate hikes. 1Y1Y OIS now reflects 2.5-3 hikes (depending on the basis assumed), with Dec 15 OIS at around 1.27%. We retain the view that there is room for the front end to flatten, though this is likely to require a step up in hawkish rhetoric and/or a turnaround of inflation dynamics. 2Y5Y10Y vs 1Y1Y

May Aug Nov 2.50 Quarterly OIS % 1.86

2.00

2.00

2.09

2.23

2.5

60 1Y1Y OIS

2.0

1.44

1.50

1.22 1.02

1.00

0.81 0.43 0.50

0.50

40

2Y5Y10Y sw

1.65

0

1.0

-20 -40

0.62

0.00 0.00

Source: Deutsche Bank

20

1.5

0.5 0.0 Mar-10

-60 Mar-11

Mar-12

Mar-13

-80 Mar-14

Source: Deutsche Bank, Bloomberg Financial LP

Data Recap: The PMI surveys declined, though remain above longer term averages (eg composite new orders z-score +1). The drivers of weakness are not solely from activity balances; the decline in price balances has also contributed, now well below long term averages (input prices z-score -0.9) The decline in imported inflation pressures is a positive supply shock, helping support real incomes. At the same time the employment component, though posting a monthly decline, remains also well above long term averages (composite employment z-score +1.2). The Q1 credit conditions survey showed a continued improvement in availability of household credit, with the HtB cited as the main factor behind the “significant increase in availability “of high LTV products. Demand conditions were also reported to have increased, with a pickup in remortgaging in particular expected to drive demand going forward (though arguably there has been limited accuracy in the forward looking components).

Deutsche Bank AG/London

Page 53

4 April 2014 Global Fixed Income Weekly

Credit conditions 50 40

0.50

1.0 0.0

0.00

-1.0 -2.0

0 -10 -20

Availability Next 3M

-50

Source: Deutsche Bank, Haver Analytics

Jun-11

Dec-11

Jun-10

Dec-10

Jun-09

Input Px Output Px

Dec-09

Emp

-60 Jun-08

-1.00

Dec-08

Z-Score rhs

Availability Past 3M

-40

Jun-07

-5.0

-0.50

Dec-07

-4.0

Current vs Ave New Orders

10

-30

-3.0

1mChange

Output

20

Jun-14

2.0

Mortgage availability

30

Jun-13

1.00

3.0

Dec-13

4.0

Jun-12

1.50

5.0

Dec-12

PMI: changes vs levels

Source: Deutsche Bank, BoE

Corporate credit conditions were also positive, though as has been the case previously, varies based on size. Demand was also reported to be strong, particularly for larger corporates, and expected to remain so going forward. Non price lending terms were also reported to have eased, alongside a decline in spreads. Spreads tightening, but sector

Strong demand seen persisting

dependent 80 60

Demand Past 3M

60

Demand Next 3M

40

40

20

20

0

0

-20 -40

Medium Next 3M

Feb-14

Apr-13

Aug-11

Mar-11

Oct-10

May-10

Jul-09

Dec-09

Feb-09

Apr-08

Sep-08

Sep-13

Large Next 3M

-80 Jun-07

Jun-14

Jun-13

Dec-13

Jun-12

Dec-12

Jun-11

Dec-11

Jun-10

Dec-10

Jun-09

Dec-09

Jun-08

Dec-08

Jun-07

Dec-07

-80

Source: Deutsche Bank, BoE

Small Next 3M

-60

Jun-12

Mortgage demand

Nov-07

-60

Nov-12

-40

Jan-12

-20

Source: Deutsche Bank, BoE

Overall, it’s clear the improvement in credit conditions has continued, with respondents citing the improvement in economic outlook and risk appetite and little apparent impact of the FLS pull back announced in November 13. Thus far, BoE comments have recognised "increasing momentum" in the housing market (and commercial real estate) but point to a number of indicators being below long run averages. There seem to be a few flashing lights, enough to warrant monitoring and a pull back of FLS but no further action at this juncture. The stress tests and focus on testing against the resilience against a housing or rates shock do point at an attempt to tackle the issue from another angle in the mean time. It was announced last week that the April meeting would be shortened to a one day meeting due to clashes with international meetings; no changes in policy expected. RV: UKT 18-19 slope has flattened, 19 look rich vs surrounding bonds.

Page 54

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

18_19_20 Fly

Ave 18s-Ave19s vs 10Y

0.16

y = 0.1523x - 0.0803 R² = 0.7106

0.4

0.14

0.35

0.12

0.3 0.25 0.2

18s_19svs10Y:0

0.10

Current

0.08

0.15

0.06

0.1

0.04

0.05

Fly UKT_3_18 UKT_3_19 UKT_3_20

0.02

0 2.5

2.6

2.7

2.8

2.9

3

Source: Deutsche Bank

3.1

0.00 Oct-13

Nov-13

Dec-13

Jan-14

Feb-14

Mar-14

Source: Deutsche Bank

The issuance calendar for Q2 was released, with little clarity on the maturity of the long to be syndicated in June. The weighted average duration of conventional supply assuming it is a new 30Y is 10.6Y and around 11.5Y if it is a tap of the 68 (vs 7.8Y last quarter, and 11Y in same period in 2013). However, after the tap of the UKT 44 this week, there will be no 30Y+ nominal supply until the syndication. The curve remains extremely rich relative to levels; but the lack of supply in the near term could keep valuations rich in the mean time. Residual around long end supply

Little change in the flatness 20 15

60

Residual: Dec 44 vs Sep 23 or Sep 22

40

10

20

5

0

0

-20

-5

-40

-10 -15 -20 Oct-12

Mar-22

-60

Sep-22

-80

Jan-13

Apr-13

Jul-13

Source: Deutsche Bank

Deutsche Bank AG/London

Oct-13

Residual 10Y30Y on 10Y 30Y+ Syndications

Jan-14

Source: Deutsche Bank

Page 55

4 April 2014 Global Fixed Income Weekly

Japan

Rates Gov. Bonds & Swaps

Makoto Yamashita, CMA Strategist (+81) 3 5156-6622 [email protected]

Japan Strategy Overview „

Manufacturers have been very cautious in their production activity of late, with no sign of inventory buildup despite a surge in demand ahead of the April 1 consumption tax hike. We therefore see little risk of production adjustments subtracting from economic growth going forward. There are also some signs that a recovery in wage growth might soon signal an exit from deflation, with the March Tankan survey results pointing to significant personnel shortages (although many firms do expect business conditions to deteriorate now that the 8% consumption tax has taken effect).

„

The majority view among market participants is that the BOJ will deploy further monetary easing measures before the end of this year, with June or July considered the most likely timing. This apparently reflects a belief that the central bank will take further action if and when it becomes apparent that the impact of the consumption tax hike on economic activity and prices will make it difficult to achieve +2% inflation. However, it is also possible that the BOJ will launch a preemptive strike at the first sign its inflation target might be in jeopardy, or that the economy and prices will continue to follow the trajectories projected by the BOJ. In any case, we see very little prospect of the Kuroda-led BOJ falling behind the curve.

Recession risk limited, but still waiting for wages to rise Industrial production data for February were indeed indicative of recent weakness, but signs of caution over the near-term outlook are perhaps worthy of greater attention. Output was cut back 2.3% month-on-month in February as snowstorms seemingly had an impact, but the survey of production forecasts points to unchanged output in March followed by a 0.6% MoM decline in April. This would leave average output for 1Q 2014 up just 2.8% from the previous quarter. Inventories also fell 0.6% MoM in February as shipments (–1.0%) were cut back by less than output. It would therefore appear that the supply side of the economy has remained very cautious despite a surge in demand ahead of the April 1 consumption tax hike. The flipside is of course that a 0.6% MoM contraction in production activity for April would be relatively small, and that the economy appears to be at little risk of falling into a recession due to inventory-driven production adjustments. That said, domestic production and exports have clearly risen by less than many would have expected. The capacity utilization for manufacturers was 101.3 as of January, almost 10% short of the 110 threshold beyond which firms typically start to ramp up their capital investment. It should therefore come as little surprise that the corporate sector as a whole has remained reluctant to increase its capex. This sort of cautious behavior may help to prolong the current economic recovery, but is also likely to prevent growth from accelerating sufficiently to generate significant inflationary pressure. The recent stability of domestic interest rates is consistent with this state of affairs and, under more “ordinary” circumstances, might be seen as a good justification for the BOJ deploying additional easing measures.

Page 56

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 1: Domestic production capacity and utilization 2010=100 130

Capacity utilization (lhs)

Production capacity index (rhs)

120

2010=100 110 105

110

100

100 95 90 90

80

85

70 60

80 80

85

90

95

00

05

10

Source: Ministry of Economy, Trade and Industry, Deutsche Securities

Wage growth also remains weak at best. Monthly Labour Survey data for February showed total cash earnings unchanged from a year earlier. Scheduled pay declined 0.3% YoY, but this did represent somewhat of an improvement from –0.7% YoY in 4Q 2013. Moreover, scheduled pay of fulltime workers fell 0.2% YoY after rising 0.1% in January. An acceleration of growth in the number of full-time workers to +0.9% YoY was a positive sign, but it is important to recognize that the February data are still preliminary, which makes it difficult to conclude that deflationary pressures have eased to any meaningful degree. However, wage hikes should not be considered at all surprising in the current climate of personnel shortages. Quarter-ahead forecasts in the March Tankan survey showed that many firms expect business conditions to deteriorate now that the consumption tax has been raised to 8%, but the all-enterprises employment conditions DI (“excessive employment” minus “insufficient employment”) actually fell from –10 in the December survey to –12, indicating that many firms consider themselves shorthanded. Moreover, just a 1pt improvement to –11 is projected for the quarter through June, with only large manufacturers forecasting a positive DI. A level of –12 on an all-enterprises basis was last seen in the March 2007 survey, and before that in September 1992. Increases in base pay arising from shunto spring wage talks appear set to boost the scheduled cash earning of full-time workers by around 0.4% to 0.5%, and wages for irregular workers also appear to be facing upward pressure, with a number of major temporary staffing firms reportedly seeking substantial hourly pay increases from their clients and figures for part-time workers in Japan’s three major metropolitan areas pointing to growth in the order of +3.0% YoY. We therefore believe that macro-level wages are much more likely to rise than fall going forward.

Deutsche Bank AG/London

Page 57

4 April 2014 Global Fixed Income Weekly

Figure 2: Tankan employment conditions DI vs. scheduled cash earnings (year-on-year, %) 6.0

Change in scheduled cash earnings (lhs)

(% pt)

Employment conditions DI (rhs, inverted)

-50

5.0

-40

4.0

-30

3.0

-20

2.0 -10 1.0 0

0.0

10

-1.0

20

-2.0 -3.0

30 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13

Source :Bank of Japan, Nikkei NEEDs, Deutsche Securities

Will the BOJ take preemptive action, or no action at all? The nationwide core CPI inflation rate for February of +1.3% YoY was in line with the BOJ’s projections, and Governor Haruhiko Kuroda indicated in his March press conference that he currently sees no need to adjust monetary policy. However, just 15% of respondents in the Nikkei QUICK monthly bond survey for March said that they expect monetary policy to remain on hold throughout at least the remainder of 2014, with 34% anticipating a July easing announcement while 16% opted for June and 11% for October. This “gap” suggests that many market participants believe that overseas economic uncertainty and the impact of the April consumption tax hike will cause domestic GDP growth and inflation to drop below the BOJ’s projected trajectories. Many market participants expect the pace of CPI inflation to slow from around May as the impact of previous yen depreciation continues to dissipate, in which case the BOJ could be confronted with some unwelcome data (including weak consumer spending figures) at the end of June. It is therefore possible to make a credible case for the BOJ revising its forecasts and announcing further easing measures in either July or October. However, the BOJ policy board is of course well aware that the yen has stopped weakening, that its growth target for FY2013 is now out of reach, and that negative growth in the April–June quarter would increase calls for additional easing. The BOJ’s April 2013 Quantitative and Qualitative Monetary Easing (QQE) measures were so drastic that the scope for further action is limited, which suggests to us that timing will be of the utmost importance and that an April easing announcement should not be entirely ruled out. A preemptive strike of this nature should help to prevent weak economic indicators from April onwards from causing too much market anxiety, and would indeed be consistent with the Kuroda-led BOJ’s proactive approach to date. Finally, it is also possible that the BOJ—rather than the market consensus—will turn out to have been correct, with the economy and prices continuing to follow the trajectories projected by the central bank and thereby rendering additional action unnecessary.

Page 58

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Figure 3: When will the BOJ ease again? A pril

6%

May

4%

June

16%

July

34%

A ugust

5%

September

8%

October

11%

November-December

1%

2015 or later

2%

No further easing

13%

Source: The Nikkei Quick monthly survey for March, Deutsche Securities

Deutsche Bank AG/London

Page 59

4 April 2014 Global Fixed Income Weekly

Global

Rates Gov. Bonds & Swaps Inflation Rates Volatility

Jerome Saragoussi Strategist (+33) 1 4495-6408 [email protected]

Global Relative Value We recommend two trade ideas in Europe. The first trade is a way to position asymmetrically for a Fed-style QE by the ECB; the second trade is a carry efficient way to position for a recovery in risk premium in Europe. „

Conditional EUR 5s10s30s butterfly with 3M receivers: In a world of anchored policy rates, the 5s10s30s 50/50 fly is an excellent gauge of QE expectations, in our opinion. In the US the butterfly has been an outstanding proxy for the 10y real rate in the post Lehman regime, richening in a buildup of QE expectations (ahead of a QE announcement) and cheapening gradually post QE announcement as the market prices out the remaining amount of purchases. If the ECB embarks in a Fed-style QE it should also richen the EUR butterfly in a rally: (1) QE would reinforce the low for long policy and bull-flatten the 5s10s sector (2) QE would increase the inflation risk premium at the long end and 10s30s would tend to steepen relative to 5s10s. The EUR 5s10s30s fly has already richened in anticipation of a more dovish ECB and it is currently at the level of the USD fly ahead of QE2 announcement. History shows it could richen much more if QE expectations were to increase further (USD fly was richer by 7bp ahead of QE infinity). From a RV perspective, 5y sector looks rich, and 10s30s looks too flat outright and against 5s10s, which should help 10s outperforming in further rally, maintaining the strong recent directionality of the fly with 10s. The conditional fly with receivers is an excellent proxy for a 3m10y receiver but the cost is reduced by 60% and carry is improved by 25%.

„

EUR 1Y2Y-10Y20Y steepener: The ECB bias towards further monetary policy easing implies an asymmetrical steepening risk in 2s30s and 1y fwd 2s30s. The front end can be compressed further by a lower for longer policy resulting from the likely implementation of a form of QE, but also potentially by direct asset purchases at the front end, by rate cuts or by liquidity injections. At the time of writing the Eonia strip is only pricing in a 1-year lag between the first ECB hike and the first Fed hike. The market is indeed pricing in the first ECB hike in Sep-2016, in about 2.5 years. Another round of stimulus could well delay policy normalization in Europe by 6 months which would be particularly beneficial to the 3y sector hence a preference for 1y fwd 2s30s steepeners. As for the long end, it should underperform as the market starts pricing in more inflation risk premium. The EUR 2s30s slope stands out as being excessively flat by more than 10bp relative to the level of short rate, long term inflation expectations, expected Eurozone fiscal deficit, implied rates vol driving convexity at the long end, and risk aversion. A recovery in European breakevens should create steepening pressures. While a 5s30s steepener would offer more positive carry, the 5y sector stands out as being rich across the curve: the 2s5s10s remains close to historically rich levels while the 5s30s slope looks too flat by 5bp only. In other words we prefer optimizing this slope view by expressing it with a carry-efficient forward slope proxy. We recommend the 1y2y-10y20y steepener an excellent proxy with 50% additional carry. Moreover the 10s30s slope is currently too flat: its normalization and repricing of risk premium is likely to bear steepen 10s30s, leading to a likely underperformance of the 10y20y rate vs. 1y30y.

Page 60

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

#1- Buy EUR 5s10s30s conditional fly with 3M receivers We first recommend buying the EUR 5s10s30s conditional fly with 3M receivers, i.e. buy 3M10Y receivers against 3M5Y and 3M30Y receivers (50/50 delta split). The butterfly is increasingly directional with the 10y rate and tends to be an excellent gauge of QE expectations in a world of anchored policy rates. The ECB has made it clear that further deterioration of data in a context of persistently soft inflation could trigger an unconventional policy response. If this response takes the shape of a Fed-style QE, there is still room for further richening of the fly in a rally. In a sell-off, driven by improvement in economic data leading to a pricing out of QE expectations, the butterfly would likely cheapen but receivers would end up out of the money. (1) 5s10s30s fly as a gauge of QE expectations: In a world of anchored policy rates, we note the 5s10s30s 50/50 fly is an excellent gauge of QE expectations. In the US, this butterfly has richened systematically ahead of a QE announcement and cheapened thereafter with the gradual decline of QE expectations (see figure 1). Any Fed-style QE involving a concentration of purchases of government bonds in the 5y-10y sector tends to create flattening pressures in 5s10s while QE purchases tend to increase the likelihood of ‘reflation’ of the economy and boost the inflation risk premium leading to a relative steepening of 10s30s vs. 5s10s, hence a richening of the 5s10s30s fly. Conversely the decline in QE expectations steepens back 5s10s and reduced the risk premium in 10s30s hence a 5s10s30s cheapening. The fly has been such an efficient barometer of QE expectations than it has been more directional with the 10Y real yield than with the 10y swap itself (see figure 2). Figure 1: The US experience shows that the 5s10s30s fly

Figure 2: US 5s10s30s has been the best proxy for a 10Y

tends to be an excellent gauge of QE expectations

real yield, the ultimate asset to benefit from QE

30 25 20

1.50

30 UST 5s10s30s butterfly as QE expectations barometer QE2

Delay in QE tapering QE tapering

Twist

15 10

25

US 5s10s30s 50/50 fly (lhs)

1.25

TIPS 10Y real yield (rhs)

1.00 0.75

20

0.50 0.25

15

QE 3

0.00

5

10

0

-0.25 -0.50

5

-0.75

-5

-15 Jan-10 Jul-10

-1.00

0

-10

-1.25 Jan-11 Jul-11

Source: Deutsche Bank

Jan-12 Jul-12

Jan-13 Jul-13

-5 Nov-10

Jul-11

Mar-12

Nov-12

Jul-13

-1.50 Mar-14

Source: Deutsche Bank

(2) The EUR 5s10s30s can embed more QE expectations – If the ECB embarks in a Fed-style QE, the same kind of dynamics as in the US should be observed (richening in a rally as QE expectations compress 5s10s relative to 10s30s which benefits from higher inflation risk premium), particularly as the ECB would not commit to buying long assets in the 30y sector. The EUR 5s10s30s fly has already richened in the past few months, partly in anticipation of a more dovish ECB and it currently trades at the level of the USD fly prevailing ahead of the QE2 decision (QE2 size was moderate in terms of 10y-equivalents). Based on the USD fly history, the EUR fly could keep richening if QE expectations were to increase further: when the US curve was pricing in the QE infinity announcement, the USD 5s10s30s butterfly was trading 7bp richer than the EUR butterfly. Deutsche Bank AG/London

Page 61

4 April 2014 Global Fixed Income Weekly

Figure 3: US experience suggests that further richening of theEUR 5s10s30s is possible if QE expectations increase further 40

EUR 5s10s30s USD 5s10s30s

35 30 25 20 15 10 5

QE2

0 -5 Mar-09

QE inf inity Dec-09

Sep-10

Jun-11

Mar-12

Dec-12

Sep-13

Source: Deutsche Bank

(3) .From a RV perspective, the 5y sector looks rich, and 10s30s looks too flat outright and against 5s10s, which should help 10s outperforming in further rally, maintaining the strong recent directionality of the fly with 10s. The richness of the 5y sector can be observed via the outright level of the 2s5s10s fly which is at the bottom end of the historical range and looks excessively rich against money market slopes, but also against 1y1y Eonia and ECB excess liquidity (see figure 4 below). The excess flatness of the 10s30s slope by more than 5bp can be seen in our market model which captures the directionality with the short rate, with long term inflation expectations, risk aversion, implied rates vol to capture convexity, and deficit expectations. It is also confirmed by the relationship between the 5s10s slope and the convexity-adjusted 10s30s slope (using 1-factor Merton model) which shows some residual excess flatness of the convexity-free 10s30s slope relative to the 5s10s slope (see figure 5). Figure 5: EUR 10s30s (convexity adj.) too flat vs. 5s10s

Figure 4: EUR 5Y sector is on the rich side 5.0 4.0

EUR 5Y -10Y swap slope

150 EUR 2s5s10s f ly v s. model f it (1y 1y eonia, ECB excess liquidity

130

3.0

160

Conv exity adjusted 10Y -30Y slope (rhs)

140

110

120

90

100

70

80

50

60

30

40

(3.0)

10

20

(4.0)

(10)

2.0 1.0 (1.0) (2.0)

(5.0) Jan-13 Source: Deutsche Bank

(20)

(30) Apr-13

Jul-13

Oct-13

Jan-14

Apr-14

03

04

05

06

07

08

09

10

11

12

13

14

Source: Deutsche Bank

(4) The conditional fly with receivers is an excellent proxy for a 3m10y receiver, in our opinion, but the cost is reduced by 60% and carry is improved by 25%. Indeed the EUR 5s10s30s fly has moved with a stable 20% beta over the past 12 months so that we can compare the carry on the 3M forward fly to the 3M carry on the beta-adjusted equivalent position in a 3M10Y swap. 3M carry on the fly is 1.7bp which should be Page 62

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

compared to 20% x 6.8bp = 1.35bp. In other words the 3m forward butterfly provides a 25% improvement in positive carry relative to an outright long position in 3m10y. The cost of the conditional fly with ATMF 3M receivers is 0.8bp running which should be compared to the cost of the beta-adjusted equivalent position in a 3M10Y receiver i.e. 20% x 10.6bp = 2.1bp running. In other words the conditional fly is 60% cheaper than the equivalent position in a 3M10Y receiver. Figure 7 – Conditional fly cheap proxy to 3m10y receiver

Figure 6: Strong directionality of the fly with 10s 25

2.6

EUR 5Y -10Y -30Y butterf ly EUR 10Y rate (rhs)

4.0 3.5

Cost of equiv alent beta weighted ATMF receiv er in the belly Conditional butterf ly ATMF cost in bps

2.4

20

3.0 2.2

15 2.0 10

2.5 2.0 1.5

1.8 1.0 5

0 Jan-13

1.6

Apr-13

Jul-13

Oct-13

Jan-14

Source: Deutsche Bank

1.4 Apr-14

0.5 0.0 Aug-13

Oct-13

Dec-13

Feb-14

Apr-14

Source: Deutsche Bank

Risks of the trade: If data improves and the market prices out the likelihood of a QE program then 10s should underperform across the curve hence a cheapening of the butterfly but the receivers will expire out of the money limiting the loss to the small premium paid. A second risk to the trade would be a breakdown of directionality due to an outperformance of the 30Y sector in a context of systemic crisis and forced receiving by PFs.

#2 – EUR 1Y2Y-10Y20Y steepener We recommend a EUR 1Y2Y-10Y20Y steepener as a carry efficient way to position for a repricing of the risk premium embedded in the EUR curve on the back of heightened QE expectations in Europe. The ECB bias towards further monetary policy easing implies an asymmetrical steepening risk in 2s30s and 1y fwd 2s30s. (1) The front end can be compressed further by a lower for longer policy resulting from the likely implementation of a form of QE, but also potentially by direct asset purchases at the front end, by rate cuts or by liquidity injections. At the time of writing the Eonia strip is only pricing in a 1-year lag between the first ECB hike and the first Fed hike (priced in by the market for Aug-2015). The market is indeed pricing in the first ECB hike in Aug-2016, in about 2.5 years. Another round of stimulus could well delay policy normalization in Europe by 6 months which would be particularly beneficial to the 3y sector hence a preference to receive 1Y2Y. Given the particularly slow pace of rate hikes priced in around 40bp per year beyond 2016, the 5Y sector is already pricing in an extremely soft rate hike cycle and we find some richness in the 5y sector against the wings (see figure 2 below). From this perspective the additional carry offered by forwards in the 5Y sector instead of 3Y sector is not worth the risk of underperformance.

Deutsche Bank AG/London

Page 63

4 April 2014 Global Fixed Income Weekly

1-

2-

Eonia strip could be lower in the 3Y sector

1.25

25

40bp hikes per y ear

Eonia strip

Richness of the 5y sector makes us favor 3Y sector 180 160 140 120 100 80 60 40 20 0 -20 -40 -60 -80 -100

EUR 2s5s10s f ly

20

ER4-ER8 money market slope (rhs)

15

1.00

10

0.75

5

1st hike in Aug 2016

0

0.50

-5

Remov al of excess liquidity by Oct 2015

0.25

-10 -15 -20

0.00 Jul-14 Oct-14 Jan-15 Apr-15 Oct-15 Apr-16 Aug-16 Apr-17 Apr-18 Source: Deutsche Bank

-25 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 Source: Deutsche Bank

(2) As for the long end, it should underperform as the market starts pricing in more inflation risk premium. The EUR 2s30s slope stands out as being excessively flat by 12bp (see figure 3) relative to the level of short rate, long term inflation expectations, expected Eurozone fiscal deficit, implied rates vol driving convexity at the long end, and risk aversion. A recovery in European breakevens should also create additional steepening pressures. An update of our 5s30s slope model indicates much milder excess flatness by ~5bp confirming the relative richness in the 5Y sector vs. 2-3Y sector. (3) Carry optimization: We prefer optimizing the 1Y fwd 2s30s steepening view by expressing it with a carry-efficient forward slope proxy rather than via 5s30s steepeners. We recommend the 1y2y-10y20y steepener as an excellent proxy for 1y fwd 2s30s with a 95% R-2 and 1.25 beta over the past few years (see figure 4) and with the benefit of 50% additional positive carry (26.4bp on 1y2y-10y20y to be compared to equivalent beta adjusted carry on 1y2y-1y30y i.e. 1.25 x 14bp=17bp). The additional carry comes from the remarkable 3bp 1Y roll up on 10Y20Y which is in sharp contrast with the 10bp of 1Y roll down on 1Y30Y. Moreover the 10s30s slope is currently too flat by 5b. Its normalization and the repricing of the inflation risk premium is likely to bear steepen 10s30s. There is therefore room for an underperformance of 10y20y vs. 30y. 3-

4-

Excess flatness of 2s30s in our model by 12bp

300 250

Theoretical slope implied by the model

300

EUR 2Y-30Y swap slope

270

1y2y-10y20y is perfect proxy for 1y fwd 2s30s 250

EUR 1Y 2Y -10Y 20Y slope (lhs) EUR 1y f wd 2s30s

240

200

210

200

180

150

150

150 120

100

100

90

50

50

60 30

0

0

0

-50

-30

-100 05

06

Source: Deutsche Bank

Page 64

07

08

09

10

11

12

13

14

-60 Oct-05

-50 Apr-07

Oct-08

Apr-10

Oct-11

Apr-13

Source: Deutsche Bank

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

The risk of the trade would be a policy mistake by the ECB leading the market to price in higher Eonia strip at the front end and a higher risk of deflation at the long end via a lower 30y sector, leading to further flattening pressures.

Deutsche Bank AG/London

Page 65

4 April 2014 Global Fixed Income Weekly

Asia

Rates Gov. Bonds & Swaps Rates Volatility

Asia „

„

China: We believe there are two implications of PBoC’s Q1 MPC statement: (a) monetary policy in Q2 is likely to remain prudent and similar as it was in Q1; (b) given the relaxation of liquidity condition on the back of the 2.64% YTD depreciation in RMB against the USD, we think the government is more likely to focus on fiscal stimulus in Q2. India: We close our long bond position and switch to marketweight on duration, given the persistence of a broadly hawkish monetary policy bias amidst an uncertain inflation outlook, rising funding costs, resumption of supply with further upside risk to the issuance numbers, and a likely binary election outcome with risk-reward increasingly disfavoring rupee bulls at current levels.

Sameer Goel Strategist (+65 ) 64236973 [email protected] Linan Liu Strategist (+852) 2203 8709 [email protected] Kiyong Seong Strategist (+852) 2203 5932 [email protected] Swapnil Kalbande Strategist (+65) 6423 5925 [email protected]

China No change in monetary policy bias PBoC held its Q1 monetary policy meeting on April 3rd and we highlight three points from the statement: (a) the central bank thinks economic growth remains within the reasonable range, financial market is broadly stable and general price level remains stable; there are both downside and upside risks to the current economic conditions; globally, it sees diverging economic performance with more signs out of US/Europe pointing to positive growth momentum, while some EM economies continue to decelerate; (b) compared with the previous MPC statement at the end of last year, there is no change in the bias of monetary policy; (c) the only change in the specific wording of monetary policy is that the central bank revised “push forward reforms during macroeconomic rebalancing” in the previous statement to “continue to deepen financial structural reforms”. We believe there are two implications: (a) monetary policy in Q2 is likely to remain prudent and similar as it was in Q1; (b) given the relaxation of liquidity condition on the back of the 2.64% YTD depreciation in RMB against the USD, we think the government is more likely to focus on fiscal stimulus in Q2. Commercial banks favored higher carry investments in Q1. Q1 interbank bond holdings statistics show that the total holding balances in CGBs and medium term notes fell by RMB39bn and RMB111bn respectively, reflecting the slower pace of supply relative to redemption. The balance of policy bank bonds and enterprise bonds were up by RMB441bn and RMB136bn respectively. Out of the top three domestic bond investors, namely commercial banks, insurance companies and fund management companies, commercial banks reduced their holding of CGBs and MTNs in Q1, while increased investment in policy bank bonds and enterprise bonds. Specifically, commercial banks bought 58% of net supply of policy bank bonds in Q1, and 30.5% of net supply of enterprise bonds. Insurance companies bought 4% of net supply of policy bank bonds and 6.2% of net supply of enterprise bonds. Fund managers bought 16% of the Q1 net supply of policy bank bonds and reduced their investment in enterprise bonds and MTNs in Q1. We think the statistics is very consistent with what we have seen in the market, particularly for commercial banks, where allocation demand for their investment books has been relatively slow compared with Page 66

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

last year. We think unstable deposit base and cost pressure on their funding will continue to depress their demand for CGB bonds. Q1 cash bond holdings change (RMB bn)

Net change in cash bond holding in Q1 by investors (RMB bn) 350

500 400

changes in interbank bond holdings in Q1

300

MTNs

300

Enterprise bonds

250

Policy bank bonds

200

CGB

150

200

100

100

50 0

0 -50

-100

-100

-200 CGB

Policy bank bonds

Source: Deutsche Bank, CEIC

Enterprise bonds

MTNs

-150 Commercial Banks

Insurance

Fund companies

Source: Deutsche Bank, CEIC

Q2 CGB supply details out: The MoF today released the Q2 CGB supply plan. It plans to offer 3Y and 5Y savings CGBs on the 10th of April, May and June and we estimate the gross supply of savings CGBs to amount to RMB120bn. It will conduct fourteen CGB auctions in Q2 with an estimated gross supply of RMB392bn and an average maturity of 9.9 years. We estimate the net supply of CGBs amounts to RMB167bn, about 119% up from Q1. Specifically at the long end, there three scheduled auction in 7Y, three in 10Y, on each in 20Y and 50Y tenor. We remain of the view that cash CGB bonds, particularly long dated bonds are unlikely to rally much despite comfortable liquidity given the risk of duration supply in Q2.

India This fixed income story will have to wait. We close our long bond position, and switch to marketweight on duration. A few factors have driven our decision. RBI is unlikely to let down its guard on inflation anytime soon. RBI maintained its broadly hawkish tone in the most recent MPC meeting concluded on 1-Apr. The central bank said that recent disinflation was attributed to a helpful vegetable price dynamic, but further softening is unlikely and it sees upside risks to their central forecast of 8% CPI inflation by Jan 2015.We thus don’t see any signs of RBI letting down its guard on inflation anytime soon, especially as it remains committed to its disinflationary glide path for the CPI to 8%, and 6% by Jan 2015 and Jan 2016 respectively. Cost of funding is unlikely to ease soon. Cost of funding is unlikely to ease soon. While the central bank is likely to keep policy rate on hold in the near term, it is gradually tightening the cost of liquidity. Arguably, it’s recent move to reduce the liquidity provided under LAF to 0.25% of NDTL from 0.50% earlier and increase the same to 0.75% from earlier 0.50% for term repos, is a step towards developing term repo market in line with Urijit Patel committee recommendations. But, at the same time these measures also increase the cost of funding marginally as banks will have to resort to the term repo market for funding needs beyond 0.25% of NDTL, which comes at a higher cost than

Deutsche Bank AG/London

Page 67

4 April 2014 Global Fixed Income Weekly

the policy repo rate available under LAF. Thus, to the extent these measure impact the cost of funding, downside room for front-end rates is limited. Front-end OIS is fairly priced - implying no rate cut and no normalisation of o/n rate to repo anytime soon. 11

%

10

LAF Corridor, Current vs DB Economics forecasts O/N Mibor: 1m Avg, Current vs Implied from the OIS curve MSF, Current vs DB Economics forecasts

9 8 7 6 5 4 3 Jan-10

Jan-11

Jan-12

Jan-13

Jan-14

Jan-15

Source: Deutsche Bank, Bloomberg Finance LP

Supply for FY15 resumes today. Meanwhile, the government is set to kick-start its FY15 issuance today, targeting INR3.68trn in H1, or 61% of its total FY15 issuance set as per the interim budget. This front-loading of issuance is not new and is in line with the last few years. Also, the gross issuance is relatively higher in April-May, but this is not surprising as we have a large INR751bn of G-secs maturing in the same period. So, while supply is not an immediate concern at this stage, there remains the risk that the interim budget (and by extension, financing needs) is revised higher once the new administration is in place. The interim Budget's target of 4.1% of GDP deficit is fairly optimistic, both by way of revenue assumptions, and a realistic compression in spending by a newly installed government. Again, while the broader drive towards fiscal consolidation will likely continue, the potential for revision in issuance numbers will remain an overhang. Indeed, we note that the fiscal deficit for FY09-10 was penciled in at 5.5% of GDP (INR3.3trn) in the interim budget during last Lok Sabha election, only to be revised higher to 6.8% of GDP (INR4.0trn) in the final budget presented by the same finance minister post elections. Demand remains the key. Banks have been the main buyers of government securities owing to the mandatory SLR requirements, followed by insurance companies and the RBI. Together they own 74% of the market and another 17% is owned by Primary dealers and Provident funds.

Page 68

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Government bond ownership (%)

Relatively higher gross supply in Apr-May, but not net 900

INR bn

Gross

Redemptions

Net

800 700

Corporates, 1.3

Mutual Funds, 1.6

600

Financial Institutions, FIIs, 1.4 0.7

Others, 4.9

EPF, 7.4

500

Banks, 38.2

PDs, 9.3

400 300

RBI, 16.0

200 100

Insurance Companies, 19.3

0 Apr

May

Jun

Jul

Aug

Sep Source: Deutsche Bank, RBI Govtt bond holdings as on Dec-13

Source: Deutsche Bank, RBI

So, clearly banks, local real money and the central bank remains the key source of demand. While the former two are expected to keep their allocations steady given their regulatory requirements, support from the latter is a key consideration, especially as the RBI has reduced OMOs of government bonds this year and has instead shown its preference for term repos to add liquidity to the system. Historically, OMO announcements also served to cheer the sentiment of onshore players. As such, in the absence of this support from RBI, the issuance won’t be any easier to absorb without some cheapening. RBI picked up very little net less issuance this year in OMO buybacks compared to FY12 and FY13 1,400

INR bn

OMO purchases by RBI OMOs as % of Net Issuance (RHS)

35%

1,200

30%

1,000

25%

800

20%

600

15%

400

10%

200

5%

-

0% FY09-10

FY10-11

FY11-12

FY12-13

FY13-14

Source: Deutsche Bank, Bloomberg Finance LP

Meanwhile, demand side of the equation is likely to benefit from the recent RBI regulation to phase out FII limits in T-bills and merge these freed up limits with the normal government dated securities limits. This was done to limit hot money flows and rather encourage longer maturity flows. The net impact from this, however, is likely to be marginal as 1. Foreigners’ overall share as a % of total market is minimal and 2.these carry/arb FII investors are unlikely to chase duration and will stick to the very front end of the government bond curve. Lastly, stability of the rupee is another consideration. As we noted in our recent FX piece ‘INR & IDR: The complexion of elections’, markets have been cheering the prospect of leadership change, anticipating a shift to clean Deutsche Bank AG/London

Page 69

4 April 2014 Global Fixed Income Weekly

reformers that can overcome bureaucratic hurdles and produce better economic results. However, reality could be more complicated than what the market is positioning for. As sentiment coalesces around a market positive outcome, the balance of risks also shifts towards disappointment. Given the overhang of positioning both in equities and FX, and that markets have historically been more sensitive to election effects, the scope for disappointment relative to expectations appears higher. We would thus stay nimble at this stage and await the election outcome to re-assess the situation later.

Page 70

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Pacific Australia New Zealand

Rates Gov. Bonds & Swaps Inflation

Dollar Bloc Strategy „

10Y yields are close to recent highs as we go into the payrolls release. We think we will have to see a payrolls of well above 200k for the sell-off to extend materially. This is not out of the question when we consider Deutsche Bank’s above consensus call for a payrolls gain of 275k.

„

While the 3Y/10Y ACGB slope recently reached its lowest point for the year, before steepening in the past few days as the US long-end has sold-off, the front of the curve has been steepening since early March and is back to where it was around the middle of January.

„

If the market is repricing the medium-term outlook for the RBA but still not expecting the tightening cycle to start for some time then the front of the curve can bear steepen. This is part of what we think has been happening. Another recent steepening influence on the front of the curve is the US sell-off.

„

Our view of the RBA being on hold for an extended period couples with our expectation that the 10Y UST yield will gradually move up to produce the expectation that the front of the AUD curve will steepen rather than flatten. In hindsight we should have recommended the front-end steepener in early March when it was at its flattest point for the year. But we aren’t going to let the fact we didn’t stop us from entering the trade now given our medium-term outlook and the near-term expectation that the US nonfarm payrolls will be in the upper end of the consensus range.

„

We recommend a 3M forward 1Y/3Y steepener at its current level of +54bp. The trade is very modestly positive carry. The key risks are weak US data or the growing expectation that the RBA might tighten sooner rather than later.

„

This trade challenges the recommendation we made earlier this week to pay the 3Y swap spread. On reconsideration, especially when we think about the negative carry of this trade, we exit the paid 3Y swap spread trade.

„

The relative performance of the belly of the AUD curve remains directional with the market. We expect this to continue.

David Plank Macro strategist (+61) 2 8258-1475 [email protected] Ken Crompton Strategist (+61) 2 8258-1361 [email protected]

10Y bond yields at top of recent range as we head into the key employment data As we write 10Y bond yields across the $-bloc are around their highest levels since the second half of January. The growing expectation of a reasonably robust US payrolls report seems to be at least one factor in the recent sell-off.

Deutsche Bank AG/London

Page 71

4 April 2014 Global Fixed Income Weekly

10Y ACGB and 10Y UST 5.0 4.8

10Y NZGB (LHS)

10Y ACGB (LHS)

10Y UST (RHS)

10Y CAN (RHS)

3.2 3.0

4.6

2.8

4.4 2.6

4.2 4.0

2.4

3.8

2.2

3.6

2.0

3.4 1.8

3.2

1.6

3.0 2.8 Jan-13

1.4 Mar-13

May-13

Jul-13

Sep-13

Nov-13

Jan-14

Mar-14

Source: Deutsche Bank, Bloomberg Financial LP

There is some justification for not getting too carried away by the recent move. There has been no obvious trend in bonds since late August/early September. As such we could characterise this recent sell-off as just being a continuation of the choppy range trading that has been in place for some 6 months or so. Alternatively it could be the start of a renewed uptrend. But we think we will have to see a payrolls gain of well above 200k for the move to extend materially. Such a result is not out of the question, of course. Our US economics team has a pick of 275k, for instance. With this being the top of the range in the Bloomberg Financial LP survey something close to this outcome could see the 10Y UST push toward 3% in the near-term. If the market consensus pick of around 200k proves to be closer to the mark then we would expect the recent trading range to hold. Front of AUD curve steepening back to January levels In recent commentary (such as our Fixed Income Monthly published on 1 April) we have noted that by late March the 3Y/10Y ACGB slope (using bonds rather than futures) had flattened to its lowest point in 2014, though it has steepened a few basis points in recent days as the 10Y UST has sold-off. In contrast, the front of the curve has been steepening since early March and is back to levels that match where it was around the middle of January. 1Y/2Y and 2Y/3Y slopes 40

1Y/2Y swap slope 2Y/3Y swap slope

35 30 25 20 15 10 Sep-13

Oct-13

Nov-13

Dec-13

Jan-14

Feb-14

Mar-14

Apr-14

Source: Deutsche Bank, Reuters

Page 72

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

The steepening at the front of the curve has been happening as the market has been pushing up its expectations for the RBA. From early March to now the market has pushed up its 12 month ahead expectation for the RBA by around 15bp. 12M ahead market pricing for the RBA 2.7

Cash rate implied by 12th IB contract

2.6

2.5

2.4

2.3

2.2 Sep-13

Oct-13

Nov-13

Dec-13

Jan-14

Feb-14

Mar-14

Apr-14

Source: Deutsche Bank, Reuters

We might normally expect a front-end sell-off to flatten the curve. The relationship between market direction and the front curve slope is more complicated than this, however. If the market is repricing the medium-term outlook for the RBA but still not expecting the tightening cycle to start for some time then the front of the curve can bear steepen. This is part of what we think is happening at present. Bills/2Y slope vs 10Y UST 60 40

3.0

front curve slope as implied by IR2/3Y ACGB spread, bp (LHS) 10Y UST (RHS)

2.8 2.6

20 2.4 0

2.2 2.0

-20

1.8

-40

1.6 -60 1.4 -80 -100 Jul-12

1.2 1.0 Oct-12

Jan-13

Apr-13

Jul-13

Oct-13

Jan-14

Apr-14

Source: Deutsche Bank, Bloomberg Financial LP

Another recent steepening influence on the front of the curve is the US sell-off. The chart above shows that the correlation between the 10Y UST and the AUD bills/3Y slope has been very tight in recent months. In part we think this is because the overall move in the AUD front-end over the past few months has been relatively small.

Deutsche Bank AG/London

Page 73

4 April 2014 Global Fixed Income Weekly

Our view of the RBA being on hold for an extended period couples with our expectation that the 10Y UST yield will gradually move up to produce the expectation that the front of the AUD curve will steepen rather than flatten over the course of this year. In hindsight we should have recommended the front-end steepener in early March when it was at its low for the year. Having not done so, however, should we jump into it now? The success or otherwise of this trade in the near-term could well depend on the strength of the US payrolls report, with a weak outcome possibly triggering a US rally that flattens the curve. On the other hand, the fact that RBA pricing for the year ahead is close to its recent high provides some protection to the steepening trade in the event of a soft US report (assuming that the AUD front-end rallies in response, in part because the AUD is likely to move higher if we get a weak report). On balance we like the look of the front-end steepener even though we missed entering it at a much better level. This reflects both our medium term view of the market and our near-term expectation that payrolls will be in the upper end of market expectations. We recommend a 3M forward 1Y/3Y steepener at 54bp. The trade is very modestly positive carry. The key risks are weak US data or the growing expectation that the RBA might tighten sooner rather than later. Stopping out of paid 3Y swap spread trade Isn’t a front-end steepener at odds with our recommendation earlier this week to pay the 3Y swap spread? After all, if the front of the curve steepens then it will be difficult for the 3Y swap spread to widen. We recommended the swap spread trade even with this risk because of the size of the gap between the actual 3Y swap spread and our model estimate. Our estimate takes into account the slope of the curve and so was effectively saying that the 3Y swap spread was already pricing in a steeper front-end curve. While this may have been the case, we neglected to consider the impact of the considerable negative carry on the trade. Even if we were right that a steeper front-end curve would not compress the 3Y swap spread further, it would likely make it very difficult for the 3Y swap spread to widen by more than the negative carry. It also seems that the 3Y swap spread can compress – it has fallen a few basis points since we recommended the trade. In light of this, the negative carry and our view on the front-end of the curve we are stopping out of the trade. Belly of the AUD curve continues to perform in line with the outright direction of the market The final aspect of the curve we want to consider in this week’s edition is the 5Y. If we look at the 2Y/5Y/10Y butterfly, it dropped reasonably sharply in late February through to early March and has since rebounded.

Page 74

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

2Y/5Y/10Y curve butterfly vs 10Y swap from Jan-13 25 20

5.00

2Y/5Y/10Y swap butterfly (LHS) 10Y swap (RHS)

4.75 15 10

4.50

5 4.25 0 -5

4.00

-10 3.75 -15 -20 Jan-13

3.50 Mar-13

May-13

Jul-13

Sep-13

Nov-13

Jan-14

Mar-14

Source: Deutsche Bank, Reuters

We think the sharp drop (sharp in the context of the recent range) was mainly a reflection of the market rally, give or take, with the rebound due to the subsequent sell-off. There has been a reasonably close directional relationship between the relative performance of the 5Y and the direction of the market for some time. This is not that unusual. We find a long period of directional correlation prior to the global financial crisis, but then during the GFC and for a period following the butterfly tracked the yield curve rather than the direction of the market. 2Y/5Y/10Y curve butterfly vs 10Y swap between Jan-01 and Dec-07 60

8.0 7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8

2Y/5Y/10Y swap butterfly (LHS) 10Y swap (RHS)

50 40 30 20 10 0 -10 -20 Jan-01

Aug-02

Mar-04

Oct-05

May-07

Source: Deutsche Bank, Reuters

For now we think the relative performance of the 5Y will likely reflect market direction. This is because we see the 5Y part of the curve broadly equating to market thinking about whether or not rates will normalise. That is, if the market is confident rates will eventually normalise then the 5Y part of the curve will underperform the bear market in rates whereas a pushing back on the expectation of rates normalising will see the 5Y outperform a bullish rates market.

Deutsche Bank AG/London

Page 75

4 April 2014 Global Fixed Income Weekly

Dollar Bloc Relative Value „

Around $21b of AUD debt securities mature during June 2014, including the $13.3bn ACGB.

„

We have found in the past that new issuance increases around the time of maturities, whether that be due to borrower refinancing or new borrowers looking to capture some of the excess cash in the system.

„

The pickup in issuance can have a noticeable impact on swap spreads due to swapping of new fixed rate issuance, or use of swap to aid portfolio lengthening to match index extension. The yield curve can also flatten due to bond buying related to the index extension. Increased Kangaroo issuance can also generate basis swap receiving.

„

On average, ACGB maturity months see 3Y swap spreads widen six basis points relative to our regression model after the ACGB maturity date, whilst 10Y swap spreads tighten by 3bp over the period from a month before to a month after the maturity.

„

However, the June maturity will be followed quite quickly by the October ACGB. Both maturity clusters are of similar size. Two maturities so close together – especially of bond lines which were only six year and four year bonds at the time of original issue respectively – mean that the market impacts might be diluted a little.

„

The concession to supply ahead of the upcoming 8 April ACGB linker auction has been larger than normal, reflecting some positioning flows. With breakevens weaker despite much higher yields over the past month we think that the auction offers value.

A medium-sized heavy maturity month in June The $13.3b June 2014 Australian Commonwealth Government Bond line matures on 15 June 2014. Large ACGB maturities can be important because debt issuance tends to aggregate around ACGB maturities, using the “benchmark effect” of the ACGB to aid price discovery and hedging. AUD denominated debt maturity – including Gov’t – monthly totals (A$b) 35 30

Floating

Jun-14

Fixed

25 20 15 10 5 0 Jan-05

Jan-08

Jan-11

Jan-14

Source: Bloomberg Financial LP, Deutsche Bank

The heaviest maturity periods tend to cluster around ACGBs which had relatively long durations at issue, such as the bonds of roughly twelve year maturity that are issued to be built up into liquid benchmarks for the 10Y bond futures contract. The June 2014 ACGB, however, was issued in July 2008 and

Page 76

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

as such the total maturities during the month (around $21.3bn face value) are, whilst well above average, not “extremely” large. Semi, ACGB and SSA maturities, May-July 2014 Issuer

Coupon

Maturity

Outstanding

AUSGOV

6.250

15-Jun-14

$13.299b

NSWTC

2.750

8-Jul-14

$1.589b

SAFA

5.250

6-Jun-14

$1.778b

TASCOR

5.500

23-Jun-14

$0.962b

WATC

5.500

23-Apr-14

$1.926b

ADB

5.250

13-May-14

$1.025b

EIB

5.375

20-May-14

$2.750b

IADB

5.375

27-May-14

$1.750b

IFC

5.750

24-Jun-14

$1.800b

KFW

5.500

5-Jun-14

$1.250b

RENTENBANK

6.000

15-Jul-14

Total

$1.200b $29.329b

Source: Deutsche Bank, Reuters

Large maturity months are interesting because historically we have seen that “maturity begets issuance” – ie large maturities tend to see clusters of higher than average bond issuance. The combination of the very large volume of cash being returned to investors, as well as the possibility of refinancing, or new issuance to take advantage of investor cash, means that there are often perceptible market impacts during the months of heavy issuance. Historically one of the largest impacts has come from the extension of the duration of bond indices as the maturing issues are removed. Index tracking funds will be compelled to match this extension, and if this is achieved by buying longer maturity bonds then there is potential for the long end swap spreads to widen. Alternatively, if the duration is extended by buying shorter maturity bonds and receiving swaps then swap spreads are more likely to tighten. Fixed rate issuance is generally associated with interest rate swap receiving, especially from financials (who prefer to match their mostly floating rate liabilities) and Kangaroo borrowers, who receive swap as an intermediate step to cross-currency swapping the exposure back to their preferred benchmark currency. Thus if Kangaroo borrowers step into the fray around the June maturities, both basis swap receiving and swap receiving are likely. Our swap spread model – 3Ys tend to widen, 10Ys tend to tighten In quantifying the typical change in swap spreads during a maturity event, rather than directly observing changes in swap spreads, we instead used the residual values calculated from our swap spread model. Doing this helps to remove effects like changes in the outright level of rates, changes in the slope of the yield curve or moves in repo spreads. Because our model has no variable to act as a proxy for bond issuance, any impact from issuance should be revealed in the residuals. Higher residuals indicate that spreads are wide relative to the model, whilst lower residuals indicate that spreads are tightening relative to the model.

Deutsche Bank AG/London

Page 77

4 April 2014 Global Fixed Income Weekly

3Y swap spread residuals in “heavy” and “low” maturity months – business days before/after 8.0

3Y swap spread model residual

6.0 4.0 2.0 0.0 -2.0 -4.0

Low maturity month average Heavy maturity month average -20-18-16-14-12-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20

Source: Bloomberg Financial LP, Deutsche Bank

As the graph above shows, 3Y spreads tend to widen shortly before the middle of a big maturity month (ie when the ACGB usually matures, although the 21stof-the-month ACGB maturities will become more common in the future) and then continue widening afterward. The likely drivers of 3Y swap spread widening during big maturities are unclear. 10Y swap spread residuals in “heavy” and “low” maturity months – business days from mid-month 2.5

10Y swap spread residual

2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0

Low maturity month average Heavy maturity month average -20-18-16-14-12-10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20

Source: Bloomberg Financial LP, Deutsche Bank

At the long end of the curve, 10Y swap spreads tend to continually tighten through a large maturity month, although the magnitude is lower than the 3Y widening This indicates that, at the 10Y point, the combined use of swap for duration extension and hedging by borrowers is more significant than the effect of outright bond buying.

Page 78

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

AUD denominated issuance and maturities – including Gov’t – monthly totals (A$b) 35

Maturity

30

Jun-14

Issuance

25 20 15 10 5 0 Jan-05

Jan-08

Jan-11

Jan-14

Source: Bloomberg Financial LP, Deutsche Bank

But issuance might be spread out this time The chart above shows the relationship between maturity and issuance over time. It can be seen that the relationship is not especially robust, and that there has been significant shift in the base level of issuance since 2009. This is mostly due to increased Government issuance. During 2009-10 the AOFM issued several “infill” bonds at the front of the yield curve. This has resulted in the average maturity gap between bonds in the ACGB curve dropping from almost a year to a little over half a year at the front end. The next ACGB maturity is 21 October 2014, and total AUD maturities in that month are expected to be around $21bn. Like the June 2014 ACGB, the October 2014 bond had a relatively short duration at issue (about four years) and thus has not had as much time to attract similar maturity issuance. The swap spread impact of issuance around medium-sized maturity months (such as June and October 2014) is less than large maturity months (such as May 2013, or April 2015). This, combined with two maturities being close together may see the usual issuance cluster being spread a little thinner than has been the case in the past. We think this means that any 3Y spread widening or 10Y tightening tendency might be a little weaker this time around. 3Y spreads – currently appear quite tight 160

3Y swap spread model 3Y swap spread

140 120 100 80 60 40 20 0 -20 Jan-07

Jan-08

Jan-09

Jan-10

Jan-11

Jan-12

Jan-13

Jan-14

Source: Bloomberg Financial LP, Deutsche Bank

Deutsche Bank AG/London

Page 79

4 April 2014 Global Fixed Income Weekly

As the graph above shows, 3Y EFPs are trading very tight relative to our model at present. 10Y swap spreads are also trading a little tight to our model but to a much lesser extent than the 3Y. Given the tendency for tightening pressure on 10Y EFPs through heavy maturity periods we would be hesitant to enter a paid swap spread trade unless spreads were even further rich, or we had other reasons to expect a short-term widening in spreads.

8 April linker auction offers value Breakevens weaken despite large rally ACGB linkers have underperformed the bond selloff during the week of 31 March. Breakevens have declined across most of the curve despite 10Y bond yields selling off more than ten basis points. ACGB linker real yield and BE curves 30

Real yield

Breakeven

25

3.0 2.5

20 2.0

15

1.5

10 5

1.0

0 -5 -10

0.5

Δ1w Δ1m 4-Apr level (RHS)

0.0

Source: Bloomberg Financial LP, Deutsche Bank

The Sep-30 breakeven has dipped below 2.60% for the first time since November as the breakeven curve has flattened slightly. ACGB linker breakevens 3.2

ACGB Aug-20 ACGB Sep-25 ACGB Aug-35

ACGB Feb-22 ACGB Sep-30

3.0 2.8 2.6 2.4 2.2 2.0 Jan-13 Mar-13 May-13 Jul-13 Sep-13 Nov-13 Jan-14 Mar-14 Source: Bloomberg Financial LP, Deutsche Bank

The cheapening of breakevens is due in some part, in our view, to a combination of positioning flows and a concession to supply ahead of the 8 April ACGB linker auction. Our traders have noted greater selling interest through the belly of the curve (2020-2025) whilst support for the longer dated Page 80

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

lines (2030 and 2035) has been a little stronger. The slight underperformance of the belly of the curve over the past month supports this. ACGB linker modified duration vs z-spread 30 20

Z-spread

10 0 -10 ACGB linkers

-20

ACGB nominal -30 0

5

10

15

20

Modifed Duration Source: Bloomberg Financial LP, Deutsche Bank

Weaker breakevens, higher real yields, and positive carry – 8 April auction offers value Carry on ACGB linkers remains positive across the curve, with the impact of the above-expectations Q1 2014 print continuing to accrete into linker notionals. ACGB linker real yield carry (bp) 1M carry

35.0

2M carry

30.0

3M carry

25.0 20.0 15.0 10.0 5.0 0.0 Jan-15

Jan-18

Jan-21

Jan-24

Jan-27

Jan-30

Jan-33

Source: Bloomberg Financial LP, Deutsche Bank

We think that the cheapness that has developed ahead of the upcoming auction is overdone. We think that with positive carry, narrower breakevens and real yields on the 2030 line back above 2.00% both domestic and real yield buyers will find ACGB linkers attractive. The line to be offered will be announced at midday on 4 April. We think buying at the auction to capture the recent excess cheapening will offer value especially if 2030 or 2035 linkers are offered (as an aside, we note that the Apr33 nominal ACGB auction recently generated strong results).

Deutsche Bank AG/London

Page 81

4 April 2014 Global Fixed Income Weekly

Global

Economics Rates Gov. Bonds & Swaps Inflation

Markus Heider Strategist (+44) 20 754-52167 [email protected]

Global Inflation Update Global B/E performance this week was mixed, with valuations rebounding from low levels in EUR, remaining broadly unchanged in USD and declining mostly in GBP (chart 1). Real yields rose strongly in GBP and USD. The macro backdrop remains mixed for B/Es, with indicators of global inflation remaining low, and domestic inflation expected to recover only gradually. Some help may come from somewhat better global data in Q2.

1. EUR B/Es outperform 10

1w change in BEI on 3-Apr, carry adj., bp

8

EUR

FRF

6

GBP

USD

4 2

Central banks: Policy divergence may be increasing, with the focus in the US and UK mainly on the timing of the first rate increase and the speed of subsequent policy normalisation, whereas in EUR the focus is on the possibility of additional stimulus. This could be a relative support for EUR B/Es especially if it means a somewhat weaker currency. Economic data: Recent PMIs have shown a leveling-off in new orders (chart 2), and the loss of data momentum tends to be a negative for B/Es (chart 4). An expected pick-up in activity in the US and some reduction in geo-political risk could support data into Q2, especially if China can stabilize. This should be a positive for B/Es, especially given that valuations look somewhat low against the level of PMIs (chart 3).

0 -2 -4 5y

10y

20y

30y

Source: Deutsche Bank

2. New orders level-off, prices down 65

80

Global PMIs: principal component

75

60

70

55

65

50

60 55

45

Commodities/inflation: Ultimately, significantly higher B/Es will likely require a more visible pick-up in spot inflation. While headline y/y inflation is expected to be bottoming, this year’s recovery is expected to be only gradual. Commodity prices may be stabilizing, but are unlikely to push up CPI inflation significantly in the short-term (although we expect higher food inflation from H2). Global manufactured goods inflation remains subdued, and PMI input price balances have been turning lower again (chart 2), which is a downside risk for CPI projections and B/Es. Domestic inflation in the UK (RPI) and US is supported by a recovery in housing markets, but wage growth may rise only progressively. Against this backdrop, and given relatively low valuations, we would see scope for higher B/Es, although the near-term upside may be limited.

50

40

45

35

new orders

30

input prices (rhs)

25 Jan-07

40 35 30

Jan-08

Jan-09

Jan-10

Jan-11

Jan-12

Jan-13

Source: Deutsche Bank

4. But recent momentum has been negative

3. 10y TIPS B/Es low v level of economic indicators 3.0

2.5

US, EUR, JP & China business survey, 3m chge (lhs)

2.0

2.0

5y TIPS implied ZC BEI, 3m chge

1.5

1.5

2.5

1.0

1.0

0.5

0.5

2.0

0.0

0.0 1.5

fitted based on US, EUR, JP & China busines surveys, core inflation and VIX 10y TIPS implied ZC BEI

-0.5

-0.5

-1.0

1.0

-1.0

-1.5

-1.5

-2.0 0.5 Jan-02

Jul-03

Source: Deutsche Bank

Page 82

Jan-05

Jul-06

Jan-08

Jul-09

Jan-11

Jul-12

Jan-14

Jan-14

-2.5 Jan-03

-2.0 Jul-04

Jan-06

Jul-07

Jan-09

Jul-10

Jan-12

Jul-13

Source: Deutsche Bank

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

EUR

1. EUR B/Es rebound 12

1w change on 3-Apr ,carry adj., bp BEI Real Yld

10

RV: 5y cash B/Es, and in particular the OBLei18 have richened against swaps, and for German ILBs swap-bond B/E spreads are now tighter in 5y than in 10y which is unusual (chart 4). At current valuations we would have a preference for swaps over the OBLei18.

6 4 2 0 -2 -4

OATei40

OATei32

OATei27

OATei24

OATei22

OATei20

OATei18

OATei15

DBRei23

OBLei18

-6

DBRei16

Rationale: The market and macro backdrop could be slowly improving for EUR B/Es. The March HICP flash estimate was slightly lower than we were initially forecasting (see ‘EUR inflation watch’), but headline y/y inflation is expected to rebound to 0.7/0.8% in April, and a stabilisation in inflation should be a positive for B/Es, and so could be the improving carry into May (close to 30bp for 5y ILBs until the start of June). B/Es have rebounded since the HICP print on Monday (chart 1). The fall in crude oil prices this week remains a downside risk, but some widening in margins may cushion the impact on April HICP. With metal prices rising this week, and agricultural prices stabilizing at one-year highs, commodity price trends on average would appear supportive for B/Es (chart 3). Business surveys have leveled-off in March, but on average in Q1 remain consistent with another slight rise in GDP growth, while February retail sales point to strengthening consumption growth. A renewed pick-up in US demand and some reduction in geo-political risk may support business surveys in Q2. China and a renewed weakening in upstream price indicators (chart 2) are downside risks. Against that, the ECB appears to be stepping up (verbal) efforts to lean against a strengthening currency and falling inflation expectations, and B/Es have rallied following the ECB press conference. Meanwhile valuations remain low when measured against baseline HICP forecasts; we have a positive bias on B/Es, in particular 5y swaps.

Nom Yld

8

DBRei20

View. With valuations low, commodities less of a drag, global data momentum expected to improve, carry positive, spot inflation likely to rebound in April and the ECB stepping up (verbal) support, we have a positive bias on B/Es. We see fair value for the new DBRei30 at a real yield 42/45bp over the DBRei23.

Source: Deutsche Bank

2. PMI prices down, but B/Es cheap 2.5

US, EA, CH PMI: input prices

2.80

2.0

10y OATei implied ZC BEI

2.60

1.5 1.0

2.40

0.5

2.20

0.0

2.00

-0.5 -1.0

1.80

-1.5

1.60

-2.0 1.40

-2.5 -3.0 Jan-03

1.20 Jul-04

Jan-06

Jul-07

Jan-09

Jul-10

Jan-12

Jul-13

Source: Deutsche Bank

4. OBLei18 rich v swaps

3. Commodity trends more supportive

40

2.5 DEMBE5Y

2.3

fitted, f(oil, agriculture, metals, spot HICP)

30

2.1 20

1.9 1.7

10

1.5 1.3

0

1.1

OBLei18, swap richness -10

0.9 0.7 Jan-10 Jul-10

Jan-11 Jul-11

Source: Deutsche Bank

Deutsche Bank AG/London

Jan-12 Jul-12

Jan-13 Jul-13

Jan-14

-20 Apr-12

DBRei23, swap richness

Aug-12

Dec-12

Apr-13

Aug-13

Dec-13

Apr-14

Source: Deutsche Bank

Page 83

4 April 2014 Global Fixed Income Weekly

DBRei30: Germany will launch a new DBRei 0.5% 15-April 2030 via auction next week (EUR2bn announced). This will be the longest German ILB to date, and in duration terms will be longer than the OATei32. Demand is expected to be strong, as for example suggested by the experience of the DBRei23, which for the first few months after issuance had mostly traded richer than swaps (chart 4). Estimating fair-value for the DBRei30 is not straightforward, not only because it will represent a significant extension of the real yield curve, but also because the 15y-20y sector of the nominal Bund curve is populated with relatively seasoned issues and looks cheap (chart 5). This complicates pricing off the nominal curve (via B/Es or various relative ASW metrics). For example, looking at the z-spread to the nominal bond curve (and taking into account a floor value of 3bp), a real yield spread of about 50bp to the DBRei23 would be required to bring the DBRei30 in line with the OATei32. At DBRei23 +50bp, the DBRei would however look too cheap on the forward real yield curve. A real spread to the DBRei23 of just over 40bp would bring it line with swaps or OATei on the latter. We would estimate fair-value to be at a real yield 42-45bp over the DBRei23 and charts 6-8 show the DBRei30 on the ‘swap-richness’, forward B/E and forward real yield curves assuming DBRei23 +42/45bp. 6. Z-spread v nominal bond curves

5. 15y-20y Bunds relatively cheap 15

90

10

80 70

5

60 0

50

-5

40 DBR z-spreads

-15

DEM FRF DBRei30

10

5

20

OAT z-spread (rhs)

z-spread v nominal bond curve

15

30

-10

duration

10

-20

0

0

-25 -30 2013

20

-10 -20 2019

2024

2030

2035

2040

-5

2046

-10 0 Source: Deutsche Bank

10

15

20

25

Source: Deutsche Bank

7. Forward B/Es 3.2

5

8. Forward real yields 1.1

Fwd BEI, sa

2.2

Fwd real yields, sa

2.7

1.7 0.6

2.2 1.7

DEM (fwd between subsequent issues)

1.2 0.7

0.1

FRF (fwd between subsequent issues)

1.2

1y HICP

0.2 DEM (fwd between subsequent issues)

-0.4

1y real swap rates

0.7

-0.3

FRF (fwd between subsequent issues; rhs) 0.2 2013

Source: Deutsche Bank

Page 84

2019

2024

2030

2035

2041

-0.9 2013

-0.8 2019

2024

2030

2035

2041

Source: Deutsche Bank

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

GBP View. Recent survey evidence about firms’ selling price intentions has been mixed to weaker, despite still strong activity indicators. With risks to near-term RPI prints on the downside in our view we stay neutral on B/Es. Real yields are highest (just positive) 5y to 10y forward; they are the richest in 20y and 40y.

1. B/Es decline 10

BEI

Real Yld

Nom Yld

8 1w change on 3-Apr, carry adj, bp

6 4 2 0 -2

UKTi68

UKTi62

UKTi55

UKTi52

UKTi50

UKTi47

UKTi44

UKTi42

UKTi40

UKTi37

UKTi34

UKTi32

UKTi29

UKTi27

UKTi24

UKTi22

UKTi19

-4

UKTi17

Rationale: Inflation news this week has been mixed to weaker. Commodities diverged (crude down, agriculture sideways, metals up), the exchange rate was broadly stable, headline PMIs fell, while remaining at levels which are consistent with above consensus growth and rising prices, but PMI price balances fell. Indicators of firms’ price setting intentions have recently not given a consistent message; PMI output prices in March declined more than one point in manufacturing and close to 1.5 points in services, while the EC survey’s ‘selling price expectations’ balances increased in February and March, both in services and retailing (chart 2). While evidence remains inconclusive, a renewed downturn in indicators of upstream price trends would be a challenge for our view of higher inflation next year and a downside risk for B/Es, which have mostly declined this week (chart 1). 5y RPI is down 3bp and back closer to the middle of the 3% - 3.1% range seen over the past couple of months. We continue see the risk of RPI inflation falling to 2.4% in March. In this context, we maintain our neutral view on B/Es.

Source: Deutsche Bank

2. Survey price trends mixed 4.0

PMI services, output prices PMI manufacturing, output prices

3.0

z-scores

EC selling price expectations, retailing EC selling price expectations, services

2.0 1.0 0.0

RV: Real yields are off the cyclical lows seen in spring last year, but have recently stabilized at still relatively low levels; the highest point on the bond curve is marginally above zero, while real swap rates remain negative across the curve. Looking at 5y swap rates for different forward horizons shows positive yields 5y to 10y forward, while rates remain very low at the ultra longend (chart 4). Chart 4 puts these valuations in the context of trends seen over the past three years. 5y5y real rates not only are positive, but also cheaper than average over the past three years. From 10y forward 5y rates mostly remain at below average levels, despite the general sell-off in spot yields seen over the past year. 20y5y and 40y10y in particular seem expensive (chart 4).

-1.0 -2.0 -3.0 -4.0 2007

2008

2009

2010

2011

2012

2013

2014

Source: Deutsche Bank

4. Real rate RV: 5y5y cheap, 20y5y rich

3. 5y real rates 5y forward positive, ultra longs rich 0.4%

1.5

0.2%

1.0

1.0

0.5

0.5

0.0

0.0

-0.5

-0.5

0.0%

1.5

GBP real swap rates: 3y RV

-0.2% -0.4% -0.6% -0.8%

-1.0

-1.0

10y real swap, x years fwd

-1.5

-1.5

5y real swap, x years fwd

-2.0

spot

75% quart

max

min

25% quart

-1.0%

-3.0

-3.0

-1.2% 0

1

2

3

4

Source: Deutsche Bank

Deutsche Bank AG/London

5

7

10

15

20

25

30

40

-2.0 -2.5

-2.5

5y

5y5y

10y5y

15y5y

20y5y

25y5y

30y10y 40y10y

Source: Deutsche Bank

Page 85

4 April 2014 Global Fixed Income Weekly

United States

Rates Gov. Bonds & Swaps Inflation Rates Volatility

Alex Li Research Analyst (+1) 212 250-5483 [email protected]

Inflation-Linked „

We like short dated TIPS out to the five-year. Front end TIPS look cheap on our CPI forecast. The five-year TIPS have cheapened in breakevens, despite 5s underperforming on the Treasury curve.

„

There will be a new five-year TIPS auction on April 17. We expect an $18 billion auction size to be announced on April 10.

„

The Fed will conduct its TIPS purchase operation on April 8 for $0.75 billion to $1 billion. The focus is likely again to be in the long end.

Against the conventional wisdom We like short dated TIPS out to the five-year. The five-year TIPS have cheapened in breakevens, despite 5s underperforming on the Treasury curve. There will be a five-year TIPS auction on April 17. We expect an $18 billion auction size to be announced on April 10. Conventional wisdom suggests TIPS breakevens usually widen on higher Treasury yields. The cheapening of 5s on the Treasury curve, such as in 2s-5s10s, has been well advertised. Even after adjusting for the 2s-5s-10s spread and the five-year Treasury yield level, five-year TIPS breakevens appear low. The seasonally positive TIPS carry also supports holding front end issues. In our CPI forecast, the March NSA index is projected to rise to 235.95, a 0.5% increase from the February’s level, partly thanks to the recent rise in gasoline prices. If this forecast is correct, it will be the largest monthly increase since February 2013. The five-year TIPS will have a carry about +18bp from now through the end of May, the ten-year TIPS about +9bp, and the 30-year TIPS about +4bp in this scenario. 5yr TIPS breakevens versus 2s/5s/10s Treasury spread 2.05

Last 6 months' data

2.00

4/4/2014

y = 0.1974x + 1.8655 R² = 0.2639

5yr BE

1.95 1.90 1.85 1.80 1.75 1.70 (0.40)

(0.20)

0.20 2s-5s-10s Treasury spread

0.40

Source: Bloomberg Finance LP and Deutsche Bank

Page 86

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

Front end TIPS look cheap on our CPI forecast. For example, 2015 TIPS are about 8-10 ticks cheap, and have highly positive carry near term. The 0.5% of 4/2015 TIPS, e.g., have a projected carry around 60bp through the end of May. 5yr TIPS breakevens versus 5yr Treasury yield 2.05

y = 0.1404x + 1.6419 R² = 0.0879

Last 6 months' data

2.00

4/4/2014

5yr BE

1.95 1.90 1.85 1.80 1.75 1.70 1.20

1.30

1.40

1.50 1.60 5yr yield

1.70

1.80

1.90

Source: Bloomberg Finance LP and Deutsche Bank

Projected TIPS carry through the end of May 2014 90 80 TIPS carry from 4/8/14 to 5/31/14

70 60 50 40 30 20 10

TII 0.75% 02/42

TII 0.625% 02/43

TII 2.125% 02/41

TII 2.125% 02/40

TII 3.375% 04/32

TII 2.5% 01/29

TII 3.875% 04/29

TII 1.75% 01/28

TII 3.625% 04/28

TII 2% 01/26

TII 2.375% 01/27

TII 2.375% 01/25

TII 0.625% 01/24

TII 0.375% 07/23

TII 0.125% 01/23

TII 0.125% 07/22

TII 0.125% 01/22

TII 0.625% 07/21

TII 1.25% 07/20

TII 1.125% 01/21

TII 1.375% 01/20

TII 1.875% 07/19

TII 2.125% 01/19

TII 1.375% 07/18

TII 0.125% 04/18

TII 1.625% 01/18

TII 2.625% 07/17

TII 0.125% 04/17

TII 2.5% 07/16

TII 2.375% 01/17

TII 2% 01/16

TII 0.125% 04/16

TII 0.5% 04/15

TII 1.875% 07/15

TII 1.625% 01/15

0

Source: Deutsche Bank

Front end TIPS look cheap on DB CPI forecast BE Inflation

Implied CPI

DB forecast CPI

Rich/ Cheap

TII 1.625% 1/15/2015

TIPS

1.71%

237.14

237.79

Chea p: 9 ti cks (35bp)

TII 0.500% 4/15/2015

1.61%

237.89

238.64

Chea p: 10 ti cks (30bp)

TII 1.875% 7/15/2015

2.06%

240.23

240.80

Chea p: 8 ti cks (18bp)

TII 2.000% 1/15/2016

1.74%

241.36

242.43

Chea p: 15 ti cks (25bp)

Source: Deutsche Bank

The Fed will conduct its TIPS purchase operation on April 8 for $0.75 billion to $1 billion. The focus is likely again to be in the long end. In the March purchase operation, the Fed bought $492 million 3.625s of 4/2028 and $465 million 1.375s of 2/2044 TIPS.

Deutsche Bank AG/London

Page 87

4 April 2014 Global Fixed Income Weekly

Contacts Name

Title

Telephone

Email

Head of European Rates Research

44 20 7545 4017

[email protected]

EUROPE Francis Yared Alexander Düring

Euroland & Japan RV

44 207 545 5568

[email protected]

Global Inflation Strategy

44 20 754 52167

[email protected]

Covered Bonds/SSA

41 44 227 3710

[email protected]

Global RV & Rates Vol

33 1 44 95 64 08

[email protected]

Euroland Strategy/ EUR Govt. bonds

44 20 754 74458

[email protected]

UK Strategy & Money Markets

44 20 7547 3091

[email protected]

Nordic & Swiss Strategy

44 20 7545 2424

[email protected]

Nick Burns

Credit Strategy

44 20 7547 1970

[email protected]

Stephen Stakhiv

Credit Strategy

44 20 7545 2063

[email protected]

Sebastian Barker

Markus Heider Bernd Volk Jerome Saragoussi Abhishek Singhania Soniya Sadeesh Christian Wietoska

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]

Head of APAC Rates Research

61 2 8258 1475

[email protected]

Japan Strategy

81 3 5156 6622

[email protected]

ASIA PACIFIC David Plank Makoto Yamashita Kenneth Crompton

$ bloc RV

61 2 8258 1361

[email protected]

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]

Head of European FX and cross markets strategy

44 20 754 79118

[email protected]

Sameer Goel

CROSS-MARKETS George Saravelos Source: Deutsche Bank

Page 88

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

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

Deutsche Bank AG/London

Page 89

4 April 2014 Global Fixed Income Weekly

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.

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 Page 90

Deutsche Bank AG/London

4 April 2014 Global Fixed Income Weekly

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 fixed income 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.

Deutsche Bank AG/London

Page 91

David Folkerts-Landau Group Chief Economist Member of the Group Executive Committee Guy Ashton Global Chief Operating Officer Research Michael Spencer Regional Head Asia Pacific Research

Marcel Cassard Global Head FICC Research & Global Macro Economics

Ralf Hoffmann Regional Head Deutsche Bank Research, Germany

Richard Smith and Steve Pollard Co-Global Heads Equity Research

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 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

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

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 the investors' own circumstances including their tax position, their regulatory environment and the nature of their other assets and liabilities and as such investors should take expert legal and financial advice before entering into any transaction similar to or inspired by the contents of this publication. Trading in options involves risk and is not suitable for all investors. Prior to buying or selling an option investors must review the "Characteristics and Risks of Standardized Options," at http://www.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