2013 11 21 Fixed Income Insights

Published on February 2017 | Categories: Documents | Downloads: 23 | Comments: 0 | Views: 99
of 8
Download PDF   Embed   Report

Comments

Content

Q3 GDP: The final nail in the Riksbank’s coffin?
We have revised our forecast for Swedish Q3 GDP growth, due on Nov 29, down to 0.2%, 0.1% y/y following the surprisingly large decline in manufacturing inventories announced last week. Most monthly growth indicators suggest GDP at or below our estimate, which is also lower than that of the Riksbank (0.5%, 0.4% y/y). The main upside risk is that stagnating GDP so far this year contradicts signs of a recovering labour market and improving sentiment indicators. Also, the October survey suggests that fixed investment in H1 2013 may have declined less than earlier perceived, indicating that GDP for H1 2013 may be revised slightly upwards. With markets having moved rapidly in our direction over the past week to currently discount a 68% probability of a December rate cut, Q3 GDP may well end up being the final nail in the Riksbank’s coffin, paving the way for it to lower rates next month. Page 5

THURSDAY 21 NOVEMBER 2013

EDITOR Jussi Hiljanen +46 8 50623167 CONTRIBUTORS Olle Holmgren +46 8 7638079 Lauri Hälikkä +46 8 506 23044 Anders Söderberg +46 8 50623021

Short rates to remain depressed, yield curves to steepen
In our upcoming Nordic Outlook to be published on Nov 26, we discuss probable ECB measures to fight growing deflationary pressure within the Euro-zone. With the ECB increasingly flirting with the possibility of negative rates and the Riksbank expected to cut the repo rate in December, we expect the short end of the Swedish curve to remain under downward pressure. We reiterate our 0.75-0.80% target for SGB1049 (2015). Given upcoming Fed tapering and our views on US long yields, we prefer to position for a steeper curve also in Sweden. Pages 4 and 6

Establish a conditional 3m4y vs. 3m10y steepener in the SEK IRS curve
We regard 4y vs. 10y steepeners as attractive. Central bank policies in the Euro-zone and Sweden favour lower short rates while Fed tapering poses an upside risk for long rates. Due to uncertainties concerning near term developments in the long-end of the curve, we prefer to express our steepener view conditionally by selling an ATM payer in 3m4y and buying an ATM payer in 3m10y in SEK IRS. At maturity, the position is not exposed to a risk of a bull flattening except a 3bps premium. Page 7

Inflation-linked bonds: Upcoming coupon payments and switch auctions
Swedish I-L bonds will pay coupons totalling SEK 4.9bn on Dec 1. Switch auctions from SGB3105 (2015) to SGB3107 (2017) will take place on Dec 5-6 and increase OMRX Real index duration by 0.07y effective on Dec 11. For a market update, click here.

You can also find our research materials at our website: www.mb.seb.se. This report is produced by Skandinaviska Enskilda Banken AB (publ) for institutional investors only. Information and opinions contained within this document are given in good faith and are based on sources believed to be reliable, we do not represent that they are accurate or complete. No liability is accepted for any direct or consequential loss resulting from reliance on this document. Changes may be made to opinions or information contained herein without notice.

Fixed Income insights

ECB Current repo Next rate ann. Action expected 0.25% Dec 5 Unch

Fed 0.00-0.25% Dec 18 Unch

Bank of England 0.50% Dec 5 Unch

Riksbank 1.00% Dec 17 -25bps Duration vs. BM Return YTD SEB portfolio Index

Norges Bank 1.50% Dec 5 Unch 0.25y short -1.70% -1.79% Levels & P/L 30 cents Dynamic 46 cents +16 cents 0ps 10bps Dynamic +1bps Status Hold Hold

Strategic duration
Swedish government bond portfolio (Oct 3). We track the OMRX T-bond index and may deviate +/-1 year from BM duration. The recommendation is strategic, being intended mainly to capture quarterly trends in bond yields. The strategic duration recommendation may differ from shorter term views. We changed the duration from 0.5y short to neutral on Feb 7, 2013 and to 0.25y short on Oct 3, 2013.

Trades
Directional Buy a 2.40/2.65/3.00% seagull downside in 3m10y SEK swaps (Fixed Income Focus , Upfront premium Oct 29). Expiry date on Jan 29, 2014. With 10y SEK swap at technically important levels, a Stop break below 2.54% could send 10y swap rate sharply lower. We keep the position open as Now a hedge to our short bond portfolio. P/L Buy 1.10/1.30% risk reversal downside in Dec13 SEK FRA (Jun 13). The position was Upfront premium established based on a possible 25bps repo rate cut before December 2013. We now Target expect the Riksbank to lower the repo rate at its meeting next month. This could result in a Stop small profit, depending on market pricing on the Dec13 fixing day as the rate P/L announcement occurs the following day. Relative value Sell 3m4y ATM payer (1.77%) and buy 3m10y ATM payer (2.65%) in SEK swaps (Nov Upfront premium 21). Expiry date on February 21. We believe that the Riksbank and the ECB keep short rates Stop under downward pressure while Fed tapering poses a risk of higher long rates. Now P/L Buy C1583 (06/20) vs. swaps (Quantitative Insights, Nov 4). At trade entry, forward Established pricing (C1581/C1583) implied that a 1.5y bond on Stadshypotek’s CB curve was priced to Target trade at 146bps vs. swaps in 5.1y and a 2.2y bond was priced to trade 123bps vs. swaps in Stop 4.4y (C1580/C1583). Such ASW spread levels did not exist even at the height of the Now financial crisis. 3m spread carry is 2.5bps. P/L Buy Sp183 (09/2015) vs. Sp182 (03/2015) and pay fixed in a Sp183/Sp182 fwd SEK Established IRS (Oct 31). Current pricing on the Swedbank Mortgage curve implies that a 6m bond is Initial target priced to trade at 28-29bps in 1.4y, which we think unlikely. We therefore recommend Stop buying Sp183 (09/15) vs. Sp182 (03/15) nominal vs. nominal and paying fixed in a Now Sp183/Sp182 forward SEK IRS with an initial target of 15bps. Carry over 3m is 1.4bps. P/L Buy SGB3102 (2020) vs. OATei 2020 in PVBP risk neutral terms (Oct 10). Our Established conditional entry at 54bps was triggered on Oct 10. The spread reached our target and was Target closed on Nov 19 with a profit of 12bps, including a carry of -2bps. Stop Closed P/L Buy Jun15 vs. Sep14 SEK FRA (Sep 26). Historical precedents suggest that once the rate Established hike cycle begins the 1st vs. 4th FRA curve will trade substantially steeper than current Target pricing implies. The position was established at a time when we expected the Riksbank to Stop initiate a hiking cycle in December 2014. Currently, we would prefer expressing the Now steepening view further out on the curve, for example in Dec14 vs. Dec15 SEK FRA. P/L Receive fixed in 10y in 10y SEK vs. EUR swap forward (Sep 19). With the new discount Established for lifers taking effect as of year-end, we expect increased receiving interest from lifers in New target swaps longer than 10y. We change the target to -15bps. Stop Now P/L Pay fixed in the 5y SEK swap, receive fixed in a bfly of 2y and 10y SEK swaps (Aug Established 29). The trade has moved against us since the Fed refrained from meeting market Target expectations of tapering announcement in September. Recent increased market Stop expectations of a Riksbank cut in December has further triggered outperformance of 5s vs. Closed wings. We close the trade with a loss of 10.7bps including carry. Butterfly (bfly) pick-up P/L incl. carry calculated as 5y swap -0.5*2y swap -0.5*10y swap. Buy SGB3108 (2022) vs. SGB1054 (2022) in PVBP neutral terms (Aug 20). As a result of Established another downside surprise in Swedish CPI, the BEI gapped through our stop of 145bps. We Target regard the BEI as too low, expecting it to rise in coming months. Over the last two years, Stop SGB3108 BEI has repeatedly bottomed at the 130-135bps area. We therefore keep the Now position open with a new stop at 130bps. From Aug 20, the position has a carry of -2.3bps P/L to Dec 1 and -6.2bps to Jan 1.

Hold

Levels & P/L 25cents Dynamic -25 cents 0 cents 56.5bps 40bps 50bps 46bps +11bps 28bps 15bps 28bps 18bps +10bps 54bps 40bps 62bps 40bps +12bps 39bps 55bps 30bps 33bps -6bps 13bps -15bps 5bps -9bps +22bps 10bps 25bps 0bps 0bps -10.7bps 155bps 165bps 130bps 140bps -15bps

Status New

Hold

Hold

Closed

Hold

Change

Closed

Change

2

Fixed Income insights

Buy Sp187 (2018) vs. swaps (Aug 8). With a rate cut from the Riksbank in December now our main scenario and with issuers nearly fully funded for this year, we believe demand will remain supportive for Swedish 5y covered bonds going forward. As of Aug 8, the position had a spread carry of 2.4bps.

Established Target Trailing stop Now P/L Buy Sp183 (09/2015) vs. SGB1049 (08/2015) (Apr 25). Despite the trade reaching its Established target, we still see potential for tightening as we expect a Riksbank rate cut in December Target and believe remaining CB supply for the rest of this year is limited, especially in the 2015 Trailing stop segment. As of Apr 25, carry over 3m was 4.4bps. Now P/L Buy SEB568 (2015) vs. SEB567 (2014) and pay fixed in an SEB567/SEB568 fwd SEK Established IRS (Dec 6). Despite the trade reaching its target, we still see potential for tightening Target based on our expectations of a Riksbank cut in December. As of Dec 6, carry over 3m was Stop +2.6bps. Now P/L Note: Until the position is closed, P/L is expressed without taking carry into account and using mid-prices.

37bps 28bps 3bps 31.5bps +5.5bps 55bps 30bps 3bps 35.5bps +19.5bps 46bps -8bps 8bps -1bps +47bps

Hold

Hold

Hold

3

Fixed Income insights

RIKSBANK MOVING TOWARDS DECEMBER RATE CUT. Following the release of Swedish October inflation data on Nov 12, the scales tipped in favour of a rate cut and we changed our forecast to accommodate a 25bps cut to 0.75% in December. During the past week, markets have moved rapidly in our direction with current market pricing of RIBA futures and SEK FRAs implying a 68% (-17bps) likelihood of a December rate cut and a cumulative 82% probability (-20.5bps) that the Riksbank will cut the repo rate either in December or February. Our case for a rate cut has received support from the latest comments by two Riksbank board members, Skingsley and Jansson, both of whom we regard as key to determining the outcome of the next rate decision. Following his speech on Nov 18, Deputy Governor Jansson said to reporters that “clearly lower than expected inflation (CPIF 0.4%-points below the Riksbank’s forecast) has increased the probability of a rate cut.” In her speech on Nov 19, Deputy Governor Skingsley stated that the Riksbank “needs to ensure there’s confidence in the price target” and that the probability of a rate cut “certainly hasn’t diminished” following the latest inflation figure. While both Jansson and Skingsley said that little weight should be given to any single data release, apparently both are becoming more concerned by low inflation. They are therefore, in our opinion, likely to support a rate cut in December, alongside Deputy Governors Ekholm and Flodén. MORE TO COME FROM THE ECB. Recent comments by the ECB have fuelled expectations that it is preparing to take further action. Indeed, we see more to come from the ECB and will present a detailed discussion of the subject in the upcoming Nordic Outlook due on Nov 26. SUMMARY OF OUR VIEWS. To summarise our views discussed in this edition of Fixed Income Insights: • Slower than expected Swedish Q3 GDP growth is likely to confirm expectations that the Riksbank will cut the repo rate in December (page 5). The SGB1049 (2015) target of 0.75% remains intact (page 4). We see potential for further tightening of the 10y Sweden vs. Germany spread (page 6). We recommend steepening positions on the SEK IRS, preferably with swaptions (page 7). Jussi Hiljanen

In focus

TECHNICAL OUTLOOK: GERMAN 10Y BUND. The rejection from 1.77-1.79% has given the bears some respite. To reignite selling the market must break above last week’s high point, 1.80% (and by doing so exert pressure on the topside key point, 1.95%, the barrier to the 2.30/2.40% area). The above scenario can only survive if the market stays above 1.50%. A break below 1.68% will signal a move to around 1.501.64%. Therefore, considerable uncertainty will remain between 1.68/1.80%.
Yield

2
1,95 1,9 1,85 1,8 1,75 1,7 100,0% 1,64 1,65 1,6 16 23 30 07 14 21 28 04 11 18 25 02

september 2013

oktober 2013

november 2013

10Y SEK IRS. The answer to last week’s question (asking whether or not a correctional low had been set) was clearly and quickly answered as the market broke below the previous low point, 2.60%, paving the way for a new cycle low (2.52%). With the market immediately responding to important support (2008 trend line, 233d ma band and the July low), the set up for a possible base formation is better than last week. Ending the current week above 2.63% will make a possible upturn increasingly credible, while a break above 2.78% will largely confirm it.
Price SEK

3
2,9 2,8 2,7 2008 trend line 2,6 2.54% 2,5 2,4 08 15 22 29 05 12 19 26 02 09 16 23 30 07 14 21 28 04 11 18 25 02

• • •

jul 13

aug 13

sep 13

okt 13

nov 13

Anders Söderberg

4

Fixed Income insights

Swedish economy

GDP - FINAL NAIL IN THE COFFIN FOR THE RIKSBANK? Previously, we highlighted the mixed picture for Swedish macro indicators with, on the one hand, lacklustre monthly growth metrics, and on the other improving sentiment indicators and a surprisingly strong labour market. We expect this situation to continue with weak Q3 GDP growth (due Nov 29) expected to be 0.2%, 0.1% y/y. Following the large decline in manufacturing inventories announced last week, we now expect Q3 growth below the Riksbank’s forecast (0.5%, 0.4% y/y). A slightly complicating factor is that the investment survey from October (released this morning) suggests that business sector fixed investments may have declined less than earlier perceived implying that GDP in H1 2013 could be revised upwards by 0.2-0.4%-points. These means that the Riksbank’s forecast for average growth in 2013 (0.7%) probably is on track even if Q3 turns out to be weaker than expected. As so often, different indicators paint a mixed picture regarding the outlook for GDP, although on this occasion growth looks weak from both the production and expenditure side. This time around, uncertainty arises from the fact that very weak GDP growth contradicts positive signals from the labour market and recovering economic sentiment indicators. Regarding the outlook going forward, we attach greater credibility to these indicators than volatile quarterly GDP data and do not intend to substantially revise our 2014 GDP growth forecast of around 2.5%. FIRM CONSUMPTION GROWTH NOT ENOUGH. The table below shows our GDP forecast from the expenditure side. Despite expected satisfactory growth in both private and public consumption as well as fixed investments, predicted decreases in exports and inventories mean that working day adjusted GDP should remain unchanged y/y.
SWE: Q3 GDP forecast, % y/y Q3 f/c Private consumption Public consumption Fixed investments Exports Goods Services Imports Goods Services Inventories GDP GDP working day adj. Working hours (wda) Employment 2.1 1.5 1.0 -0.3 -1.9 2.0 -1.9 -3.6 3.0 -1.5 0.6 0.1 0.8 1.0 Q2 1.9 2.0 -3.0 -2.3 -3.9 1.3 -1.1 -2.6 3.7 0.3 0.6 0.1 1.2 0.7

As we project imports to fall faster than exports, the contribution from net foreign trade should be positive. However, this assessment is uncertain, due to the current large discrepancy between net exports calculated using monthly trade statistics and PPI versus the net trade contained in the National Accounts. This situation has occurred previously, especially when there have been major fluctuations in export and import prices. Our forecast assumes that the contribution from the net trade of goods both closes the gap and turns positive in line with the trade statistics, though this represents a downside risk for our Q3 GDP growth forecast.

A large decline in manufacturing inventories is another downside risk for Q3 growth. While we expect inventories to add -1.5%-points y/y, our models indicate a negative contribution closer to -2%-points y/y, despite a small increase in trade inventories.

PRODUCTION DATA ALSO WEAK. In addition, production data indicate weak GDP with a large decline in manufacturing production, sluggish activity within the service sector and slightly positive business sector output. With only a limited contribution from the public sector, production-side GDP appears largely consistent with our forecast. A low reading for Q3 growth could possibly be partly off-set by an upward revision to GDP for H1 2013. The October investment survey indicated that manufacturing fixed investment declined by 2%, which is significantly less than indicated in the May

5

Fixed Income insights

survey. We think that GDP growth in H1 2013 could be revised upwards by approximately 0.2-0.3% points. This means that the Riksbank forecast for average growth in 2013 (0.7%) is assessed to be on track, even if Q3 disappoints in line with our expectations. To conclude, monthly growth indicators suggest GDP will grow in line with or slightly below our projection of unchanged GDP y/y in Q3. Upside risks come mainly from the suggestion that both the labour market and sentiment indicators are at odds with GDP being largely unchanged so far this year. Finally, on Nov 29 Statistics Sweden will also present some revisions for last year’s GDP (currently estimated to be 1% y/y), which means that the scope for adjustments to historical GDP is larger than normal. Olle Holmgren

SGB1049 scenario analysis

TARGETING 0.75%. With market pricing having moved rapidly in our expected direction, further downside potential in the short end is becoming limited. In the chart below, we illustrate target levels for SGB1049 after the Riksbank’s December meeting based on alternative scenarios concerning market pricing of future repo rate expectations at that time. Alt 1 (main scenario): The Riksbank cuts the repo rate to 0.75%, the market prices the first rate hike to 1.00% in February 2015 and a repo rate of ~1.25% in SGB1049’s maturity. SGB1049 target: 0.77%. Alt 2 (hawkish): The Riksbank maintains the repo rate at 1.00%, the market prices the first hike in December 2014 and a repo rate of 1.65% in SGB1049’s maturity. SGB1049 target: 1.07%. Alt 3 (dovish): The Riksbank cuts the repo rate to 0.75%, the market discounts a 20% probability of another rate reduction and the first hike to 1.00% in April 2015. SGB1049 target: 0.72%.

Note: We have also assumed that SGB1049 vs. to the RIBA curve trades at -10bps after the December meeting.

While an unchanged repo rate with removed rate cut expectations would easily push SGB1049 to above 1.00%, we expect pressure on the Riksbank to remain if it decides to keep the repo rate unchanged in December. Accordingly, we believe market pricing will continue to indicate a substantial likelihood of a rate cut. This conclusion is supported by our view that the ECB will remain soft while moving towards implementing additional unconventional measures (for further details, see Nordic Outlook, Nov 26). TARGETING 0.75%. Overall, we maintain our postDecember meeting target of 0.75% in SGB1049. Expecting SEK swap spreads to remain near current levels over the next 1-2 months, we forecast 2y SEK IRS to decline to 1.10-1.15%. Jussi Hiljanen

6

Fixed Income insights

Swedish curve

US LONG RATES TO RISE, SPREAD TO GERMANY AND SWEDEN TO WIDEN. Our base case remains that US long yields will increase over the next few months as the Fed moves towards tapering. With the ECB remaining soft, we expect the 10y spread between the US and Germany to increase towards 120-130bps. Despite the spread widening, German and Swedish long rates are also likely to rise, although clearly less than those in the US. SWEDEN VS. GERMANY 10Y SPREAD LIKELY TO TIGHTEN FURTHER. The chart below shows the 10y Sweden vs. Germany government spread and our model fair value. Both models show that following recent tightening the spread level is now fair. Going forward, both models suggest further tightening potential over the next 3-6 months, followed by widening to around current levels in late-2014. For 2015, the model indicates further widening, driven by higher Swedish inflation and Riksbank rate hikes.

in coming months nearer to levels seen historically when the repo rate has been low.

CONDITIONAL STEEPENER IN 3M4Y VS. 3M10Y SEK SWAPS. Due to the uncertainties on long rates, we prefer to seek alternative ways to position for a steeper curve while hedging against lower long rates. Given our central bank views, we opt to express steepening by going slightly further out on the short end of the curve, to 3-4y instead of the 2y segment. We therefore express our view conditionally by selling an ATM payer in 3m4y SEK IRS and buying an ATM payer in 3m10y SEK IRS with DV01 neutral nominals at maturity for 3.0bps (as of November 21, 10.00 CET).

Note: Model 1 uses the relative short rates, growth, inflation and public finances to explain the 10y zero coupon yield spread between Sweden and Germany. Model 2 uses the EONIA rate instead of the ECB refi rate.

STEEPER CURVES. With the Riksbank expected to cut the repo rate and the ECB likely to present additional measures, our main scenario involves short rates remaining low in Germany and declining in Sweden (see “SGB1049 scenario analysis” above). Expecting long rates to rise, we foresee steeper 2y vs. 10y curves in coming months. With such a position largely directional to movements in long rates, a continued bond market rally (bull flattening) would present the greatest risks for a steepening position given the limited potential for 2y rates to decline further. We therefore regard 2y vs. 10y steepeners mainly as directional bets on the 10y rate. The chart below shows historical steepness of the 2y vs. 10y SEK swap curve and the repo rate. With central banks lending support to the short end and yields expected to increase, we expect the curve to steepen

Volatilities have generally increased during November, with mid-curve volatility catching up long-end volatility. Indeed, 3m5y normalised volatility is now a little higher than the 3m10y, though we prefer to sell a 3m4y payer instead because the 4 year rate is slightly more sensitive to central bank policy and less sensitive to long rates than the 5y rate. 3m4y normalised volatility is a few basis points lower than 3m10y, incurring a cost of 3.0bps with both legs ATM and DV01 neutral nominals at maturity. Selling the 3m10y some 6bps OTM would result in a zero cost strategy.

7

Fixed Income insights

Trade details: Payer Fwd Sell 3m4y 1.77 Buy 3m10y 2.65 88bps

Spot 1.69 2.60 91bps

Delta -47k +46k -1k

Vega -32k +46k +14k

Nom.* 230m 100m

*DV01 neutral at maturity.

Jussi Hiljanen & Lauri Hälikkä

8

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close