VIEWS: 0 PAGES: 73 POSTED ON: 5/5/2013
• The decline in commodity prices and inflation expectations gives room for central banks to be more accommodative. We recommend reinitiating 7s-30s curve flatteners in the US. Given current valuations, we also recommend being long 10y US Treasuries and JGBs relative to France. • In Europe, core yields look low, but with the BoJ moves and weak data, they are unlikely to back up much near term. In Sweden, we stay positive following the dovish Riksbank. • In the UK, data are coming in weaker as well, but the MPC is unlikely to give any more stimulus for now: we stick with steepeners.
Interest Rates Research 18 April 2013 Global Rates Weekly Summertime blues • The decline in commodity prices and inflation expectations gives room for central banks to be more accommodative. We recommend reinitiating 7s-30s Views on a Page 7 curve flatteners in the US. Given current valuations, we also recommend being Trade Portfolio Update 58 long 10y US Treasuries and JGBs relative to France. Global Supply Calendar 68 • In Europe, core yields look low, but with the BoJ moves and weak data, they are Global Bond Yield Forecasts 69 unlikely to back up much near term. In Sweden, we stay positive following the dovish Riksbank. United States • In the UK, data are coming in weaker as well, but the MPC is unlikely to give any TIPS: Global linkers Risk-Parity 11 more stimulus for now: we stick with steepeners. Agencies: A dam too far 19 Global Volatility: For a flatter tomorrow 23 Money Markets: Clearing the fog in interbank Summertime blues 2 lending 26 The decline in commodity prices and inflation expectations gives room for central banks to be more accommodative. We recommend reinitiating 7s-30s curve flatteners Europe in the US. Given current valuations, we also recommend being long 10y US Treasuries Sovereign Spreads: Political rally in Italy 32 and JGBs relative to France. Money Markets: The ECB’s options 35 United States SSA: EFSF – to tap or not to tap 42 Treasuries: Opportunity knocks twice 8 Scandinavia: Unchanged policy rates, but We recommend initiating 7s30s curve flatteners as the term premium embedded in the significantly softer forward looking policy guidance 48 long end looks too high, given subsiding inflation worries, as evidenced by the sharp decline in long-term breakevens. We also reiterate our long Tsys vs OIS view in the front European Inflation-Linked: Viva l’Italia… again! 50 end and believe the 4y sector looks too cheap with respect to the intermediate sector. UK Inflation-Linked: Supply me to the Euro Area moon 52 Volatility: Conditionally hedged 53 Resiliency 30 The resiliency of peripheral markets recently has been impressive, with yields/spreads to Germany near YTD lows. The IMF Fiscal Monitor and Global Financial Stability Reports (FM www.barclays.com and GFSR) include analysis that help explain this, and other moves in financial markets. UK INSTITUTIONAL INVESTOR ALL-AMERICA FIXED INCOME RESEARCH TEAM SURVEY Decline and fall 40 2013 As realised volatility declines, increased interest in positive roll-and-carry structures is Voting has begun in the Institutional becoming a driver of investment, this favours the 10-15y sector of the gilt curve and Investor All-America Fixed Income Research Team Survey 2013. Barclays consequentially, a re-steepening over Q2 13. would welcome your support. If you have not received a ballot, please click Japan on Institutional Investor's Rankings Assistance Page to request one. How the BoJ can achieve a ¥175trn current account balance 55 The BoJ plans to increase its current account balance to ¥175trn by end-2014. Domestic banks will have to cut their ¥160trn in JGB holdings to around ¥70trn. For this to be accomplished smoothly, yields might have to decline somewhat further. PLEASE SEE ANALYST CERTIFICATIONS AND IMPORTANT DISCLOSURES STARTING AFTER PAGE 70 Barclays | Global Rates Weekly GLOBAL THEMES Summertime blues Rajiv Setia The decline in commodity prices and inflation expectations gives room for central banks +1 212 412 5507 to be more accommodative. We recommend reinitiating 7s-30s curve flatteners in the firstname.lastname@example.org US. Given current valuations, we also recommend being long 10y US Treasuries and JGBs relative to France. Laurent Fransolet The rally in global rates that began in the middle of March continues, with 10y US yields +44 (0)20 7773 8385 about 40bp lower than their recent peak and 10y bunds within striking distance of their all- email@example.com time lows from last summer. The latest leg of the rally has largely been the result of economic data surprising to the downside globally: GDP growth in China surprised to the Amrut Nashikkar downside (at 7.7%, compared with consensus of 8%), while signs of an economic +1 212 412 1848 slowdown have emerged in the US. The regional manufacturing indices (Empire State and firstname.lastname@example.org the Philadelphia Fed) were weaker than expectations. Retail sales, which had shown signs of strong growth in the first quarter, were revised lower, and it is now clear that the payroll tax hikes are beginning to take effect. Also, as our economists have indicated, consumer spending in the first quarter was buoyed by higher-than-expected utility spending due to the cold winter, posing downside risks to Q2 consumption. Several leading indicators of activity – such as new orders, consumer confidence indices, and base metals prices – have declined, suggesting that data are likely to be soft for the next few months. Moreover, the drag from sequestration has yet to show up in the data. In response to the softer tone of recent data, consensus forecasts for Q2 GDP have The rally in rates has continued decreased by almost 50bp (in line with the Barclays view) over the past few months; because of weaker-than- however, in our view, consensus estimates for the second half still look too high (Figure 1). expected US economic data, Given the backdrop of slowing global growth, we are hard-pressed to see US rates move declining commodity prices higher on a sustained basis over the next few quarters. Rather, we believe that rates have and signs of a global economic room to rally, especially if the commodity complex continues to soften and the stock market slowdown also undergoes its customary summer swoon, as has been the case over the past few years. FIGURE 1 FIGURE 2 Consensus growth forecasts have decreased, in line with our Commodities back to their lows of last year, while rates, expectations stocks and inflation expectations have held up Real GDP Growth expectations, % Apr 2012 = 2.8 2.70 100 2.60 120 2.6 2.50 115 2.4 2.30 110 2.2 2.10 105 2.00 2.00 2.0 100 1.8 95 1.60 90 1.6 1.50 85 1.4 80 1.2 Apr-12 Jul-12 Oct-12 Jan-13 Apr-13 Ind metals Gold 1.0 Q213 Q313 Q413 CRB S&P500 Barclays Consensus (Latest) Consensus (1M Ago) 10y breakevens Source: Bloomberg, Barclays Research Source: Bloomberg, Barclays Research 18 April 2013 2 Barclays | Global Rates Weekly FIGURE 3 FIGURE 4 Inflation expectations declining across developed markets Wage growth has not accelerated through the recovery, thus with the exception of Japan tempering real growth and inflation bp change yoy % yoy % since 1/18 15 5.0 60 50 10 4.0 40 5 30 3.0 20 0 2.0 10 -5 0 1.0 -10 -10 -20 -15 0.0 -30 1990 1993 1996 1999 2002 2005 2008 2011 Bank loans, LHS Jan-13 Feb-13 Mar-13 Mar-13 Apr-13 JP BE swap France BE swap Avg hrly earnings - prod & n.s., RHS US BE swap UK BE swap Avg hrly earnings, all pvt, RHS Source: Barclays Research Source: Haver Analytics, Barclays Research The most significant market move over the past week was the dramatic decline in gold prices and the overall commodity complex (Figure 2). While commodity prices have likely declined for a variety of reasons, expectations of slower global growth and a stronger USD have likely contributed. Despite worries about a mid-year slowdown, growth prospects in the US look far better than those in Europe or Japan. Also, with the Fed arguably closer to the end of its easing cycle than other central banks, dollar strength should persist in coming quarters. The BoJ will likely be expanding its balance sheets for the next few years, while at the ECBs press conference, President Draghi opened the door to rate cuts if growth continues to underwhelm. In contrast, although we think it is premature, the FOMC minutes revealed that members are busy discussing the tapering of asset purchases. We believe that the decline in Despite this extensive discussion of tapering, speeches that have followed the most recent inflation expectations gives payroll report suggest that any such move remains distant. Fed Vice-Chair Yellen suggested room for the Fed to be more there were few signs that the low rate environment was leading to excessive risk taking. St. accommodative. Louis Fed President Bullard suggested that the Fed might need to defend its inflation target from the downside, and that a further decline in inflation expectations might even lead to an increase in the pace of asset purchases. Our view on near-term growth has long been that the market was incorrectly concerned with tapering; rather, given the subdued inflation backdrop, the real question should be what the Fed would do next if growth and/or inflation stayed below desired levels. More generally, inflation expectations have been declining globally, with the notable exception of Japan (Figure 3). This has occurred even as central banks globally have continued to provide unprecedented amounts of monetary accommodation. In our view, the failure of extremely accommodative monetary policy to translate to inflation is likely the result of two factors. First, financial systems have been constrained because of private sector deleveraging in the aftermath of the financial crisis; further, the pro-cyclical regulatory response – raising capital and liquidity requirements via Basel 3 – has not helped either. For instance, Figure 4 shows that bank loan growth has been stuck well below pre- crisis levels. Second, wage growth has been extremely low even during the recovery (Figure 4), translating into low real income growth as well as downward pressure on inflation. A recent IMF study also suggests that an inflation spike is unlikely even as the recovery 18 April 2013 3 Barclays | Global Rates Weekly FIGURE 5 FIGURE 6 The IMF forecasts debt load to be higher in France and France 10y rates looking rich relative to US and Japanese Belgium relative to other core European countries yields net debt/GDP 2.4 4.0 100 2.2 90 3.5 80 2.0 70 3.0 1.8 60 50 1.6 2.5 40 1.4 30 2.0 20 1.2 2010 2011 2012 2013 2014 2015 2016 2017 2018 1.0 1.5 Belgium Netherlands Apr-11 Aug-11 Dec-11 Apr-12 Aug-12 Dec-12 Germany France Austria JGB+US 10y avg, LHS France 10y, RHS Source: IMF WEO, Barclays Research Source: Barclays Research progresses1 because inflation has been less sensitive to economic slack through this recession and recovery than was historically the case. From a policy perspective, this implies global central banks should remain accommodative. Trade: Long 10y US and 10y JGBs versus 10y France If the commodity complex continues to weaken and stocks falter, we see some room for 10-year breakevens to cheapen; however, given the large decline over the past few days, we see far more room for 10-year real yields (currently at -58bp) to decline. If worries about tapering recede, and growth slows as we expect, real yields could easily decline 25bp or more from current levels. However, with nominal yields below 1.7% despite no signs of risk aversion from Europe, we are wary of initiating outright long duration trades in nominal treasuries at current levels. We would rather recommend 7s30s flatteners in the US (see the US Treasury section of this publication). The prospect of a real yield- We also believe the sharp moves in global duration markets in recent weeks have opened a driven rally in the US, a reversal few interesting trades. First, after a sharp decline in the immediate aftermath of BoJ of Japan’s most recent rate announcing it would expand the pace of its asset purchases from JPY2trn to JPY7trn per sell-off and the richness of core month, JGB yields have backed up significantly. 30y JGBs are now about 1.6%, above their Euro rates leads us to level before the BoJ surprised markets, making them attractive once more to Japanese recommend being long USTs private buyers. As portfolios rebalance on the back of JGB purchases, Japanese yields should and JGBs against France drift lower over time. Second, our European strategists believe that core European rates are too rich relative to fundamental value and have also recommended selling France against Netherlands (See “The new range”, April 11, 2013). Buying an equally weighted basket of Japan and US 10y treasuries and selling French debt is also attractive, in our view (Figure 6), and we are initiating this trade in our portfolio at a level of 65bp and a target of 80bp (the pre-BoJ level). We believe that amongst the core European countries, selling France is better. If the decline in inflation and growth expectations continues, it would worsen the fiscal outlook for countries with higher debt loads. This could again bring about greater differentiation within core European sovereigns, which would cause France to underperform in a rally (Figure 5). 1 See “The dog that didn’t bark: Has inflation been muzzled or was it just sleeping?”, from the IMF’s World Economic Outlook, April 2013. 18 April 2013 4 Barclays | Global Rates Weekly Further, as we suggested in Shock and Awe, Kuroda style, April 5, 2013, the rally in French debt that immediately followed the BoJ’s announcement was likely the market overreacting to the recent history of private portfolio purchases from Japan. At current levels, it is quite possible that actual flows from Japan in the weeks ahead are focused more on either the US (where MBS remains an attractive asset class) or domestically, which should also favor our recommended trade. Europe: Resilient and watching the PMIs European rate markets continued to rally across the board, in line with their global counterparts as equities and commodities came under pressure. Remarkably, though, peripheral markets have remained very resilient, generally outperforming the core countries. We look at some of the reasons behind this in the Euro rates strategy section of this publication. The bottom line is that we believe this good performance of peripherals will continue, with the belly of the curves in particular likely to outperform. In our view, in a low- yield environment with limited volatility and relatively few event risks on the horizon, the attraction of the carry available in peripheral curves will continue or actually widen, with some foreign investors (e.g., Japanese, North Europeans) who sold in 2011 eventually showing signs of coming back. The resilience of peripheral Some technical factors have also helped this good performance, particularly in Italy. First, markets should continue Italy managed to raise EUR17bn of retail funds in two days via the so-called BTP Italia. In the Euro inflation-linked section, we look at the implication this may have on other Italian inflation BTPs. A second positive factor is that the post-election uncertainty might come to an end (albeit temporary) soon, as a compromise seems to have been reached between the two biggest political parties regarding a presidential candidate (See the Euro Sovereign Spread section of this publication). Lastly, the recent volatile moves in commodities, currencies and stocks might have led to some leveraged investors closing positions, which have typically been short on peripheral debt markets. We stay long Italy and Spain in the 3- 5y sector. Core yields look low, but with Core markets bond yields have continued to benefit from the spill-over of the BoJ the BoJ moves and weak data, announcement: they are trading at their lows and look rich against fundamentals (by up to they are unlikely to back up 50bp in 10y Bunds). To be sure, there has not been a wall of buying: the moves have been much near term driven by expectations rather than actual flows at this stage, as investors are a bit reluctant to chase the market at current low yields. However, with the economic data having lost the little positive momentum from earlier in the year and the ECB open to additional easing, we do not expect much of a back-up in coming weeks. The ECB is not out of options The PMI data in the coming week will be important in that respect: a further weakening yet, and the PMIs will be key would strengthen expectations of an ECB response. In the Euro Money Markets section, we look at the ECB’s various options. While the issue of SME financing is key, there is no ‘magic wand’ to solve it. In our view, the most likely response would be a refi rate cut as early as May (although its effect would be limited, there would be some signalling value in it) or new LTROs around the summer (to counteract the passive tightening of liquidity conditions that is likely to occur as banks keep repaying the maturing LTROs). Both actions would support the back end of the money market curve. In any case, we expect short-end rates to remain low for a long period of time. In Sweden, we stay positive Following the Riksbank meeting, in which the forward-looking policy guidance was following the dovish Riksbank significantly softer than the market expected (but in line with our view), we continue to see value in receiving SEK10y swaps versus EUR despite recent outperformance (See the Scandinavia Rates Strategy section). We prefer expressing a bullish view in such cross market positions (e.g, Sweden, peripherals) rather than by being long outright 10y Bunds or core markets. 18 April 2013 5 Barclays | Global Rates Weekly In the UK, data are coming in In the UK, the Minutes of the April MPC meeting offered little news to the market as the weaker as well, but the MPC is Committee seems at an impasse between members looking for more support for the economy unlikely to give any more and those concerned by the persistence of inflation. We expect Q1 13 GDP to be 0.0% q/q but stimulus for now: we stick with with risks to the downside. While carry and roll are highest in the steepest part of the curve, steepeners further curve steepening may be driven by longs in the 10-15y part of the curve, where carry- vol ratios are historically high. In the UK inflation linked section of this publication, we also cover the upcoming heavy supply and highlight some relative value opportunities. 18 April 2013 6 Barclays | Global Rates Weekly VIEWS ON A PAGE US EUROPE JAPAN Direction • Economic data in the US remain modest, with the fiscal deal and • The EGB market continues to be resilient, supported by the bold • The decline in liquidity resulting from the BoJ’s latest the sequester likely to exert significant drag. BoJ action. Peripheral supply is limited to an Italian linker and easing has raised volatility across the entire yield curve. If • Bigger-than-expected easing by the BoJ is likely to be supportive CTZ zero coupon auctions next week. As such, the focus will yields through the intermediate sector stabilize, we also of US yields as well. stay on political developments ie, the election of a new President expect long-term yields to stabilize, which would make a in Italy, which could lead to a new government being bull-flattening position attractive. In terms of direction, we • We remain neutral on duration. established, likely further supporting Italian and other perpiehral recommend a long position. We also see room for 10y- EGB markets. 20y flattening. Curve/ • We recommend 7s30s curve flatteners, as the curve looks too • Hold on to receive EUR 5y5y/5y10y/5y15y fwds. • 5x5-10x10 flattener curvature steep, given inflation expectations. • UK: longer-dated nominal gilt yields remain rich, underpinned by • 5s10s steepener • We maintain our long front-end Tsy vs. OIS view, given low real rates. Reset Gilt 5/30s or Gilt 10/30s steepeners into Q2 • 10s20s flattener improving financing conditions. The 4y sector looks cheap. 13 supply. Declining realised volatility supports carry-and-roll • Shorten on the Cs STRIPS curve into the 10-12y area; switch out structures – carry: vol ratios are at their cheapest in the 10-15y of rich 20y Tsy P STRIPS to 20y REFCO P STRIPS. part of the curve. Swap • Neutral on 30y spread wideners, considering risks from risky • EUR: Hold on to short Bobl ASW on Ger 5s/30s ASW box • 20s30s box (20y long) spreads asset underperformance. • GBP: APF transfer may improve the fiscal position, but • 10s20s box (20y long) • 1y1y Libor-OIS tightener hedged with 1y1y 3s1s widener. fundamentals remain poor. We remain negative on 10y gilt ASW. Long 5y ASW versus OIS or versus 10y (both versus 6mL). Other • We continue to favor long-end agency-Treasury spread • SEK: Hold SEK/EUR 10y tighteners in swaps and longs in SEK Sep • Pay USD/JPY 1yx1y basis spread tighteners but find the most upside potential only in the super- ’13 3m FRA tighteners. • Pay USD/JPY 4y basis. sectors long end. 7s have underperformed along the curve and now • Hold Spain and Italy 2s/5s/10s. offer over 20bp of spread pickup to Treasuries; shorten duration to 10s with no spread give-up. • Hold Spain 5s/10s/30s. • We remain constructive on Canadian covered bonds, given their • Long 5-8y Netherlands versus France relative isolation from Europe and continued significant spread • Long FRTR Oct 19/Oct 22/Apr 26 fly pickup to agencies. Pockets of value persist in USD SSA space. Inflation • Long 5y5y breakevens, given the recent decline and dovish Fed. • Long 10y Euro HICPx swaps • Breakevens have started to rise even with negative carry Neutral on spot breakevens. 5y and under breakevens could • IL22/IL62 breakeven steepeners or real yield flatteners attractive intact, and we expect this trend to continue for now. narrow further. in UK. The long-end remains cheap on the curve, and supply Over the short term, there is probably an opportunity for • The recent uncertainty argues for lower real yields, given the favours an outperformance of long linkers versus nominals. capital gains. However, it is difficult to determine where slower payroll growth and low realized inflation. levels will settle over the medium to long term. In establishing long positions over such a horizon, we • Should risk aversion increase, we would expect April18s to recommend paying attention to levels. richen versus Jan18s as the real yield curve flattens and the floor on April18s richens. Volatility • Sell 1y*10y straddles to benefit from range-bound rates and • Buy EUR1y*30y 100bp wide risk reversal (long receivers) to • Look to long vol when market stabilizes in late April supply from callables. hedge a risk flare in the eurozone. • Long 3y*2y vs. 3E and 4E mid curve straddles as a limited-loss • Buy 1y*5y receivers funded with 1y SL 5-30y curve cap to way to benefit from a tight range in rates. benefit from EUR rates staying low for long. • Long 3x13 Libor cap-floor vs 3y10y swaption straddle to • Buy EUR 6y*5y versus 1y*(5y5y) to position for steepening of position for a steepening of the vol surface. the vol surface and monetise the range in rates. Source: Barclays Research 18 April 2013 7 Barclays | Global Rates Weekly UNITED STATES: TREASURIES Opportunity knocks twice Anshul Pradhan We recommend initiating 7s30s curve flatteners as the term premium embedded in the +1 212 412 3681 long end looks too high, given subsiding inflation worries, as evidenced by the sharp email@example.com decline in long-term breakevens. We also reiterate our long Tsys vs OIS view in the front end and believe the 4y sector looks too cheap with respect to the intermediate sector. The Treasury market rallied on The Treasury market rallied over the week as economic data surprised to the downside and the back of weaker economic inflation worries subsided. Retail sales at the end of last week came in much weaker than data and subsiding inflation expected, with revisions to previous months’ prints suggesting that the tax hike was finally worries having an adverse affect on consumer spending; our economists’ estimate for personal consumption in Q1 was revised lower to 2.3% from 3% and they expect it to fall further to 1% in Q2. Michigan consumer confidence also surprised to the downside. The highlight of the week, however, was the move in commodities and inflation expectations. Figure 1 shows that the aggregate commodities index has declined 5% since the beginning of the month, with gold still 13% lower. While weaker-than-expected global growth is certainly playing a role, it seems that investors are less worried about the unintended consequences of easy monetary policy, given that inflation has been well behaved, declining to below the Fed’s target; core PCE inflation has been only 1.25% over the last year. We recommend initiating Not surprisingly, the latest decline in nominal rates has been driven by TIPS breakevens; 7s30s curve flatteners, as the Figure 2 shows that even as 10y real yields have risen modestly, 10y breakevens have curve looks too steep, given tightened 25bp from the beginning of the month (Figure 2). We had turned neutral on our inflation expectations 7s30s Treasury curve flattening view after the sharp reaction to BOJ’s easing on April 5, which brought the curve closer to fair value (see Bold action, bold response). However, with long-term inflation expectations having fallen quite sharply since then, we believe the long end is again looking attractive. We recommend reinitiating 7s30s curve flatteners, which should also benefit if renewed concerns about global growth filter through into the equity markets. Figure 3 shows that after accounting for inflation expectations and the monetary policy embedded in the intermediate sector, the 7s30s curve is trading 9bp too steep. With limited room for front-end rates to rally, were 10y inflation expectations to decline further to their July 2012 level, the fair value of the curve would be another 3bp lower. FIGURE 1 FIGURE 2 Commodities have sold off across the board The rally has been driven by decline in inflation expectations; real rates have risen modestly 310 1800 2.7 -0.4 305 1700 2.6 -0.5 300 2.5 1600 -0.6 295 2.4 1500 -0.7 290 2.3 1400 -0.8 285 2.2 280 1300 2.1 -0.9 275 1200 2.0 -1.0 Jan-13 Feb-13 Mar-13 Apr-13 Oct-12 Dec-12 Feb-13 Apr-13 CRY Index, lhs GOLD ($/oz), rhs 10y breakevens, %, lhs 10y real rates, %, rhs Source: Bloomberg Source: Barclays Research 18 April 2013 8 Barclays | Global Rates Weekly FOMC members have recently Further, the most hawkish FOMC member, St. Louis Fed President Bullard, has suggested that become vocal about defending the Fed should defend inflation from the low side as well. He noted that were inflation to fall inflation from the low side further, he would favor more purchases. With modest inflation trends, it is likely that the term premium at the long end of the curve could decline substantially in that scenario. The latest primary dealer survey suggests that expectations are actually skewed toward a tapering of Fed purchases, so an increase in purchases would be a surprise to the market. Figure 4 shows that that median and even the 75th percentile expectations are for the status quo until the end of the year and the 25th percentile is for the Fed to taper Treasury purchases to an average of $30bn/m in Q4 13 (from the current $45bn/m) and agency MBS purchases to an average of $28bn/m (from the current $40bn/m). The median expectation is for the Fed to not buy any agency MBS after 2013. Term premium at the back end Figure 5 shows the term premium embedded in the long end of the curve against 10y CPI of the curve is too high swap rates. Even as long-term inflation expectations have fallen, the term premium has remained elevated; the last time 10y CPI swap rates were trading at similar levels, the slope FIGURE 3 FIGURE 4 7s30s curve is trading too steep, given inflation expectations Expectations are skewed towards a tapering of purchases 15 Monthly Run Rate, $bn Q313 Q413 10 Tsys Agency MBS Tsys Agency MBS 25th Pctl. 45 40 30 28 5 Median 45 40 45 40 0 75th Pctl. 45 40 45 40 -5 -10 -15 Apr-12 Jun-12 Aug-12 Oct-12 Dec-12 Feb-13 Apr-13 7s30s curve residual, bp Source: Barclays Research Source: New York Fed FIGURE 5 FIGURE 6 Term premium at the back end is too high 2y Tsy-OIS basis is too wide, given near-term financing conditions 0.70 3.05 14 10 12 0.65 2.95 10 8 0.60 2.85 8 6 0.55 2.75 6 4 0.50 2.65 4 2 0.45 2.55 0 2 -2 0.40 2.45 -4 0 0.35 2.35 Jul-12 Sep-12 Nov-12 Jan-13 Mar-13 2y Tsy-OIS basis, bp, lhs 0.30 2.25 Apr-12 Jul-12 Sep-12 Dec-12 Mar-13 3M Futures implied GC-FF spread, bp, lhs 10y1y-7y1y curve, %, lhs 10y cpi swap rates, %, rhs Spread, bp, rhs Source: Barclays Research Source: Barclays Research 18 April 2013 9 Barclays | Global Rates Weekly of the back end of the curve was 15bp lower. Also, as Figure 2 shows, real yields have not declined in this move, which suggests that were expectations of asset purchases to rise, there would be room for real yields to fall. Hence, even though the curve is trading close to the lows following the BOJ announcement, we believe it looks quite steep and recommend re-initiating 7s30s curve flatteners. Opportunities in the front end We reiterate our front-end Tsy- We continue to recommend being long Treasuries vs. OIS in the front end, and with the OIS view relative cheapening of the 4y sector, we believe investors should also consider overweighting the 4y sector vs. the 6y sector. We have argued in the past that the funding premium embedded in 2y Tsy-OIS is too wide, Figure 1 shows that the even as the 2y Tsy-OIS basis has steadily tightened from Operation Twist levels, the funding premium has actually widened, as the basis has not kept pace with the improvement in financing conditions. Even though the 3m GC-FF basis is priced to trade at close to 0 over the next three months (and 1bp over the next year), the 2y Tsy-OIS basis still looks wide at ~7bp, as it implies a 1y1y spread of almost 12bp. Alternatively put, it is highly unlikely, in our view, that were GC to average close to FF for a year, the market would still price it to trade 12bp above FF over the subsequent one year. If the forward basis was wide due to the expectation of reserve draining, then it should have tightened, as weaker- than-expected economic data and subsiding inflation worries have pushed out hike expectations. We therefore continue to like being long Tsy vs. OIS in the front end. The 4y-6y inversion of the Tsy- Similarly, the 4y sector now appears too cheap versus the intermediate sector. Figure 7 OIS curve is overdone shows that the 4y Tsy-OIS basis is now trading at 10bp, compared with just 4bp for securities in the 6y area; it used to trade 1-2bp cheaper. The discrepancy is even starker in the forward space. Figure 8 shows that the 1y Tsy-OIS basis 3y forward is still trading at 15bp, even as the 1y basis 5y forward has tightened to -10bp. The improvement in Treasury borrowing prospects with the sequester going into effect should have benefited the 4y sector as well. Under the CBO’s baseline, net borrowing needs over the coming years would be lower than net coupon issuance were the Treasury to keep auction sizes unchanged. This would either translate into negative net T-bill issuance or the Treasury would end up reducing front-end auction sizes, in our view. Either of these scenarios is likely to benefit the front end as well. Hence, we think the inversion of the Tsy-OIS basis curve is overdone. FIGURE 7 FIGURE 8 4y sector looks cheap on the Tsy-OIS curve, especially with Forward Tsy-OIS curve is sharply inverted (bp) respect to the 6y sector (bp) Spd vs OIS 40 12.0 30 10.0 20 8.0 10 6.0 0 4.0 -10 years to maturity 2.0 1.0 1.4 1.7 2.1 2.6 3.0 3.5 4.0 4.5 5.0 5.8 -20 Oct-12 Nov-12 Dec-12 Jan-13 Feb-13 Mar-13 2y series 3y 5y 7y 1y Tsy-OIS basis, 3y fwd 1y Tsy-OIS basis, 5y fwd Source: Barclays Research Source: Barclays Research 18 April 2013 10 Barclays | Global Rates Weekly INFLATION-LINKED MARKETS: UNITED STATES Global linkers Risk-Parity Michael Pond This article was previously published in the Global Inflation-linked Monthly, 18 April 2013. +1 212 412 5051 We apply a Risk-Parity framework to a global linkers portfolio. Here, our findings indicate firstname.lastname@example.org that equal risk allocations would have outperformed market-value based portfolios in spite of having an almost identical vol profile. Similarly, for each of the global regions, we Chirag Mirani found that Risk-Parity portfolios outperform market-value portfolios. +1 212 412 6819 email@example.com Continuing on the Risk-Parity theme In The Real Risk Parity article, we highlighted that a TIPS-Equities Risk-Parity (RP) portfolio outperforms an Equities-Nominals RP portfolio, and the latter also outperforms a 60/40 Equities-Nominals portfolio. In other words, over the past 15 years (and over a 40-year period using simulated TIPS returns), a TIPS-Equities RP portfolio significantly outperformed a traditional 60/40 Equities-Nominals portfolio, while maintaining the same level of risk. In our previous exercise, we recognised that such allocations may not always be optimal given varying economic conditions (especially in terms of absolute returns) and the use of leverage may also be suboptimal in flight-to-liquidity events, as TIPS are less liquid than Using US TIPS-nominals Risk- comparable nominals. However, monthly resets on relative RP weights and a close watch on economic conditions should give active investors some comfort in this asset allocation Parity as a framework, we set approach. In this article, we set out to explore and answer a number of questions, namely: out to explore Risk-Parity 1) with the significant improvements in the Sharpe ratio with RP-based allocations in the allocation results in the global US, whether the RP approach backtested well within a global linkers market; 2) within linkers market… individual global linkers markets, do RP linkers-nominals portfolios outperform market- value(MV) based linkers-nominals portfolios? and 3) if they do, what linkers versus nominals allocation should each of the global region have? Does RP approach backtest well within a global linkers market? Yes. The key premise of a RP portfolio is that for a portfolio to be truly diversified, the diversification must occur at the risk level and not at the dollar amount invested. In other words, the risk contribution from each asset must be identical or close to it in an ex-post analysis. FIGURE 1 FIGURE 2 WGILB MV portfolio risk factors dominated by the asset with WGILB Risk-Parity portfolio is truly diversified in terms of the highest market value, the portfolio is not diversified in risk in an ex-post historical analysis terms of risk MV Risk RP Risk Profile Profile 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Jan-02 Jul-03 Jan-05 Jul-06 Jan-08 Jul-09 Jan-11 Jul-12 Jan-02 Jul-03 Jan-05 Jul-06 Jan-08 Jul-09 Jan-11 Jul-12 Aus CAD Sweden UK US EUR Govt Aus CAD Sweden UK US EUR Govt Source: Barclays Research Source: Barclays Research 18 April 2013 11 Barclays | Global Rates Weekly A market value portfolio’s A global linkers portfolio (WGILB), like all benchmark portfolios, is weighted to each asset’s variance is typically explained total market-value (MV) outstanding (certain other conditions also apply in terms of securities by the largest market values; that make it into the benchmark portfolio). Thus, a single asset class can potentially explain a the portfolio is not diversified in large amount of portfolio variation depending on its market value relative to total MV. terms of return volatility Currently, in the global linkers portfolio (ex-Japan), Australian, Canadian, Sweden, UK, US and Euro govt linkers are weighted 1.42%, 3.25%, 1.80%, 29%, 48.6% and 16%, respectively. Building from intuition, all else equal, in a WGILB MV portfolio’s historical returns variation timeseries, we would expect US, UK and EUR government linkers to contribute most of the variation. Indeed, Figure 1 shows, ex-post, that UK, US and EUR Govt linkers dominated the risk-contribution in the MV WGILB portfolio. Risk-Parity portfolios are Meanwhile, Figure 2 shows that the RP portfolio risk decomposition is starkly different than balanced in terms variance the MV WGILB portfolio. In the RP portfolio’s risk profile, over time, the risk contribution contributions… from each linker’s asset class has been roughly the same. Specifically, we constructed an RP portfolio using equal vol weights. Monthly reweighting were such that each asset class would have a reciprocal weight with respect to the previous year’s standard deviation (this is just a monthly return standard deviation over the previous 12 months), in the context of total standard deviations. In this exercise, we use 1y rolling standard deviation rather than 3y (as was done in the previous US focused exercise) rolling because of the shorter total return series available for most ex-US linkers markets. Simply stated, in a RP portfolio, a risky asset will have a lower weight while a less risky asset would have a higher weight. For this intuition development exercise, we operate under the assumption that all global linkers are independent and therefore have a low correlation and covariance structure. At the current vol levels (at the end of March 2013), a RP portfolio assigns Australian, Canadian, Swedish, UK, US and Eur Govt linkers 13.6%, 12.6%, 21.9%, 9.9%, 23% and 19% weights, respectively. We realize that such weights may not be feasible for a very large notional portfolio given the smaller market size of Australian, Canadian and Swedish linkers. However, for this exercise, we use these RP weights to gain intuition about how RP linkers portfolios have behaved over time. Next, we examine return and risk characteristics of aggregate RP portfolios versus MV WGILB portfolio. We are diversified in terms of risk in a RP portfolio, but did that deliver better returns versus market value portfolio? Yes. While the RP portfolio’s risk profile is aesthetically appealing (Figure 2), the key question investors may have is whether the RP portfolio outperforms a MV WGILB portfolio? Figure 3 FIGURE 3 FIGURE 4 RP WGILB portfolio has significantly outperformed MV RP WGILB portfolio broadly has the same vol profile as MV WGILB portfolio WGILB portfolio, in spite of a leverage factor 240 12% 220 10% 200 8% 180 160 6% 140 4% 120 2% 100 Dec-00 Jun-03 Dec-05 Jun-08 Dec-10 0% Dec-01 Dec-03 Dec-05 Dec-07 Dec-09 Dec-11 MV WGILB Risk-Parity WGILB WGILB Vol Risk-Parity Vol Source: Barclays Research Source: Barclays Research 18 April 2013 12 Barclays | Global Rates Weekly shows that, indeed, an RP WGILB portfolio has significantly outperformed MV WGILB portfolio since 2000. But, a RP portfolio does use leverage and so the risk must be higher. In fact, in Figure 4, we see that the ex-post standard deviation of a RP WGILB portfolio is almost identical to a MV WGILB portfolio’s historical standard deviation, notwithstanding the 2008 crisis. In other words, RP portfolios offer higher risk-adjusted returns or Sharpe ratios relative to standard MV portfolios. As a reminder, the leverage factor is calculated by dividing the 1y standard deviation of the monthly RP returns by 1y standard deviation (of monthly returns) of the MV WGIB portfolio. Despite having the same total Figure 5 summarizes key aspects of the RP versus MV WGILB portfolios over the past decade. variance as market-value A few highlights stand out in this table: one is that the RP portfolio has outperformed the MV global linkers portfolio, RP portfolio in seven out of the past 10 years and by a significant margin (which is confirmed in global linkers portfolios have Figure 3 as well). Returns have always been positive, though this is a function of global real delivered higher returns yields declining sharply over the past decade. Second is that Swedish linkers have typically held the highest weight, given their low volatility, while UK and European linkers have had lower weight recently, likely on European uncertainty and, for the UK, the recent formula uncertainty. One caveat to note is that in 2012, the RP WGILB portfolio underperformed the MV WGILB portfolio by almost 240bp; however, the leverage factor for the portfolio was only 0.85, which means that the 15% of RP portfolio was in cash or it could have been invested somewhere else. Based on the table below, it is likely that RP investors can choose the lowest risk-profile linkers such as the US TIPS and Swedish linkers and significantly outperform the MV WGILB portfolio when a relative vol leverage factor is applied. 2) Within individual global linkers markets, do RP linkers-nominals portfolios outperform market-value (MV) based linkers-nominals portfolios? and 3) if they do, what linkers versus nominal allocation should each of the global region have in a nominal/linkers portfolio? Yes, and see table and discussion below for RP linkers-nominals weights in individual markets. From the above exercise, we learned that there is historical merit to balancing risk across assets in a global linkers portfolio. In “The real Risk Parity” article, we only tested the RP approach in a US TIPS/Nominals portfolio. There, we showed that US TIPS should have close to 55% allocation versus comparable nominals in a RP portfolio, with a leverage factor of 1.12. A 12% additional cash-borrowing is needed, at which point the RP portfolio would have the same risk level as the market-value (MV) TIPS-Nominals portfolio. FIGURE 5 Risk-Parity (RP) versus Market-Value (MV) WGILB portfolio details, vol, returns, RP leverage factor, RP weights for each region Vol Returns RP Weights RP WGILB Return Stdev (with Differential MV WGILB leverage MV WGILB RP WGILB between MV Leverage Date Stdev factor) returns returns and RP factor Aus CAD Sweden UK US EUR Govt 12/31/2002 5.0% 5.6% 11.6% 14.4% 2.8% 1.20 26 14 16 15 13 17 12/31/2003 6.7% 6.6% 6.5% 6.5% 0.0% 1.19 18 18 18 17 12 17 12/31/2004 4.4% 4.4% 8.0% 11.3% 3.3% 1.17 16 14 14 20 12 25 12/31/2005 3.5% 3.7% 5.6% 8.6% 3.0% 1.08 19 13 15 21 16 17 12/31/2006 4.0% 4.0% 1.4% 1.8% 0.4% 1.02 18 12 19 15 18 18 12/31/2007 3.9% 3.5% 7.6% 5.8% -1.7% 1.16 20 15 17 14 14 21 12/31/2008 10.5% 10.0% 0.5% 4.3% 3.8% 1.44 25 9 29 10 11 16 12/31/2009 5.8% 5.8% 8.5% 7.3% -1.2% 1.41 19 10 26 11 14 20 12/31/2010 3.7% 4.1% 5.3% 6.8% 1.5% 1.07 17 19 19 13 16 17 12/31/2011 2.9% 3.5% 10.5% 10.5% 0.0% 0.85 17 13 23 14 21 13 12/31/2012 3.6% 3.2% 5.4% 3.0% -2.4% 0.93 14 14 24 12 22 13 Source: Barclays Research 18 April 2013 13 Barclays | Global Rates Weekly Global linkers-nominals RP Given the outperformance in the US RP linkers-nominals portfolio, we wondered how RP portfolios show higher linkers/nominals portfolios have fared in the UK, European, Sweden, Australian and risk-adjusted returns relative Canadian markets versus the respective MV portfolios. The table below confirms the same to their counterpart findings for non-US linkers markets: in each region, RP portfolios deliver higher returns market-value portfolios adjusted for risk. Clearly, for a large portfolio and/or portfolios that do not allow for leverage, RP allocations may not be feasible. But, the method and results are certainly worth investigating as a starting point. One can augment this analysis by forecasting individual market economic regimes (inflation, real growth) and adjusting RP weights based on anticipated asset class volatility. Barring flight-to-liquidity episodes, smaller portfolios stand to benefit from a RP weighting approach, in our view. In the subsequent section, we discuss nuances, including vol profiles, of each of the regional RP linkers-nominals portfolios. We also learn that relative liquidity of assets is a key factor in determining subsequent periods’ RP portfolio returns. FIGURE 6 RP Linkers-Nominals portfolio outperformance, weights and leverage recommendations Risk-Parity Weight MV Weight RP outperformance Leverage to versus MV portfolio replicate MV Country Linkers Nominal Linkers Nominal (cumulative) Portfolio US 55% 45% 14% 86% 25% 1.13 UK 47% 53% 26% 74% 10% 1.02 Euro Govt 47% 53% 5% 95% 4% 1.00 Sweden 55% 45% 27% 73% 8% 1.05 Australia 51% 49% 11% 89% 33% 1.00 Canada 46% 54% 14% 86% 32% 0.95 Source: Barclays Research UK: RP UK Linkers/Nominals portfolio outperforms MV UK Linkers/Nominals In the UK, linkers make up about 26% of the total government bond market, in terms of market-value (MV). Our analysis indicates that since 1998, UK fixed income investors forming a linkers-nominals bond portfolio would have benefited significantly by employing the risk-parity (RP) weighting approach. Relative to a MV weight approach, the RP portfolio has had a cumulative outperformance of almost 975bp since 1998 (Figure 7), while FIGURE 7 FIGURE 8 UK RP linkers-nominal portfolio outperforms MV linkers- UK RP portfolio achieves this outperformance while nominals portfolio maintaining the same level of risk 350 12% 300 10% 8% 250 6% 200 4% 150 2% 100 0% Jan-98 Jul-00 Jan-03 Jul-05 Jan-08 Jul-10 Jan-13 Jan-99 Jul-01 Jan-04 Jul-06 Jan-09 Jul-11 UK MV (Linkers-Nominal) Index vol (annualized) UK MV (Linkers-Nominal) Index UK Risk-Parity vol (annualized) UK Risk-Parity Source: Barclays Research Source: Barclays Research 18 April 2013 14 Barclays | Global Rates Weekly maintaining the same risk-level as the MV portfolio. This outperformance is again similar to what we found in the US RP TIPS-Nominals portfolio. In the UK, the RP portfolio (on 1y standard deviation rolling basis) recommends having 47% of the allocation to UK linkers and the rest in comparator nominal index, with the current leverage factor of 1.02 (2% additional cash-borrowing). In Figure 8, we see that the outperformance was achieved while maintaining the same level of risk. UK RP linkers-nominals Like all RP linkers-nominal portfolios, the UK RP minus MV cumulative outperformance portfolio outperformed MV profile indicates a significant drop in gains during the 2008 financial crisis (Figure 9). linkers-nominals portfolio from Subsequently, the RP portfolio regained almost all of the returns. In 2012, the RP portfolio 1998-2012 took some losses as UK linkers suffered versus nominals on the back of UK inflation formula uncertainty. Our conclusion from these performance gains is that the relative dislocations in the underlying assets matter a great deal in determining performance of an RP portfolio. We recommend that investors suspend RP portfolios if they anticipate significant dislocations between the underlying assets. The RP approach recommends overweights (relative to MV allocations) for both UK and Swedish linkers. These are the two markets where the linking inflation measure incorporates mortgage interest payments, and therefore insulates to some extent performance against central banks hiking rates. FIGURE 9 UK RP linkers-nominals portfolio returns minus UK MV Portfolio returns, profile shows gains overtime 12% 10% 8% 6% 4% 2% 0% -2% -4% -6% Jan-98 Jul-00 Jan-03 Jul-05 Jan-08 Jul-10 Jan-13 UK Risk-Parity Cumulative Outperformance Source: Barclays Research European: RP portfolio outperforms but during liquidity events it underperforms sharply In European linkers versus nominals, we have the most interesting case of RP implementation because of the relative illiquidity faced by linkers versus nominals in the 2008 financial crisis and European sovereign debt crisis. On a MV basis, European linkers constitute about 5% of total sovereign coupon debt, which makes linkers-nominals RP portfolio very susceptible to systemic liquidity changes between linkers and nominals. From 2001, the RP portfolio has outperformed (Figure 10) but since the financial crisis, the RP portfolio has significantly underperformed a market-value based linkers-nominals portfolio (Figure 12). The underperformance of the RP portfolio (Figure 12) coincided with spikes in RP vol relative MV vol (shown in Figure 11). 18 April 2013 15 Barclays | Global Rates Weekly FIGURE 10 FIGURE 11 European RP linkers-nominals portfolio outperforms European RP portfolio mostly tracks MV portfolio but during European MV linkers-nominals portfolio pre 2008 but has the 2008 financial crisis and European sovereign debt crisis, faltered recently due to relative liquidity changes RP vol spiked relative to MV vol 200 9% 190 8% 180 7% 170 6% 160 5% 150 4% 140 130 3% 120 2% 110 1% 100 0% Dec-01 Jun-04 Dec-06 Jun-09 Dec-11 Dec-01 Jun-04 Dec-06 Jun-09 Dec-11 Euro MV (Linkers-Nominal) Index Euro MV (Linkers-Nominal) Index vol (annualized) Euro Risk-Parity Euro Risk-Parity vol (annualized) Source: Barclays Research Source: Barclays Research Just prior to the 2008 financial crisis, the European RP portfolio had a linkers weight of 55% and had a leverage factor of 1.17 (17% additional cash borrowed). In the financial crisis, the sovereign bond market became more fragmented due to lower liquidity of linkers relative to nominals and as inflation expectations declined, the RP portfolio underperformed significantly. Similar issues also transpired around the European sovereign crisis in 2010- 2012. We think investors should exercise a good amount of caution and economic intuition in the face of liquidity events and opt for the most liquid assets if they are likely to face acute cash needs during those times. Clearly, the RP employed leverage and the relative liquidity of assets can be the biggest pitfalls in a RP-based allocations. In fact, a diversified European-based RP portfolio likely underperformed in the European sovereign financial crisis, as peripheral yields sold off, and equity markets and commodities declined. In summary, while RP linkers-nominal portfolio have outperformed since 2001 in Europe, over the past four years it has underperformed sharply, especially during European sovereign liquidity events. Currently, in a linkers-nominals RP portfolio, European linkers are assigned a weight of 47% with a portfolio leverage factor of 1 (No leverage is employed). FIGURE 12 European RP linkers-nominal portfolio has lagged MV linkers-nominal portfolio since 2008 financial crisis due to relative illiquidity between linkers and nominals 14% 12% 10% 8% 6% 4% 2% 0% -2% Dec-01 Jun-04 Dec-06 Jun-09 Dec-11 Euro Risk-Parity Cumulative Outperformance Source: Barclays Research 18 April 2013 16 Barclays | Global Rates Weekly Australia: RP AUD linkers/nominals portfolio versus MV AUD linkers/nominals In Australia, the linkers (Treasury Indexed Bonds, TIBs) market is about 10% of the government bond market, with a total TIBs market value of about AUD27.28bn (USD28.66bn). Using a MV approach, investors can construct a 10% AUD linkers and 90% comparable nominal portfolio. Our analysis (Figure 13) indicates that investors in the Australian linkers and comparable nominal bond market could have benefited significantly by employing a RP portfolio approach when assigning relative asset allocation weights. The outperformance of a RP weighting approach relative to a MV approach is very similar to the one we found in the US TIPS/Nominals portfolio in “The real Risk-Parity” article. This gives us some confidence that small active investors can benefit significantly by assigning higher weight inflation-linked assets relative to the nominal assets. In this case, a RP asset allocation approach assigns a 51% allocation to linkers versus 49% to comparable nominals, with a leverage factor of 1 (ie, no additional borrowing). Also worth noting is that the RP portfolio attains this outperformance with the same level of risk as the market-value portfolio (Figure 14), ie, the RP linkers/nominals portfolio has a higher Sharpe ratio relative to a MV linkers/nominals portfolio. Around 2003, the RP portfolio had large spikes in vol relative to the more nominal concentrated MV linkers/nominal portfolio. This is because in 2003, the Australian TIB program was suspended (until 2009) due to the lack of fiscal funding needs. As we saw earlier, liquidity is an important factor to keep in mind for RP investors. Given the leverage factor applied in a RP portfolio, relative liquidity differential changes (between linkers and nominals) can cause a large draw-down. Canada: RP Canada linkers/Nominals portfolio versus MV Canada linkers/Nominals Similar to US and Australian RP Linkers-Nominals portfolio, we also backtested Canadian linkers versus their nominal counterparts. Canadian linkers make up about 14% of total coupon bonds outstanding. Here too, as can be seen in Figure 15 and Figure 16, a Canadian RP portfolio significantly outperforms MV portfolio while maintaining the same level of risk since 1998. Looking at the past year’s standard deviations, the RP portfolio recommends a 46% weight in Canadian linkers and 54% in Canadian nominals, with a leverage factor of 0.95, which means the left over 5% cash can be invested in cash or other instruments. Here too, the smaller size of the FIGURE 13 FIGURE 14 Australian RP (Linkers/Nominals) portfolio significantly Australian RP (Linkers/Nominals) portfolio maintains the outperforms Australian MV Linkers/Nominals portfolio same risk level as the MV (Linkers/Nominals) portfolio 350 10% 9% 300 8% 7% 250 6% 5% 200 4% 3% 150 2% 1% 100 0% Apr-99 Jan-02 Oct-04 Jul-07 Apr-10 Jan-13 Apr-99 Jan-02 Oct-04 Jul-07 Apr-10 Jan-13 AUS MV (Linkers-Nominal) Index AUS MV (Linkers-Nominal) Index vol (annualized) AUS Risk-Parity AUS Risk-Parity vol (annualized) Source: Barclays Research Source: Barclays Research 18 April 2013 17 Barclays | Global Rates Weekly FIGURE 15 FIGURE 16 Canadian RP (Linkers/Nominals) portfolio significantly Canadian RP (Linkers/Nominals) portfolio generally outperforms Canadian MV Linkers/Nominals portfolio maintains the same risk level as the MV (Linkers/Nominals) portfolio. 400 12% 350 10% 300 8% 250 6% 200 4% 150 2% 100 0% Dec-98 Sep-01 Jun-04 Mar-07 Dec-09 Sep-12 Dec-98 Sep-01 Jun-04 Mar-07 Dec-09 Sep-12 Canada MV (Linkers-Nominal) Index Canada MV (Linkers-Nominal) Index vol (annualized) Canada Risk-Parity Canada Risk-Parity vol (annualized) Source: Barclays Research Source: Barclays Research Canadian linkers market may prevent large investors from executing the RP methodology but to the extent US real yields move with Canadian real yields, large investors can construct a US TIPS-Canadian nominals proxy RP portfolio. Sweden: Swedish RP linkers-nominals portfolio outperforms MV portfolio The Swedish linkers market, in terms of market-value (MV), is about 27% of the central government debt. For Sweden, the RP linkers-nominals portfolio optimal allocation is also 55% in linkers and 45% in nominals with a leverage factor of 1.05 (5% additional cash borrowing). Similar to other sovereigns, here too, the risk-parity portfolio has outperformed over the past 15 years, though the large part of the performance came in the early years (+800bp, 1998-2008) and post the financial crisis (+300bp, 2009-2012). During the 2008 financial crisis, the Swedish RP portfolio underperformed by 250bp. FIGURE 17 FIGURE 18 Sweden (linkers-nominals) RP portfolio outperforms a Sweden Risk-Parity (linkers-nominals) portfolio maintains market-value portfolio the similar risk-level to the MV portfolio – ie, RP portfolio offers a higher Sharpe ratio 270 9% 250 8% 230 7% 210 6% 190 5% 170 4% 150 3% 130 2% 110 1% 90 0% Dec-98 Sep-01 Jun-04 Mar-07 Dec-09 Sep-12 Dec-98 Sep-01 Jun-04 Mar-07 Dec-09 Sep-12 Sweden MV (Linkers-Nominal) Sweden MV (Linkers-Nominal) vol (annualized) Sweden Risk-Parity Sweden Risk-Parity vol (annualized) Source: Barclays Research Source: Barclays Research 18 April 2013 18 Barclays | Global Rates Weekly UNITED STATES: AGENCIES A dam too far James Ma We believe potential TVA privatization is politically infeasible and would encounter +1 212 412 2563 fiscal and practical obstacles. As such, we recommend fading the budget-related firstname.lastname@example.org widening, but acknowledge an unclear near-term impetus for valuations to improve. Significant obstacles to privatizing TVA After the President’s 2014 Budget included an intent “to undertake a strategic review of options for addressing TVA’s financial situation, including…possible divesture…in part or as a whole,” the market for TVA paper has remained dislocated, as the name widened 10-20bp in the immediate aftermath (Figure 1). In the absence of further details on the strategic review, including a timeline and action plan, Moody’s and S&P maintained their ratings for TVA in line with the US government’s. However, the former warned that a multiple-notch downgrade was possible if the entity was sold in full to a private company, consistent with its previous analyses showing TVA’s financial ratios would imply a standalone rating of A2 (Figure 2). We believe there are enough significant obstacles to TVA privatization (political and otherwise) that investors should ultimately fade the paper’s widening to Treasuries and expect spreads to return to pre-budget levels. Of course, in the near term, the paper could remain at or near presently wide levels, potentially until it becomes apparent that further steps to privatizing TVA are not forthcoming or are politically infeasible. A toxic political environment Privatization would encounter In our view, while the administration has presented the idea of TVA privatization in the fierce political resistance and a budget, its fiscal motivations to pursue TVA reform may be stymied by political lack of bipartisan support considerations. Powerful Republicans representing TVA’s coverage area have expressed their reservations, including Senate Banking Committee (SBC) Ranking Member Shelby (R- AL) stating, “I don’t believe TVA’s going to be sold right now…that’s just a gesture.” Tennessee Senators Corker, an SBC member, and Alexander (on the Appropriations Committee) have concurred, with the latter describing it as “one more bad idea in a budget full” of them. TVA privatization has been repeatedly resisted, from as recently as the Clinton administration to as far back as helping end Barry Goldwater’s presidential aspirations. FIGURE 1 FIGURE 2 TVA dislocations along the curve TVA metrics comparisons Gov spd, bp CFO ex- CFO ex- Debt to WC Int WC to Book Cap 200 Coverage Debt 180 160 TVA 2.5x 7.6% 85.6% 140 A Alliant 5.6x 21.9% 40.6% 120 100 Southern 6.2x 21.7% 46.2% 80 PSE&G 6.2x 30.9% 39.3% 60 40 BBB Con Edison 5.2x 20.0% 44.4% 20 Dominion 4.6x 16.8% 54.4% 0 0 10 20 30 40 50 Duke Energy 4.6x 18.3% 41.2% Maturity, y TVA 4/17 TVA 4/10 FRE 4/17 Elec Util A 4/17 Source: Barclays Research Source: Moody’s 18 April 2013 19 Barclays | Global Rates Weekly To add to Congressional resistance, the legislative environment includes a lack of bipartisanship across all issues, from the budget and housing finance reform to gun control and the economy. In our view, TVA’s operational independence ranks it far down the list of priorities. As any changes to the TVA Act would require Congressional approval, we find progress on any privatization proposals extremely unlikely, especially if pushed beyond the next 18 months and into the following election cycle. Precedent may outweigh fiscal considerations Furthermore, we believe fiscal motivations for TVA privatization are more optical in nature. The budget cites TVA’s capex spending expectations of $25bn as conflicting with its Congressionally set $30bn limit on debt outstanding (of which it uses $25bn). Moreover, because TVA is fully US government owned, its obligations are included in “debt held by the public,” although they do not count towards the debt ceiling. At the same time, TVA is operationally independent, and the Treasury receives no part of cash flows from its operations, save for a periodic repayment of direct appropriations (roughly $7mn/y) as a return on $258mn of paid-in capital from pre-1999 activities. This is in direct contrast to the situation at FNM/FRE, which are not included in the public debt. Despite this, the Treasury owns a $180bn combined senior preferred equity stake in the two housing GSEs, and all but $3bn each from their net assets each quarter is swept to the Treasury beginning in 1Q13. Government protection of As such, we would not be surprised if the administration’s goal was simply to avoid having implicitly guaranteed bond to ask Congress to increase the $30bn debt cap. However, precedent suggests the holders undoes the fiscal benefit government would move to protect existing holders of TVA debt issued as a GSE: • When REFCORP’s wind-down ended more quickly than expected, its liabilities’ principal became fully defeased with Treasury securities, while the interest payments were paid by the FHLB system (and any shortfall was explicitly guaranteed by the Treasury). • Sallie Mae’s transition to a private company left it with explicitly guaranteed debt to 1982 and implicitly guaranteed debt issued during 1982-1996; all of this was either bought back or fully defeased, and a firewall cut off recourse of SLM Corp (the private entity) creditors to the GSE debt. Of course, Sallie Mae shares were privately held instead of 100% government owned, and these shareholders pushed for full privatization. FIGURE 3 FIGURE 4 Subsidized TVA service area boasts below-average rates… …as electricity costs remain a high percentage of incomes Average retail rate, $0.01/kWh Average electric bill per disposable income 2008 2009 2010 2011 2008 2009 2010 2011 New York (max) 18.3 17.5 18.7 18.3 Alabama (max) 5.2% 5.3% 5.7% 5.4% US Average 11.3 11.5 11.5 11.7 S. Carolina 4.8% 5.1% 5.6% 5.3% Alabama 10.4 10.7 10.7 11.1 Mississippi 5.4% 5.3% 5.6% 5.3% Georgia 9.9 10.1 10.1 11.1 Georgia 4.3% 4.5% 4.9% 4.9% Virginia 9.6 10.6 10.5 10.6 Tennessee 4.3% 4.4% 4.7% 4.7% NC 9.5 10.0 10.1 10.3 N. Carolina 4.0% 4.4% 4.8% 4.4% Mississippi 10.4 10.2 9.9 10.2 Kentucky 3.9% 4.0% 4.4% 4.2% Tennessee 8.9 9.3 9.2 10.0 Virginia 3.5% 3.9% 3.9% 3.7% Kentucky 7.9 8.4 8.6 9.2 US Average 3.4% 3.6% 3.7% 3.6% Idaho (min) 7.0 7.8 8.0 7.9 DC (min) 1.8% 1.9% 2.1% 1.8% Note: TVA services AL, GA, KY, MS, NC, SC, and TN. Source: BEA, Moody’s Note: TVA services AL, GA, KY, MS, NC, SC, and TN. Source: BEA, Moody’s 18 April 2013 20 Barclays | Global Rates Weekly In either case, investors in implicitly guaranteed debt were eventually protected by the government in various ways, a solution echoed in the present FNM/FRE conservatorships. If defeasance were the method of choice for TVA, government spending and the size of the public debt would remain unchanged because of the additional Treasuries sales needed. On the other hand, the government could reduce its ownership of TVA to the same 79.9% stake it holds in FNM/FRE and skirt concerns over the public debt, but would be unlikely to receive a price that would be accretive to taxpayers. In exchange for TVA’s effective monopoly over its geographic operating area, it adheres to the Congressional mandate that “power shall be sold at rates as low as feasible” and, as such, has kept its rates below the national average for its entire service area (Figure 3). Practical considerations may hinder acquisition Similarly, TVA sells its power to distributors (mainly regional co-ops) under contracts that require 5-15y notice before termination, and it is unclear if the contracts could be renegotiated in the event of a full divesture. While residents of TVA’s service area receive a clear subsidy, electricity is still a disproportionately high percentage of their disposable income relative to the national average (Figure 4). Lastly, because of its mission and government ownership, TVA’s leverage ratios and other financial metrics suffer compared with publicly traded utilities within its A rated peer group. These would presumably worsen if funding costs rose to become more in line with that peer group. Any potential acquirers would have to address both issues while continuing to meet the 8-12% ROE targets common in the industry. Thus, we believe the TVA strategic review indicated in the president’s budget should not be interpreted as an imminent move towards privatization. The budget itself provides only the barest of details and no timeline, and any plan that emerged would likely encounter considerable, consistent political resistance. Expect TVA eventually to As such, we would fade related cheapening of TVA paper to pre-budget levels, which return to pre-budget spread would provide 10-20bp of upside at roughly T+70bp from the 15y sector on out the curve. levels We acknowledge that the market could remain dislocated until further details emerge or the plan’s political infeasibility becomes apparent. FIGURE 5 FIGURE 6 Central bank demand for 5y agency bellwethers increases 5-7y sector shows lingering signs of indigestion 100% A-T spd, bp 90% 50 80% 45 70% 40 60% 35 30 50% 25 40% 20 30% 15 20% 10 10% 5 0% 0 2008 2009 2010 2011 2012 2013 FN 5y 2y 3y 5y 7y 10y 20y Fund Mgr Comm Bk Ctrl Bk Other 4/3/13 4/10/13 4/17/13 Source: Barclays Research Source: Barclays Research 18 April 2013 21 Barclays | Global Rates Weekly Relative value update Primary activity remained continued in the agency bellwether market last week, with Fannie Mae placing a new 5y (5/18s, $5.5bn). Intermediate spreads seemed to reflect some indigestion of the new issue, despite high (foreign) central bank demand (Figure 5). We continue to recommend owning 5y agencies at T+19bp matched-date and 7y at T+27bp, which appear cheap to surrounding sectors and offer considerable spread roll-down (Figure 6). 18 April 2013 22 Barclays | Global Rates Weekly UNITED STATES: VOLATILITY For a flatter tomorrow Piyush Goyal Buy 3m*(3y1y) – 3m*30y bull flattener to hedge a further significant decline in yields. +1 212 412 6793 The trade seeks to benefit from attractive rate and vol entry levels. email@example.com Given the ongoing occurrences in the commodity market and the uncertain political landscape in Italy, we like hedging a large decline in yields. However, rates are already quite low; as a result, outright receivers do not offer attractive risk-reward. To come to this conclusion, we compared a few structures across the yield curve (Figure 1). As shown in the column labelled p&l (bp, carry), most of the receivers do not re-coup the cost of the option in a carry scenario and some are barely profitable even if yields decline to the low levels seen mid last year. For example, a 3m*7y ATM receiver would cost about 11bp (in underlying swap01) and would recoup only 9bp in a carry scenario, implying a loss. Further, if 7y swaps decline to their lowest level ever of 1.14%, the receiver would generate only 2.55 times the premium outlay. The main reason outright receivers do not offer attractive risk-reward is the high implied vol. Figure 2 plots the implied to realized vol ratio across tails on the gamma surface. Despite the recent decline in implied vol, the implied to realized vol ratio is quite high, especially in 5y and 10y tails. Essentially, option investors remain wary of a change in Fed stance, and keep implied vol from coming off to levels where owning options to position for a drift in rates (lower or higher) would be profitable. Admittedly, as highlighted by Figure 1, some structures are better than others to position for a decline in yields. 3m*30y receivers offer Across tails, in general, options on longer rates offer better risk-reward. Within vanilla the best risk-reward options, 3m*30y will recoup four times the premium outlay if the lows of July 2012 are for a decline in yields achieved. On the other hand, a 3m*2y receiver would barely profit if yields rally significantly. Similarly within mid-curve options, 3m*(5y5y) will perform best. Essentially, while yields have come down, the yield curve is still quite steep. FIGURE 1 3m*30y receiver offers the best risk-reward if rates fall to record lows; even better 3m*5y5y receiver spot fwd vol cost ATM 3m Carry p&l (bp, lowest rate p&l if lowest p&l if lowest rate / vanilla options (3m) rate rate (bp/y) receiver (bp) (bp) carry) level rate cost 2y 0.37% 0.40% 20 4 3 (1) 0.35% 5 1.24 5y 0.86% 0.95% 42 8 9 1 0.74% 21 2.57 7y 1.31% 1.41% 53 11 9 (1) 1.14% 27 2.55 10y 1.85% 1.93% 62 12 8 (4) 1.55% 39 3.12 30y 2.78% 2.82% 70 14 3 (11) 2.26% 56 4.01 mid-curve options (3m) 2y1y 0.68% 0.78% 40 8 10 2 0.56% 22 2.74 3y1y 1.16% 1.30% 65 13 14 1 0.94% 36 2.76 2y2y 0.92% 1.04% 51 10 12 2 0.75% 29 2.81 5y5y 2.94% 3.01% 90 18 7 (11) 2.38% 63 3.50 Note: As of 4/18/13. Source: Barclays Research 18 April 2013 23 Barclays | Global Rates Weekly Across vanilla and mid-curve options, 3m*30y is better than 3m*5y5y to own protection against a large decline in yields. It is noteworthy that 5y5y has more room to decline compared to a 30y swap rate (63bp vs. 55bp); however, due to much higher vol, a 3m*(5y5y) receiver offers less attractive risk-reward. Mid-curve options are more The relatively higher implied vol for mid-curves is probably driven by demand for these expensive than vanilla options as investors move further out on the yield curve for Fed-on-hold or hawkish-Fed swaptions trades. As a result, in general, the entire mid-curve sector is at a higher premium to realized vol compared to vanilla swaptions. Figure 2 shows the implied vol of mid-curve options on RED, GREEN, BLUE and GOLD eurodollar futures. While 30y tail vanilla options are at a discount to delivered vol, the mid-curve options are expensive across the underlying rates. Put together, a steep yield curve and relatively expensive mid-curve vol bring us to the following bull flattener: • Long $100mn 3m*30y low-strike receiver • Short $2bn2 3m*3y1y ATM receiver Buy 3m*30y low-strike For historical context, Figure 3 compares the 30y swap rate with 3y1y forward swap rate receivers funded with 3m*3y1y since early last year. The key takeaway from the chart is the two rates decoupled in mid-curve receivers September last year, probably on the back of Fed’s open-ended QE. Due to the decoupling, 30y rates have more room to rally than 3y1y rates. In other words, a large rally, maybe due to a risk-flare in the eurozone would likely be led by the 30y rate. Accordingly, 3m*30y vol should be much higher than 3m*3y1y vol. However, current pricing suggests the former is only at a slight premium to mid-curve option vol. Thus, we like the bull flattener. The chief risk to the trade is if the Fed is able to communicate the possibility of Fed-on-hold beyond what is currently priced. A new dovish tool might cause the 3y1y rate to fall more and 30y rates to lag the decline as inflation expectations rise. This would mean a bull steepening of the curve, hurting the above structure. While possible, the Fed came up with its latest tool of “data-base guidance” only four months ago and may wait awhile before pumping more monetary stimulus. As the above structure is short-dated, it would have expired in a few months and would be able to avoid such scenario. FIGURE 2 FIGURE 3 Implied vol is higher than delivered; more in mid-tails, more 3y1y rates decoupled from 30y swaps last September, have in mid-curve options little room to rally Imp Vol 2y 5y 10y 30y Imp Vol 1st 2nd 3rd 4th 2.1 3.2 1m 19 40 61 70 RED 17 20 28 30 3.1 3m 20 42 62 70 GREEN 38 39 47 53 1.9 3.0 6m 23 48 68 73 BLUE 59 68 74 78 2.9 1y 32 57 74 76 GOLDS 76 85 90 91 1.7 2.8 20d rlzd 20d rlzd 1.5 2.7 2y 5y 10y 30y 1st 2nd 3rd 4th vol vol 2.6 1m 17 33 57 78 RED 13 15 18 24 1.3 2.5 3m 18 34 58 79 GREEN 29 36 40 47 2.4 6m 18 38 61 80 BLUE 52 57 62 67 1.1 1y 22 47 66 83 GOLDS 71 74 77 78 2.3 0.9 2.2 Imp/rlz Jan-12 Apr-12 Jul-12 Oct-12 Jan-13 Apr-13 2y 5y 10y 30y Imp/rlzd 1st 2nd 3rd 4th d 3y1y (L) 30y (R) 1m 1.11 1.22 1.06 0.90 RED 1.31 1.34 1.52 1.28 3m 1.15 1.22 1.07 0.89 GREEN 1.32 1.10 1.19 1.13 6m 1.26 1.26 1.11 0.90 BLUE 1.12 1.19 1.20 1.17 1y 1.42 1.23 1.12 0.92 GOLDS 1.07 1.15 1.17 1.16 Source: Barclays. As of 4/17/13 Source: Barclays. Data period: 1/2/12 – 4/18/13 2 3y1y has a dv01 of 1cts vs. 20cts for 30y swaps, implying a hedge ratio of 20 18 April 2013 24 Barclays | Global Rates Weekly Change to exchange The trade could also be initiated via exchange future options, in the following form: • Long $100mn 6/14/13 – 30y receiver low-strike • Short 4000 3EM3C 99 call • Short 4000 4EM3 98.375 call The expiry on the 30y swaption receiver is changed to June 14, 2013 to match the expiry of the exchange options. More importantly, the reason a mid-curve option on the 3y1y rate can be replicated with the exchange option on EDH7 and EDH8 is the correlation and beta between the 3y1y rate and average of 3y3m, 4y3m is ~ 1. So, a long position in the 3y1y option is the same as a long position in the two options on the exchange. 18 April 2013 25 Barclays | Global Rates Weekly UNITED STATES: MONEY MARKETS Clearing the fog in interbank lending Joseph Abate There is surprisingly little data on tenors, volumes or borrowing rates in the interbank +1 212 412 7459 loan market. Some of this information can be imperfectly inferred from other markets. firstname.lastname@example.org But data from Fedwire might provide more clarity. • Detailed data on term interbank lending activity is not available. Even information on the overnight fed funds market is sparse. • Some sense of overall interbank activity can be inferred from the commercial paper market, but the sample size and issuance is limited. • An algorithm designed to match, send, and receive payment instructions over Fedwire might capture broader detail on interbank loan terms, rates, and volumes. In real time, these data might be used to confirm the transacted rate figures derived from money fund holdings and the CP market. Filling in the gaps Information on interbank Despite intense scrutiny over the past five years, information on the interbank lending lending is surprisingly thin market is surprisingly sparse. The Federal Reserve collects weekly data on the total amount of interbank loans reported as assets by all US commercial banks. Fed funds and reverse repos make up nearly all of the roughly $130bn outstanding, but it is unclear what the distribution is between them, although our sense based on observations of the brokered fed funds market suggests most of this total is in the collateral market.3 Similarly, the Fed collects data on borrowing reported by commercial banks from other banks in the US (a liability). It is unclear what form this borrowing takes – for instance, commercial paper or perhaps a brokered wholesale deposit. Like the asset holdings, this, too, has fallen sharply since the financial crisis, from about $400bn to about $160bn (Figure 1). Beyond the outstanding amounts, however, neither series provides clues as to the maturity distribution of these loans or their underlying rates. FIGURE 1 FIGURE 2 Bank borrowings from US banks ($bn) Money fund financial CP holdings (% total fin CP) 600 46 45 500 44 43 400 42 41 300 40 39 200 38 100 37 36 0 35 Jan-00 Jan-03 Jan-06 Jan-09 Jan-12 Oct-11 Jan-12 Apr-12 Jul-12 Oct-12 Jan-13 Source: Federal Reserve Source: imoney.net, Federal Reserve 3 For a recent review of activity levels in the fed funds market, see, “An Empirical Study of Trade Dynamics in the Fed Funds Market”, G. Afonso, and R. Lagos, Federal Reserve Bank of New York, February 2012. 18 April 2013 26 Barclays | Global Rates Weekly To fill in the gaps, analysts have turned to other data sources, including commercial paper and money market fund holdings. Although commercial paper rates are close to unsecured term interbank loans rates, they are not exact matches, as CP is typically sold to investors outside the commercial banking sector. Indeed, except for their own acceptances, commercial banks hold no CP. Instead, a far larger share of bank CP is held by money market funds. At the end of March, financial CP outstanding totalled $550bn, and money fund holdings across all bank issuers were $230bn, or 42% of the market (Figure 2). But while money fund holdings are extremely transparent, the aggregate figures are only available once a month. Transacted rates CP rates may be a bit Given the nearly $600bn in outstanding financial commercial paper of all maturities, it misrepresentative should be fairly reasonable to use published commercial paper rates from the Federal Reserve as a proxy for interbank funding rates. But for a variety of reasons, these rates may misrepresent true interbank lending rates a bit. First, the rates are very noisy since they reflect changing samples of issuers each day; after all, the same banks do not consistently issue 3m CP every day (Figure 3). More significantly, the daily volume of newly issued CP is small. Financial paper issuance across all maturities has averaged just $1.25bn per day this year, down from an already low $1.5bn/day in 2012. And the pace of issuance for paper with more than 80d to maturity has barely exceeded $0.7bn this year (Figure 4). This thinness is compounded by the fact that the Fed publishes data only for AA issuers, even though money market funds and other investors have modestly expanded their purchases to include slightly lower-rated financial issuers. As a result, published CP rates from the Federal Reserve may not be representative of overall interbank rates, which cover a wider array of borrowers across deeper issuance. But there may be another approach. Round-tripping Much of interbank lending Fedwire is a real-time gross settlement system for large dollar value wholesale payments occurs over Fedwire between financial institutions (including the GSEs) in the US. The US branches of foreign banks are also participants.4 Given that much of the interbank lending done by banks in the US occurs over Fedwire, it should theoretically be possible to capture a large enough sample from the Fedwire payment instructions to gather a precise sense of the volume, tenor, and rate on interbank borrowing. With this in mind, the Federal Reserve has developed a modified algorithm designed to sift through the millions of daily Fedwire instructions to look FIGURE 3 FIGURE 4 Financial CP rate, 3m (bp) Daily financial CP issuance ($mm) 32 5,000 30 4,500 28 4,000 3,500 26 3,000 24 2,500 22 2,000 20 1,500 18 1,000 16 500 14 0 Jan-12 Apr-12 Jul-12 Oct-12 Jan-13 Apr-13 12 Feb-11 Jun-11 Oct-11 Feb-12 Jun-12 Oct-12 Feb-13 >80d Total Source: Federal Reserve Source: Federal Reserve 4 Fedwire is not the only payment settlement mechanism in the US; CHIPS is the other major network. 18 April 2013 27 Barclays | Global Rates Weekly for matching sets of interbank loans and their repayments.5 Essentially, the algorithm matches each set of “send payment” instructions with a future “return payment” between bank A and bank B.6 The send payment is limited to transactions in fixed increments above a fixed amount (say, $10mn) designed to capture most wholesale interbank lending. It looks for a return payment that is slightly larger than the size of the candidate send payment, by an amount equal to the interest it might have earned at an interest rate in whole basis points within, say, 100bp of the prevailing Libor rate. Kuo et al. do not attempt to identify the type of transaction, although the algorithm has been applied specifically to overnight transactions in an attempt to gauge fed funds market activity, although with disappointing results.7 Does the algorithm capture Of course, the algorithm is useful only if it captures genuine interbank lending transactions. interbank loans well? Indeed, given the huge volume of daily transactions, it is possible that unrelated matching send and receive instructions could be identified as interbank loans. Misidentification, as either a false positive (labelling a random transaction as a bank loan) or a false negative (missing interbank loans entirely) is one problem with the algorithm. Both Armantier and Copeland and Kuo et al. benchmark the output of their algorithms to known data on interbank lending from other sources. Armantier and Copeland find very large misidentification errors – particularly false positives – in analyzing overnight transactions. That is, the algorithm identifies more ‘fed funds’ trades than the true volume. But Kuo et al. seem to find fewer of these types of errors, perhaps because they are mainly interested in any term unsecured bank loan between banks A and B. Instead, Kuo et al. find that the algorithm captures a fairly large (75%) portion of actual brokered interbank lending transactions in 2000-04. There are two additional types of errors. First, the algorithm is applied just to the Fedwire data; to the extent that any interbank loans settle on CHIPS the output of the data sifting will be too small. More significantly, the algorithm cannot identify the final borrower in a transaction. Thus, in a correspondent transaction, while the payment may round-trip between Bank A and B, the ultimate borrower may in fact be Bank C, or more discouragingly, Company C, which could have a different risk profile than Bank B. It is unclear if the inability to capture intermediary transactions biases the implied rate higher or lower. Nevertheless, we estimate that omitted error bias and the difficulty in capturing intermediary transactions accounts for much of the 25% of loan transactions missed by the algorithm. The algorithm does well in In their analysis of pre-crisis term unsecured interbank markets, Kuo et al. note that the capturing interbank lending implied rates from the algorithm were clustered around the prevailing Libor rates. They trends before and after the crisis interpret this to indicate that the algorithm is not capturing random pairs of Fedwire transactions but actually tracking interbank lending. Moreover, they observe that the estimated daily volume of transactions is $90-150bn (about the size of the market as reported by the Fed) and matches the observed behavior of other lending markets such as CP in terms of post-crisis tenor shrinkage and volume declines. Real time The data might be best used Obviously, trying to match transactions instructions real time is impossible: we can monitor as confirmation of trends in CP the send legs but will not know the return legs until, say three months, later. In the and money fund holdings meanwhile, it will still be useful to examine the money fund figures, as well as the daily interest rates from the commercial paper market. Only in hindsight will the algorithm 5 See, “Identifying Term Interbank Loans from Fedwire Payments Data”, D. Kuo, D. Skeie, J. Vickery, and T. Youle, Federal Reserve Bank of New York, March 2013 and the original algorithm in “The Microstructure of the Fed Funds Market,” Financial Markets, Institutions, and Instruments, 1999. 6 We ignore the technical issue of what to do when a send instruction matches several potential return instructions. In these situations, some kind of “tiebreaker” rule needs to be applied. These decisions are fairly common, probably because of the standardization or clumping in transaction sizes. 7 See, “Assessing the Quality of Furfine-based Algorithms”, O. Armantier and A. Copeland, Federal Reserve Bank of New York, October 2012. 18 April 2013 28 Barclays | Global Rates Weekly provide more clarity on interbank lending activity in terms of the distribution of rates and tenors. Nevertheless, we think that making this data available will improve the market’s sense of the overall level of activity over the current mix of money fund, CP and bank asset/liability data, even if they might be stale. 18 April 2013 29 Barclays | Global Rates Weekly EURO AREA: RATES STRATEGY Resiliency Laurent Fransolet This is an edited extract from the Global Macro Daily, 18 April 2013 +44 (0)20 7773 8385 The resiliency of peripheral markets recently has been impressive, with yields/spreads to email@example.com Germany near YTD lows. The IMF Fiscal Monitor and Global Financial Stability Reports (FM and GFSR) include analysis that help explain this, and other moves in financial markets. The frontloading of the budget Figure 1 (Figure 10 of the FM) shows the 10 advanced economies that require the biggest adjustments helps to explain changes in cyclically adjusted primary balance. The right-hand axis shows how much of the some of the resilience of total adjustment is being frontloaded (in 2011-13) vs what remains to be done in 2014-20. peripheral markets Italy (and Greece) fare well on that measure (with about 70% of the primary balance adjustment done), with other euro countries (Spain, France, Belgium) around the 50% mark. In our view, this fundamental improvement, also apparent on the current account front, is one of the reasons why these markets have been quite resilient in the past few quarters, even as headline news on the eurozone has picked up (eg, Cyprus, Italian elections, etc). FIGURE 1 Advanced economies with largest adjustment requirements in the cyclically adjusted primary balance (% of GDP) 18% 2011–13 2014–20 % frontloaded (RHS) 80% 16% 70% 14% 60% 12% 50% 10% 40% 8% 30% 6% 4% 20% 2% 10% 0% 0% -2% -10% France Greece Portugal Japan Spain Ireland Belgium Italy United States United Kingdom Source: IMF Italy is in the best position on The chart also illustrates the relatively better position of Italy compared with Spain in terms that metric, in both absolute of progress and the absolute size of the adjustment that is needed. Indeed, despite the and relative terms political uncertainties and the slower progress on the structural reform side, international investors have been, and will likely remain, much more comfortable with the Italian government debt market than with the Spanish. While Spain has tightened vs Italy in recent months, it is still higher in yields, and anecdotal evidence suggests that the search for yield (from displaced ‘safe-haven’ investors, including Japanese ones) will benefit Italy more than Spain in the near to medium term. In contrast, Portugal (and, to a lesser extent, Ireland) is lagging on these metrics, and international investors will likely remain much more cautious than in Spain and Italy until there is a clear turnaround in the economy (even if the data show some small net foreign buying since September 2012). Large developed countries The other large developed countries (the US, the UK, Japan) have a lot of fiscal adjustment to do, have a lot more to do, but have but the bond markets there are (and should remain) fundamentally more stable than they were in more stable investor bases the euro area periphery. Hence, the necessity for a large and fast adjustment is less urgent, and the risk of a sudden stop ‘demand’ induced sell-off is inherently much lower, even without the prospects of QE buying. This can be seen in Figure 2 (Box 3 of the FM), which shows the (pre- crisis) proportion of ‘stable’ ownership of each government bond market (defined as domestic 18 April 2013 30 Barclays | Global Rates Weekly and foreign central banks’ ownership, as well as domestic non-bank financial institutions’ ownership) and the volatility these markets experienced in the past few years. Clearly, Japan, the US and the UK benefitted (and continue to benefit) from much more ‘stable’ investor bases than most euro countries and vs peripheral countries’. The IMF notes, though, as we have frequently highlighted, that foreign ownership of peripheral markets has stabilized, if not slightly increased, recently (eg, Figure 1.12 of the GFSR, updated in Figure 3 below), and this has been a key driver of the resiliency of peripheral markets, even at times of risk-off sentiment. FIGURE 2 Real-money investors before the crisis, 2007 Central banks, national and foreign Domestic nonbank financial institutions 80% 16 Bond yield volatility (RHS, bp) 70% 14 60% 12 50% 10 40% 8 30% 6 20% 4 10% 2 0% 0 GRC PRT IRL ITA BEL ESP FRA DEU SWE CAN JPN USA UK Source: IMF Some technical factors have Several more short-term technical factors also help to explain this resiliency. First, Italy has just helped peripheral markets raised €17bn of a retail BTP Italia, highlighting the strength of domestic demand (and almost as well enough to fully cover the increased issuance needs due to the decision to repay arrears to SMEs, a positive for the economy). Second, as we have highlighted before, leveraged investors still have short positions in peripheral debt markets, and the recent moves in equities and commodities might have induced some knock-on short covering. We stay positive on peripheral We stay positive on peripheral bond risk, with a focus on the short ends and 5y sectors in bond markets Italy and Spain; the forward spreads embedded in the 3-5 year part of the spot curve are still elevated and offer attractive yields compared with the safe-haven markets, as illustrated in Figure 4 (Figure 1.13 of the GFSR). FIGURE 3 FIGURE 4 Foreign ownership of Spanish and Italian government debt 2y bond spreads vs Germany (bp) 70% 800 60% 700 600 50% Forwards 500 40% 400 30% 300 200 20% 100 10% 0 0% -100 Jun09 Jan10 Jul10 Jan11 Jul11 Jan12 Jul12 Jan13 -200 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Spanish debt securities held by foreigners (ex ECB) % of BTPs owned externally (ex ECB) France Italy Spain Belgium Source: Italian and Spanish Treasuries, Barclays Research Source: IMF 18 April 2013 31 Barclays | Global Rates Weekly EUROPE: SOVEREIGN SPREADS Political rally in Italy Huw Worthington Intra-day volatility of Italian and Spanish yields has decreased recently such that the +44 (0)20 7773 1307 yields are approaching December’s very low levels. Spreads appear a little cheap, firstname.lastname@example.org particularly in 10y. Politics in Italy are likely to remain a driver of spreads and have the potential to drive spreads and yields a little lower. Cagdas Aksu Notwithstanding a small re-widening correction in the past week (likely supply related), the +44 (0)20 7773 5788 period since the end of March has been characterised by a re-tightening of EGB spreads in email@example.com highly rated core and peripheral issuers. A major catalyst in the move was the BoJ announcement a fortnight ago that in particular supported the highly rated core. France was the major beneficiary, leading the move tighter, with outperformance of 19bp in 10y bonds, followed closely by Belgium at c.17bp tighter again in 10y. In the periphery, Italian bonds have tightened across the curve, with 47bp of outperformance in the 10y area and 55bp in the front end. Spanish bonds managed similar figures. The recent moves in Spain and Italy have also been accompanied by a further decrease in the underlying intra-day volatility. Figures 1 to 2 chart 10-day moving average intra-day yield moves in 5y and 10y Spanish and Italian bonds since the start of 2011 against the yield spreads in these maturities versus bunds. Intra-day volatility has In Spain and Italy, 5y and 10y intra-day volatility has now fallen to the extent that with the fallen recently 10-day moving average is in single digits for the first time since the end of December 2012 when lows of 5-9bp in both maturities were registered, markets where at their quietest and supply was much less of a factor than now. Previous episodes have had As can be seen in the charts, past reductions and increases in intra-day volatility can be a much higher volatility in good indication of improved underlying liquidity, and thus sentiment, in EGB markets. As intra-day yields such, other things being equal, peripheral EGB spreads appear to be on the cheap side, particularly in Spain and in Italy in the 10y area. FIGURE 1 FIGURE 2 10-day moving average SPGB yield changes vs 10y SPGB 10-day moving average BTP yield changes vs 10y BTP Bund Bund spread spread 40 10y Mvg Avg yield volatility 800 45 10y Mvg Avg yield volatility 700 10y SPGB Bund Spread (RHS) 10y BTP Bund Spread (RHS) 35 700 40 600 35 30 600 500 30 25 500 25 400 20 400 20 300 15 300 15 200 10 200 10 5 100 100 5 0 0 0 0 Apr-10 Nov-10 Jun-11 Jan-12 Aug-12 Mar-13 Sep-10 Jan-11May-11Sep-11 Jan-12May-12Sep-12 Jan-13May-13 Source: Bloomberg, Barclays Research Source: Bloomberg, Barclays Research 18 April 2013 32 Barclays | Global Rates Weekly Politic and, supply dynamics The better environment in general, particularly following the BoJ announcement, has been a key driver of the lower volatility. However, several other factors seem likely to drive spreads, not least of which will be political developments in Italy. The election of a new president is a pre-requisite to forming a new administration after almost two months of impasse. In this regard, the decision by both centre-left leader Pier Luigi Bersani and centre-right leader Silvio Berlusconi to support former Senate Speaker Franco Marini as a candidate is potentially positive because it could lead to the swift formation of a new grand coalition administration with electoral law reform high on its agenda. Whether Marini could secure the two-thirds majority required in the early voting rounds remains to be seen as the centre- left PD party is split over his candidature; whether Marini would be proceed without this level of support also remains unclear. Nevertheless, a swift decision on a new president, with a new administration being formed soon thereafter, would, in our opinion, be taken positively by markets, regardless of whether the new government may be inherently unstable and, in reality, only delay elections to 2014 as opposed to later this summer. Prior to the elections in February, we felt that a clear Bersani-Monti majority in both houses could see 10y Italy rally towards 4%; at the time of writing, 10y BTPs yield 4.25%. While a relatively unstable grand coalition may not be as enticing as a stable centre-left government, 10y BTPs could still move lower from current levels on the news. In reality, investment flows into Italy from ex-euro area foreign investors (and the periphery in general) in the wake of the BoJ announcement may take longer to come to fruition. However, any political stability may mean the mere anticipation of such flows, particularly in shorter-dated bonds and bills, should provide support in Italy. Politics aside, a number of other factors may also help explain the resiliency of peripheral markets, and Italy in particular. On the supply front, this week Italy issued €17bn of a retail 4y BTP Italia, highlighting the underlying strength of domestic demand for Italian debt even at a time of political uncertainty. However, the issuance will not result in cancelled or reduced auction sizes because, at the same time, the head of the Italian Treasury, Maria Cannata, announced an increase in total gross funding requirements for this year to €450bn from €415bn previously. The revision reflects the repayment of arrears designed to act as a much- needed stimulus to the economy (€20bn this year likely in late summer and €20bn next year; see Italy: Set to liquidate EUR40bn accounts payable – Potential effects on public finances, growth and debt management, 8 April 2013), and takes into account fiscal slippage (see Italy: Risks of fiscal slippage loom this year, 12 April 2013). As such, we now expect gross FIGURE 3 FIGURE 4 Evolution of BTP forward yields vs Germany Selected core yield spreads vs Germany 1y Spain vs Germany 30 France 2y 100 Average 2 to 5y Spain fwd vs Germany Belgium 2y 8 28 95 1y Italy vs Germany France 10y (RHS) 7 Average 2 to 5y Italy fwd vs Germany 26 Belgium 10y (RHS) 90 24 85 6 22 80 5 20 75 4 18 70 3 16 65 2 14 60 1 BoJ Announcement 12 55 0 10 50 Jan-10 Jul-10 Jan-11 Jul-11 Jan-12 Jul-12 Jan-13 Jul-13 Jan-13 Feb-13 Mar-13 Source: Bloomberg, Barclays Research Source: Bloomberg, Barclays Research 18 April 2013 33 Barclays | Global Rates Weekly bond issuance of c.€215bn (net c.€57bn), with net t-bills issuance of c.€4bn (previously -€11bn). However, the frontloading of supply already this year (Italy is currently 41% funded for 2013 on the higher issuance target), the success of the BTP Italia (with another expected in the autumn) and higher CCT-eu supply mean that BTP supply will not necessarily need to increase from the levels seen in the scheduled auctions carried out thus far in 2013. As such, supply pressure on the BTP nominal curve can remain contained, in our opinion. FIGURE 5 EGB issuance progress YTD % of 2013 Actual Estimated All Net issuance estimated gross Difference funding - funding - Up to 18/04/13 2-3 5 7-10 15+ maturities YTD €bn issuance vs 2012 2012 2013 Germany 24.0 17.0 16.0 2.0 59.0 13.0 32% 0% 183 184 France (inc buybacks) 13.0 24.0 27.7 13.7 78.4 52.5 41% 8% 202 190 Italy 25.4 37.5 12.9 12.2 88.0 46.1 41% 13% 232 215 Spain 16.7 11.0 21.9 1.9 51.5 30.1 43% 1% 98 121 Belgium 0.0 6.2 5.3 1.2 12.7 -0.6 34% 4% 44 37 Holland 7.8 5.2 6.5 3.5 23.0 6.7 46% 2% 65 50 Portugal 0.0 2.5 0.0 0.0 2.5 2.4 36% NA 0 7 Finland 0.0 0.0 4.0 1.0 5.0 5.0 38% 12% 13 13 Austria 0.0 0.6 5.0 2.8 8.4 8.8 36% -6% 21 23 Greece 0.0 0.0 0.0 0.0 0.0 NA NA NA 0 0 Ireland 0.0 2.5 5.0 0.0 7.5 7.5 75% 75% 4 10 Eurozone Aggregates 86.8 106.5 104.2 38.4 336.0 171.5 40% 6% 862 850 Percentages 26% 32% 31% 11% Difference vs. 2012 -7% 6% 0% 1% Source: Barclays Research Next week’s cash flows Next week should be quiet in terms of scheduled supply, with Italy conducting a linker auction linker on Wednesday, while Germany taps its 30y bond for €2bn the same day. Support for the market will come in the form of €16.2bn of coupon and €21.7bn of redemption flows, primarily in France. FIGURE 6 Barclays cash flow expectations for week beginning 22 April 2013 Beginning Auction Date Issuance Redemptions Coupons Net Cash Flow 08-Apr 0.45 Germany 2.00 0.00 0.00 2.00 Weekly 15-Apr -22.85 France 0.00 21.73 15.62 -37.34 Net 22-Apr -32.46 Italy 3.50 0.00 0.23 3.27 Cash flow 29-Apr -15.45 Spain 0.00 0.00 0.00 0.00 06-May 10.52 Belgium 0.00 0.00 0.01 -0.01 Greece 0.00 0.00 0.08 -0.08 Net Cash Flow is issuance minus redemptions minus Finland 0.00 0.00 0.00 0.00 coupons. Negative number implies cash returned to Ireland 0.00 0.00 0.00 0.00 the market. Holland 0.00 0.00 0.00 0.00 Austria 0.00 0.00 0.29 -0.29 Total issuance 5.50 Portugal 0.00 0.00 0.00 0.00 Total redemptions 21.73 Total 5.50 21.73 16.235 -32.46 Total coupons 16.24 Net cash flow -32.46 Source: Barclays Research 18 April 2013 34 Barclays | Global Rates Weekly EUROPE: MONEY MARKETS The ECB’s options Giuseppe Maraffino Increasing 3y LTRO repayments has raised the risk of passive tightening for liquidity +44 (0)20 3134 9938 conditions. The ECB has several options to counteract any rise in short rates. The firstname.lastname@example.org current bull flattening of the curve is likely to continue. Since mid-March the pace of 3y LTROs reimbursement has increased. With last week’s Laurent Fransolet €10.8bn repayment, the total is now about €264bn (about 53% of the total net new liquidity +44 (0)20 7773 8385 injected by the two 3y LTROs). We noted also an increase in the average repayment, email@example.com suggesting that at present the payback is mainly from the biggest banks (which borrowed from the domestic central bank or via central banks in other eurozone countries) after the initial phase (in January-February) which was characterized by a larger participation of small Increase in the 3y banks. This is an important aspect to take into account: as we highlighted “The 3y LTROs’ LTRO repayments “big payback”: a bottom-up approach”, 23 January 2013, the borrowing at the ECB’ 3y LTROs is very concentrated: the top 20 banks have borrowed on average €30bn each, giving a total of €600bn. Therefore, the decision by a few banks to repay or not can have a large impact on the total. Figure 1 shows our estimates for the breakdown of repayments by country on a monthly basis up to the end of March. FIGURE 1 3y LTROs payback: estimated country breakdown (€bn) Borrowing at other ECB's Payback - end Payback - end Payback - end of Est. 3y LTRO Payback - % tot. 3y LTRO operations of January of February March borrowing borrowing Remaining (end of March) MRO LTROs DE 21 45 46 69 67% 23 1 0 FR 52 57 59 165 36% 106 10 7 IR 0 6 15 60 25% 45 9 1 SP 44 58 68 305 22% 237 21 8 GR 0 0 0 0 0% 0 75 0 PT 4 5 5 50 10% 45 3 1 BE 11 21 24 40 59% 17 0 1 IT 0 0 0 255 0% 255 6 7 HO 0 10 10 20 50% 10 0 0 Lux 3 3 3 5 50% 3 0 0 AT 3 8 10 17 56% 8 0 0 Tot 138 213 238 981 748 125 25 Note: Breakdown figures are based on the monthly changes of the borrowing in the period January-March 2013, Source: National Central Banks. Barclays Research We note the following factors. 1) For peripheral countries, there is still no evidence of repayment from Italian banks. While they are currently in a good position both in terms of liquidity and eligible collateral, their choice of not repaying yet is probably related to political uncertainty and more medium- term plans for a gradual exit. Spanish banks continued to repay in February and March after 18 April 2013 35 Barclays | Global Rates Weekly the first big repayment of €44bn on January 30 (mainly from major banks)8. Probably also medium and small Spanish banks have started to reimburse part of their 3y liquidity, given the improvement in access to the interbank funding market9. In Portugal, after the initial repayment by larger banks, Portuguese banks have not repaid any part of their 3y borrowing in February and March. In Ireland, banks have repaid a total of €15bn, this likely has come mainly from foreign banks that borrowed via the Irish Central bank. 2) Regarding the core countries, the pace of repayment by French banks slowed down in February and March, after the initial big repayment of €52bn on January 30. This is consistent with the plan announced by the major banks in January of a gradual exit from the two operations. With about €100bn of liquidity still at the 3y LTROs, there is scope for the continuation of the recent repayment trend for the French banks. March data for German banks are not available yet, but we expect a small repayment in March after the big ones in January and February (about 30% of their total 3y LTROs borrowing in each month, likely because smaller banks repaid). With just about €20bn left, further repayment by German banks should not represent a game-changer for liquidity conditions. The same holds true for banks in other core countries. Therefore, according to our analysis, banks’ repayment decisions in France, Italy and Spain (with remaining liquidity in the 3y LTROs at €106bn, €232bn and €255bn respectively) will be crucial for the evolution of the liquidity conditions. Figure 2 outlines the possible evolution of the liquidity surplus depending on the pace of repayments in the next few months. Our analysis relies on assumptions of stable autonomous factors (the main source of volatility for liquidity needs)10 and stable roll of the other ECB’s refinancing operations (ie, no change in the current outstanding balance). FIGURE 2 Liquidity surplus: possible evolution (€bn) and EONIA dynamics under the most aggressive payback scenario (bp) Liquidity Surplus €4bn payback €4bn payback till June €8bn payback till year end 1,200 Eonia vs refi rate, bp, rhs 0 EONIA vs refi, under the most aggressive scenario 1,000 800 -25 600 400 -50 200 0 -200 -75 Jun-09 Apr-10 Jan-11 Dec-11 Sep-12 Aug-13 Source: EBF, ECB, Barclays Research 8 See Focus II in the April 15 euro Money Markets Weekly 9 Indeed, activity in the repo market by Spanish banks has increased significantly over the recent months. MEFF repo data show a rising trend in market volume, with March reporting the highest level of activity since the beginning of the series (January 2010). 10 As we stressed several times in our publications, such a component of liquidity needs is hard to predict as it is the result of the sum of several factors which move in different directions. Government deposits are the main source of volatility but they would be unlikely to trigger a particular trend in autonomous factors as a whole. In our opinion, the item in the Eurosytem’s balance sheet which is worth noting is “Liabilities to non-euro area residents denominated in euro”. It posted a significant increase last summer, which has been the main driver of the rising trend posted by autonomous factors. 18 April 2013 36 Barclays | Global Rates Weekly Evolution of the liquidity The grey line refers to a scenario of total repayment by the end of Q2 of €300bn surplus under different (corresponding to an average payback of €4bn until the end of June and no repayment repayment scenarios afterwards). This would bring the liquidity surplus to about €320bn from the current level of €360bn. Such a scenario would be consistent with President Draghi’s comments at the ECB’s February press conference that “according to [ECB] estimates, even after the second repayment of the LTRO, the excess liquidity should remain well above €200 billion”. In such a scenario we would expect EONIA to remain approximately unchanged at about 8bp, with some volatility probably related to the particular microstructure of the EONIA market characterized by low (reported) volume, which would make the fixing very sensitive to any particular flows reported by panel banks. The light blue line corresponds to a scenario in which there is a gradual weekly average repayment of €4bn till the end of the year. Here, the “psychological limit” of €200bn should be breached by the end of October. Finally, the red line outlines a scenario of aggressive average weekly paybacks of about €8bn until year end: the “passive” tightening of the liquidity conditions would push the surplus below €200bn at the end of July. In such a scenario, we would expect a significant tightening of the spread EONIA vs refi as shown in the Figure 2, similarly to in the second part of 2010 when the surplus approached zero. Risks of aggressive repayment are With a total of about €730bn of 3y liquidity currently still to be reimbursed, the likelihood of not negligible the third scenario is not negligible, in our view. With the repo market for Spanish banks working again (both domestically and via bilateral transactions with international banks), the gradual and moderate repayment by Spanish banks is likely to continue. French banks should behave in the same manner. Some of the Italian banks are likely to start a gradual exit from the 3y LTROs in the next few months also to smooth the effect of the €85bn of the government-guaranteed banks bonds that from March 1 2015 will no longer be eligible at the ECB’s operations. BTP redemptions in the next few months could be taken as an opportunity to unwind carry trades and start repaying ECB liquidity. An improvement of the political situation with the formation of a new government in the very near term might also support the decision to repay part of the liquidity. ECB’s options Passive tightening could At the April meeting, the ECB remarked on its commitment to maintain an accommodative affect the ECB’s monetary policy stance as long as needed, with the fixed rate procedure with full allotment, accommodative stance and to be vigilant that liquidity conditions in the euro money markets remain consistent with its stance. The “passive” tightening of the liquidity conditions, which will develop over time if more LTROs are reimbursed as we discussed above, could represent a key issue, with the potential to limit the effect of the accommodative stance on money market rates. Notwithstanding this, we believe that the ECB still has a few options to counteract the possible tightening through measures that would affect the liquidity needs (lower reserve requirements) and the supply of liquidity (new LTROs). In addition, the ECB could cut the refi rate. In our view, the ECB will be more open to reiterate actions that have clearly worked in the past: rate cuts or LTROs. The room for large changes in collateral eligibility is limited in our view, given that already, there is a lot of flexibility on that side, and the ECB is likely unwilling to change haircuts much (although we would note that there is a review of the collateral framework during the summer). 18 April 2013 37 Barclays | Global Rates Weekly Lower reserve requirement: not that likely Reduce liquidity needs, but Lowering the reserve ratio to (or close to) zero would cause an increase in the liquidity risks that this could be offset surplus, with the additional implication of freeing up eligible collateral. Here, the main risk by lower MRO borrowing would be that such a measure may not be enough to counteract the passive tightening, also because the reduced liquidity needs might be partially offset by a lower MRO borrowing11. New LTROs: quite likely New 3y LTROs with particular On this point, the ECB could decide to conduct a new LTRO, in order to reinject liquidity. features While it may be used to roll maturing 3y LTROs for a number of banks (thus not increasing the surplus, but keeping it stable, and with the same long duration), the ECB could design particular features (especially on the collateral side to take into account, for example, the differentiation introduced by the Basel3 LCR’s HQLA regulations)12 to make it attractive also for those banks that have reimbursed the 3y LTROs (or are more likely to do it), in order to increase net liquidity and push the surplus higher again. The most likely option is for the ECB to conduct 3-5y LTROs, unlimited in amount, on specific SME-type collateral, and possibly at fixed (rather than variable) rates. We would expect an announcement on LTROs in the summer. Refi rate cut: possible Cut in the refi rate but not in Finally, the ECB could decide to act on the policy rate side, by cutting the refi rate. While the depo rate such a measure would have no meaningful effect on the real economy owing to the frictions in the monetary policy transmission, on the market side, it would reduce the level towards which short rates would gravitate in a context of lower liquidity. We do not expect a cut in the depo rate, even if it is theoretically an option. A policy rate cut scenario would imply tightening the monetary policy corridor and potentially hurting interbank liquidity. However, in our view, the impact of liquidity regulations is more important in that respect, as highlighted in the ECB’s article mentioned above. LCR regulations push banks to borrow (and redeposit) at the ECB, on essentially non LCR eligible collateral. Forward guidance: implicit or explicit? Precise forward guidance As President Draghi recently mentioned ”during crisis times, when short-term nominal rates are at zero or close to zero, they cannot be adjusted further down. The central bank may then engage in active communication reassuring markets that the future path of policy rates would not deviate from the current low level for a certain period or until certain observable conditions are verified” 13. The ECB has conducted its operations with the Fixed rate full allotment procedure since October 2008, rolling its commitment approximately on a quarterly or six-month basis. It will likely continue to do so, but could engage in more precise forward guidance, for example by conducting fixed rate LTROs – a move towards Fed-like guidance is much more unlikely. SME lending: ‘it’s complicated’ Measures to support SMEs While the issue of access (and pricing) of credit to SMEs has been very much in focus, and lending: no magic wand there is a good understanding of what the problems are, there is no ‘magic wand’ as Coure 11 Note that, as of the end of March, out of a total MRO borrowing of €125bn, Greek banks were the main borrowers with €75bn (they use the MRO as a flexible source of funding rather than for liquidity management). The remaining €50bn is spread over the other countries, whose banks use the weekly operations for their near-term liquidity management needs (like a reserve requirement) 12 See the article “Liquidity Regulation and monetary policy implementation”) in the April monthly ECB bulletin 13 Speech by Mario Draghi, President of the ECB, at the “Room for discussion” of the Study Association SEFA and the Faculty of Economics and Business, Amsterdam,15 April 2013 (http://www.ecb.int/press/key/date/2013/html/sp130415.en.html) 18 April 2013 38 Barclays | Global Rates Weekly stated in a recent speech14 and a solution that the ECB or other policymakers could put in place. We expect a number of small (but important) moves to improve the access to finance and lower its cost, but not a big announcement or scheme at this stage. In our view, a combination of EIB-type guarantees with attractive ECB liquidity (from a rate and collateral perspective) would be the quickest way to potentially improve the situation – assuming the economy does pick up. Market implications Bull flattening trend A bull flattening trend has characterized the price actions in the euro money markets over the last few weeks. Two aspects are worth highlighting - see below. 1) Since the ECB’s April meeting, when President Draghi left the door open for a refi rate cut, the front end of the money market rates have not changed significantly. The stability of the EONIA curve is probably related to the market having already priced out the possibility of a cut in the deposit facility. However, the lack of any significant reaction in white contracts on the Euribor future curve is probably mostly because after the sharp rally since the beginning of the year, the white contracts already incorporate some probability of a 25bp refi rate cut by June. 2) The increase in the 3y LTRO repayment has not affected the term premium on the curve. Unlike in January, when the expectations for aggressive repayment caused a significant steepening of the curve, at present, the curve has continued to bull flatten probably discounting some ECB’s actions to counteract the passive tightening. Not easy to extract market It is hard to extract precise market expectations on the ECB action from the current prices. expectations on the ECB’s action The EONIA forward for ECB’s June 2015 meeting is at about 25bp. This could be the result on the policy rates of some term premium (or the likelihood of a depo rate hike) in a context of still abundant liquidity conditions after the maturity of the 3y LTROs, or a normalization of EONIA to a level of policy rates at 25bp (therefore two cuts from the current level) given a low liquidity surplus. A continuation of the current bull Having said that, in the very near term, we expect the recent trend of moderate bull flattening trend is likely flattening of the money markets curve to continue, accompanied by some re-tightening of the FRA/EONIA spread in the forward space. A possible rate cut in May could cause an adjustment of white contacts to a new level of the 3m Euribor (which we project at about 15bp if the refi is cut 25bp, as a consequence of the tightening of the FRA/EONIA rather than of the reduction in the EONIA leg of the spread). This could lead to a moderate, but temporary, bull steepening of the curve. In this respect, next week the focus will be on the Flash Eurozone PMI data as a further decline (the third in a row) might reinforce expectations of a rate cut being delivered in May. 14 Speech by Benoît Cœuré, Member of the Executive Board of the ECB, Eurofi High Level Seminar organized in association with the Irish Presidency of the Council of the EU, Contribution to plenary Session 11: Challenges and feasibility of diversifying the financing of EU corporates and SMEs, Dublin, 11 April 2013 (http://www.ecb.int/press/key/date/2013/html/sp130411.en.html) 18 April 2013 39 Barclays | Global Rates Weekly UNITED KINGDOM: RATES STRATEGY Decline and fall Moyeen Islam As realised volatility declines, increased interest in positive roll-and-carry structures is +44 (0)20 7773 4675 becoming a driver of investment, this favours the 10-15y sector of the gilt curve and firstname.lastname@example.org consequentially, a re-steepening over Q2 13. There was little news from the The Minutes of the April MPC meeting offered very little fresh insight as to the future policy April MPC Minutes and the direction. For the third month in a row, the vote on whether to undertake a further £25bn of market remains listless asset purchases was split 6-3. Despite the much-discussed change in the BoE’s remit in the Budget, the Committee saw no reason yet to change its stance. Reflecting this sense of ennui, the market has seemed unable to start the quarter with any sense of conviction, with yields grinding lower. Figure 1 shows the evolution of the cumulative change in 10y yields In previous years, 10y yields over the year to date and compares it with the previous three years. Typically, the market have traded with a bullish bias has traded with a bullish bias in the early part of the year, with the bigger moves in the in the second quarter of the second half. Clearly, there was greater scope to rally in 2010 and 2011, when the outright year market yield was materially higher. So perhaps 2012 is more of a salutary guide to what the evolution of yield might be. Here, we can see that the trading range for yields was extremely limited. Thus with outright yields only c.25bp from the historical lows, the risk to the market is that yields grind lower during the year and look to revisit these lows. With the market range bound, The corollary of a market trading in a limited range is a continuation of the secular decline in realized volatility has been realised volatility. Realised volatility in 10y gilts is now as low as it has been in the last 5 secularly declining years – Figure 2. This move is further reflected in the interest rate swaption markets. Figure 3 shows that here the decline in realised volatility has similarly led to a decrease in implied volatility in swaptions, where vol now sits at historical lows. The decline in realised volatility and low level of absolute rates combined with the steepness of the term structure means that we should see continued interest in positioning for positive –roll-and–carry in both cash and swaps. Figure 4 shows the current 3m –roll and carry for the gilt curve. The relative steepness of the curve means that the highest roll and carry is in 7-10y. However, equally, we note that the very short end of the curve, where gilt yields trade below cash rates, offers significant negative roll and carry, most likely indicative of the non-discretionary investment that is made here (ie, gilt and cash are not seen as substitutable asset classes). FIGURE 1 FIGURE 2 Cumulative change in 10y nominal yields over 1 year (bp) Realised volatility has declined globally (abpv equiv) 50 0 170 -50 120 -100 change 70 since start -150 of the year (bp) 20 -200 2008 2009 2010 2011 2012 2013 1 45 89 133 177 221 10yr CM Gilt realised vol (abpv equivalent) Business days since 1 Jan 10yr CM Bund realised vol (abpv equivalent) 2013 2012 2011 2010 10yr CM UST realised vol (abpv equivalent) Source: Barclays Research Source: Barclays Research 18 April 2013 40 Barclays | Global Rates Weekly FIGURE 3 FIGURE 4 Selected benchmark implied vol (abpv equiv) Gilt 3m roll and carry curve (bp) 175 10 150 125 5 100 0 75 50 -5 25 -10 0 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 -15 GBP 3m 5y ATM Normalised Vol GBP 3m 10y ATM Normalised Vol -20 GBP 3m 30 y ATM Normalised Vol 2014 2019 2024 2029 2034 2039 2044 2049 2054 2059 Source: Barclays Research Source: Barclays Research Roll and carry is likely to A prerequisite to the realisation of roll and carry is the decline in realised volatility; a more become an increasingly instructive metric than roll and carry on its own is the carry: vol ratio. Most interesting are important determinant of the areas of the curve with high carry:vol ratios (ie, those that have high roll and carry and positioning low realised volatility) as these suggest a greater probability that the roll and carry (which ultimately is a function of the forward curve) being realised. Again, this points towards a bias to positioning in the belly of the curve. Figures 5 and 6 show the current carry:vol ratios for the curve and the historic z-scores across the curve. These suggest that investors looking for long duration and to earn the roll and carry on the curve would be best served by positioning in 10-15y, where the z-scores on the ratios are higher compared with other bonds on the curve. This implies further steepening pressure on the gilt curve if the belly continues to richen. In addition, a weak Q1 13 GDP estimate next week may sway the market into believing further asset purchases are likely. This would also steepen Gilt 10/30s, as has historically been the case with previous rounds of QE. FIGURE 5 FIGURE 6 Gilt carry:vol ratio curve Gilt carry:realised vol 1-year z-scores 0.70 2 3m gilt carrry/realised vol 0.60 1 0.50 0 0.40 -1 0.30 -2 0.20 -3 Z-score 0.10 -4 2014 2022 2030 2038 2046 2054 2013 2018 2023 2028 2033 2038 2043 2048 2053 2058 Source: Barclays Research Source: Barclays Research 18 April 2013 41 Barclays | Global Rates Weekly SUPRANATIONAL, SUB-SOVEREIGN & AGENCIES EFSF – to tap or not to tap This is an edited extract from The AAA Investor, 18 April 2013. Fritz Engelhard Given the strong backing of EFSF debt by buy-and-hold investors, expected guarantee +49 69 7161 1725 changes for EFSF bond issuance could be the catalyst for spread tightening of email@example.com outstanding EFSF debt. The demand and size of last week’s EFSF 5y bond represented one of the largest bond Michaela Seimen issuances of a supranational organisation. At EUR8bn in magnitude and given a strong +44 (0)20 3134 0134 performance in the secondary market, questions have arisen around the reasons for the firstname.lastname@example.org particular market attention on this bond. We believe the reasons are manifold: Jussi Harju, CFA • The EFSF has established itself since its first issuance in January 2011 as a frequent and +49 69 7161 1781 liquid issuer. email@example.com • Confidence in the European sovereign support structures and in the eurozone seems to be holding up in spite of recent uncertainty regarding the Italian political situation and the support discussions on Cyprus. • Investors are attracted to European supranational issuers as they offer a diversified European risk with a strong guarantee participation of Germany, combined with interesting relative value options versus European sovereign debt. • Strong support from Asian investors for past issuance, as well as previous public statements by the Japanese government stressing its commitment to the eurozone, have furthered expectations that Japan, in light of the BOJ’s monetary easing policy, will focus strongly on investments in European bond markets. • The announcement by the EFSF that Cyprus applied to become a ‘stepping-out guarantor’ will change the guarantee structure of the issuer. Thus, from the point of change in the guarantee structure, the issuer will not be allowed to reopen bond issuance for existing bonds. Therefore, investors with short positions could be caught out by tighter liquidity in secondary market trading. In this week’s AAA Investor, we will assess more closely the potential effect of this latter point on outstanding bond performance. Supported by an increased and diversified investor base, we believe EFSF bonds could benefit from the guarantee change with further spread tightening versus agencies and European sovereigns. We see good value in the 3y to 5y segment, as this part of the curve is supported by a more diversified investor base, and particularly by institutional investors with a typically more long-term investment view. Upcoming guarantee change for the EFSF In line with its recent bond issuance, the EFSF announced on its webpage that on 8 April 2013, given its outstanding request for sovereign support, the Republic of Cyprus submitted a request to become a so-called ‘stepping-out guarantor’ under the EFSF Framework Agreement. According to the EFSF, this request will be considered by the remaining guarantors during a meeting planned for 25 April 2013. However, the stepping-out will only become effective upon approval by the Euro Working Group. 18 April 2013 42 Barclays | Global Rates Weekly FIGURE 1 EFSF current guarantee structure and expected change to contribution keys with Cyprus becoming a ‘stepping-out guarantor’ EFSF guarantee Adjusted contribution keys commitments (to cover bond issuance, %) Adjusted contribution keys (as of 2011 with Greece, Ireland and (to cover bond issuance, %) agreements - Portugal as ‘stepping-out with Cyprus as additional Country EUR mn) guarantors’ ‘stepping-out guarantor’ Austria 21,639 2.99 2.996 Belgium 27,032 3.72 3.728 Cyprus 1,526 0.21 0 Estonia 1,995 0.27 0.271 Finland 13,974 1.92 1.924 France 158,488 21.83 21.876 Germany 211,046 29.07 29.131 Greece 21,898 0 0 Ireland 12,378 0 0 Italy 139,268 19.18 19.220 Luxembourg 1,947 0.27 0.271 Malta 704 0.1 0.100 Netherlands 44,446 6.12 6.133 Portugal 19,507 0 0 Slovakia 7,728 1.06 1.062 Slovenia 3,664 0.51 0.511 Spain 92,544 12.75 12.777 Total 779,784 100 100 Source: EFSF, Barclays Research We do not believe there should be any obstacles to the implementation of this request, given the established procedure with previous supported sovereigns and the fact that Cyprus held overall a very small position in the guarantee structures, namely 0.21% of existing bond issuance. However, we highlight that based on a changed guarantee structure, existing EFSF bonds will not be able to be re-opened. Since the establishment of the EFSF in 2010, this is the third change of the guarantee structure, with Ireland being the first and Portugal the second steeping-out guarantor, due to their respective support agreements. However, on top of changes to the guarantee structure, the downgrade of the EFSF in January 2012 resulted in an amendment to the deed of guarantee. As a consequence, previous bond issuance became excluded from the ability of being reopened. Figure 2 gives an overview of the outstanding bond volumes previously having been affected by these changes. 18 April 2013 43 Barclays | Global Rates Weekly FIGURE 2 Review of changes to the EFSF guarantee structure and affected outstanding debt amounts Total affected Bond tap- amount (EUR) able as of as per Announcement Amount 17 April outstanding ID_ISIN Ticker date Maturity Coupon Currency outstanding 2013? bonds Event date Event EU000A1G0AA6 EFSF 25/01/2011 18/07/2016 2.75 EUR 5,000,000,000 No 5,000,000,000 17/05/2011 Portugal requested support & became stepping-out guarantor EU000A1G0AB4 EFSF 15/06/2011 05/07/2021 3.375 EUR 5,000,000,000 No EU000A1G0AC2 EFSF 22/06/2011 05/12/2016 2.75 EUR 3,000,000,000 No EU000A1G0AD0 EFSF 07/11/2011 04/02/2022 3.5 EUR 3,000,000,000 No EU000A1G0AE8 EFSF 03/01/2012 04/02/2015 1.625 EUR 3,000,000,000 No 14,000,000,000 16/01/2012 Rating downgrade by S&P with subsequent amendment of the guaranteed deed EU000A1G0AJ7 EFSF 19/03/2012 30/03/2032 3.875 EUR 2,500,000,000 Yes EU000A1G0AK5 EFSF 20/03/2012 15/05/2017 2 EUR 4,960,000,000 Yes EU000A1G0AR0 EFSF 23/04/2012 02/05/2019 2.625 EUR 5,500,000,000 Yes EU000A1G0AS8 EFSF 23/05/2012 01/06/2015 1.125 EUR 4,483,600,000 Yes EU000A1G0AT6 EFSF 12/06/2012 03/04/2037 3.375 EUR 3,500,000,000 Yes EU000A1G0AU4 EFSF 10/07/2012 15/09/2017 1.625 EUR 6,000,000,000 Yes EU000A1G0A16 EFSF 28/08/2012 05/09/2022 2.25 EUR 3,974,350,000 Yes EU000A1G0A24 EFSF 15/10/2012 30/11/2017 1.125 EUR 5,900,000,000 Yes EU000A1G0A81 EFSF 15/01/2013 22/01/2020 1.5 EUR 6,000,000,000 Yes EU000A1G0A99 EFSF 29/01/2013 05/02/2018 1.25 EUR 5,000,000,000 Yes EU000A1G0BA4 EFSF 25/02/2013 07/03/2016 0.5 EUR 4,000,000,000 Yes EU000A1G0BB2 EFSF 09/04/2013 16/04/2018 0.875 EUR 8,000,000,000 Yes 59,817,950,000 25/04/2013* Cyprus will become a stepping-out guarantor Total fixed coupon EFSF primary market issuance of EFSF EUR 78,817,950,000 Note: *Provisional affected amount to become non-tap-able at time of writing if Cyprus becomes a ‘stepping-out guarantor’. Source: Bloomberg, Barclays Research Given that previous changes to the guarantee structure of the EFSF happened in the early stages, the effect was more limited, affecting outstanding bond sizes of EUR5bn and EUR14bn as of May 2011 and January 2012, respectively. Therefore, existing investors could have expected, at those points in time, a high volume of future primary issuance to fill any short positions or investment interests in particular maturity brackets. However, since the last change in January 2012, the EFSF has issued about EUR59.8bn in additional primary market fixed coupon bonds, which are currently tap-able in the market. Once the new guarantee structure, with Cyprus being a ‘stepping-out guarantor’, is implemented, the liquidity in those bond issues could be affected. 18 April 2013 44 Barclays | Global Rates Weekly FIGURE 3 Review of EFSF tapping activities since establishment of the issuer EUR bn 12 10 8 6 4 2 0 Jan-11 Mar-11 Apr-11 May-11 Oct-11 Jan-12 Mar-12 Apr-12 May-12 Oct-12 Jan-13 Mar-13 Apr-13 Jul-11 Jul-12 Feb-11 Sep-11 Dec-11 Feb-12 Sep-12 Dec-12 Feb-13 Jun-11 Aug-11 Jun-12 Aug-12 Nov-11 Nov-12 New Issue Tap Source: Bloomberg, Barclays Research Remaining funding needs of EFSF until 2014 As per the issuance last week, the EFSF has funded about EUR25bn of its yearly EUR58bn issuance programme so far this year. Should Cyprus become a ‘stepping-out guarantor’ before the next issuance, about EUR23bn of issuance this year and EUR34bn of issuance in 2014 would fall under the new guarantee structure, providing a similar amount to the recently tap-able pool of bonds. With total expected supply volume of about EUR57bn for this year and next, there seems to be sufficient room for investors to fill their investment needs in various maturity brackets. However, this depends on potential further changes to the guarantee structure if any other sovereigns apply to become ‘stepping-out guarantors’. Also, investors need to keep in mind that due to the establishment of the permanent ESM structure, the EFSF will cease to enter new programmes from July 2013. However, the support facility will complete existing programmes for Ireland, Portugal and Greece and manage the rollover of existing debt. Following the conclusion of existing programmes, the EFSF will exist in an administrative capacity until all outstanding debt has been repaid (including refinancing operations). However, given that EFSF’s successor, the ESM, is now in place, we could very well envisage that EFSF debt could be transferred to ESM to avoid a more illiquid phase-out period for any remaining smaller volumes of bond issuance. All these factors will result in overall amounts and respective tap-able amounts of the EFSF staying rather limited. Investors therefore might need to focus more on supply in primary market issuance. As a side note, on 12 April 2013, the Eurogroup and Ecofin ministers discussed the adjustment programmes for Ireland and Portugal and agreed in principle to support the sovereigns by increasing the weighted average maturity limit of EFSM and EFSF loans by seven years provided that ‘the troika’ confirms a continued successful programme implementation in the upcoming reviews. The extension of the programmes would smooth the debt redemption profiles of both countries and therefore lower the refinancing needs in the post-programme periods. From an investor perspective, we regard these changes as supportive for the spread performance of EFSF bonds because the issuer shall roll bond facilities, and therefore provide some extended liquidity to the EFSF curve. Liquidity in bond issuance – a factor of several denominators There are different reasons issuers choose to re-open bond issuance, although the overall focus of issuers might be on supporting the liquidity of the bonds: 18 April 2013 45 Barclays | Global Rates Weekly • Tap issues are often based on reverse enquiries of investors and therefore typically suppress new issuance premium. In addition, taps allow an issuer to avoid certain transaction or legal costs, resulting in a cheaper funding option for issuers. • Issuers are committed to supporting liquidity in their bonds; re-opened bond issuance often re-attracts investors and supports a follow-on performance of respective bond spreads in secondary markets. A general assumption by market participants is that bonds with higher liquidity benefit from tighter spreads and, thus, reduce liquidity premiums for future primary market issuance. • With an increased size of single bond issuance, the liquidity in the single bonds is understood to increase, and recognition as a benchmark issuance compared with sovereign bond markets improves. Growth of outstanding EFSF bond volumes in the market since the first issuance in January 2011 and diversification of the investor base, combined with a more benign market environment and more confidence of investors in the future of the eurozone, have lead to the EFSF becoming one of the more liquid bond choices for investors in the SSA environment. However, the EFSF faces a crux in this respect. As SSA debt in general often does, EFSF debt also benefits from a relatively supportive investor base, such as central banks and other committed longer-term investors. Although this helps reduce volatility in such bonds, liquidity, particularly in smaller bond issuance, is typically undermined. It is therefore not surprising that larger issuers in the SSA segment in particular try to focus on increasing the size of individual issuance to provide better liquidity for their bonds. Large liquid bonds versus ‘rare’ bond issuance That said, the argument for larger issuance and the ability to create liquidity in particular parts of the curve via reopened bond issuance is a very feasible way to achieve tighter spreads for bond issuance, in our opinion, because certain liquidity premiums do not apply for these issuances in primary markets. However, the SSA segment provides specific examples of smaller issuers that benefit from the ‘rare issuer’ status and are partly traded at tighter levels than is typical for larger issuers. Prominent examples in this respect are the EU and RENTEN, as discussed in The AAA Investor, 31 August 2012. We now assume in this respect that EFSF bonds could change from being very liquid because of the outstanding volume to being a ‘rarer’ issuer, as investors looking for particular bonds or maturity brackets are forced into new primary market issuance. Ultimately, this could become a positive development for existing investors and the issuers alike because EFSF bond issuance could experience tightening in parts of the EFSF curve. The most recent EFSF issuance already gave some indication of investors’ strong focus on the issuer. According to information by the EFSF, the bond received orders from close to 200 investors worldwide with a total volume of about EUR14bn, enabling the issuer to supply a bond with a size of EUR8bn. However, part of the success in this recent bond issuance is also applicable to the price concession the issuer applied on marketing the bond. Initial price thoughts had been about 12bp, compared with a fair secondary market value at about 6bp; however, due to the strong demand, final pricing was set at 9bp, with a strong follow-up performance of the bond in the secondary market. Furthermore, the pricing was relatively attractive compared with European sovereigns, such as France, to which the new bond offered a pick-up of about 19bp. Investors are particularly attracted by this aspect because European supranational issuers offer a diversified European risk with a strong guarantee participation of Germany. 18 April 2013 46 Barclays | Global Rates Weekly FIGURE 4 FIGURE 5 EFSF credit term structure vs. selected SSAs EFSF credit term structure vs. selected European sovereigns PP swap PP swap spread (bp) spread (bp) 40 40 30 30 20 20 10 10 Years 0 Years 0 -10 -20 -10 -30 -20 -40 -30 -50 -40 -60 0 3 6 9 12 15 0 2 4 6 8 10 12 EFSF EU KFW EIB EFSF Germany France Netherlands Note: Click chart to view dynamic data on Barclays Live. Note: Click chart to view dynamic data on Barclays Live. Source: Source: Barclays Live - Chart Source: Source: Barclays Live - Chart 18 April 2013 47 Barclays | Global Rates Weekly SCANDINAVIA: RATES STRATEGY Unchanged policy rates, but significantly softer forward looking policy guidance Mikael Nilsson Rosell We continue to see value receiving SEK10y swaps versus EUR, despite recent +44 (0)20 7773 6057 outperformance. The inflation outlook remains benign and we fear that leading firstname.lastname@example.org indicators will soften going into the summer. As expected, the Riksbank (RB) left policy rates unchanged at 1.0% at its recent meeting, but the forward looking policy guidance was significantly softer than expected. Indeed, the RB's own policy rate path signals a 25-30% probability of a cut at the 3 July policy meeting, well in line with our expectations going into the meeting (see Unchanged policy rates, but inflation suggests room for cuts, 11 April). The RB also revised its path beyond the July meeting significantly lower, with policy rates now expected to be 0.94% by year-end 2013 (previously 1.14%) and 1.43% by year-end 2014 (previously 1.89%). The chief driver of the softer forward looking policy guidance was a downward revision of the RB inflation forecast, see Figures 1-2. Trade ideas Rates rallied across the curve following the softer than expected policy announcement, with 10y, 5y and 2y cross-market spreads versus EUR tightening between 8bp and 10bp. With the RB having opened the door for the possibility of further rate cuts, we continue to see value in holding longs in the very front end. However, rather than holding cross-market tighteners versus EUR in Sep ’13 3m FRA, we see better value in outright longs from here. We also continue to see value in receiving 10y swaps versus EUR, despite recent outperformance. Besides Sweden’s unchallenged AAA status, the spread should continue to benefit from near-term rate cut prospects and the benign inflation outlook. Indeed, we expect CPI to remain well below euro area inflation well into 2014. Going into the meeting we recommended entering tactical SGB2s5s steepeners. However, the benign inflation outlook and the RB’s soft forward looking policy guidance will likely act to nourish expectations that policy rates will remain ‘low for long’ over the coming weeks. Hence we will rather look for opportunities to enter 5y SGB/DBR tighteners (currently c. 85bp) on the back of any post-RB SGB underperformance. FIGURE 1 FIGURE 2 Riksbank’s policy rate forecast Riksbank’s CPIF forecast % % 5 3.00 4.5 4 2.50 3.5 3 2.00 2.5 2 1.50 1.5 1 1.00 0.5 0 0.50 Jul-98 Jul-00 Jul-02 Jul-04 Jul-06 Mar-97 Mar-99 Mar-01 Mar-03 Mar-05 Mar-07 Nov-97 Nov-99 Nov-01 Nov-03 Nov-05 Nov-07 0.00 Feb-09 May-10 Aug-11 Nov-12 Feb-14 May-15 Policy rate April February CPIF April February Source: Riksbank, Barclays Research Source: Riksbank, Barclays Research 18 April 2013 48 Barclays | Global Rates Weekly Pronounced downward In the policy announcement press-release, the RB concludes that growth prospects are revisions of the inflation gradually brightening, but its new assessment nevertheless assumes that it will take longer forecast, despite unchanged for inflation to reach its target (2.0%). Indeed, while the RB left its 2013 CPIF forecast growth prospects unchanged (1.0%) it lowered its 2014 CPIF forecast by 0.4 percentage points to 1.4% and kept its 2015 forecast unchanged (2.0%). This was a slightly more front-loaded revision than our expectation that the RB would lower its CPIF forecast by 0.2 percentage points in both 2013 and 2014. We continue to see downside While we continue to see downside risks to the RB inflation forecast (mainly 2015) they risks, but they are less evident seem less evident than the downside risks to the RB forecast going into the April meeting than before, although data after the revisions, even taking into account the recent benign 3-year wage deal and the still looks likely to soften near term relatively strong SEK. Similarly we don’t see any particular risks to the RB near-term growth forecast, although we expect forward looking sentiment data to temporarily soften slightly going into the summer. RB majority remains concerned In the press-release the RB also concludes that a cut today “would mean that inflation with medium-term systemic attained the target somewhat quicker, but at the same time it would further increase the risks, but now seems willing to risk of imbalances building up”. In our view, this again confirms that the majority of the tolerate moderate gains in Executive Board (Ingves, Wickman-Parak, Jansson and Jochnick) also remains concerned house prices that a ‘too low policy rate for too long’ might nourish medium-term systemic risks. However, interestingly, in its Monetary Policy Update (MPU) the RB now assumes that house prices and debt levels will increase at a faster pace, with household debt as a percentage of disposable income now expected to increase from 174% to just over 177% by the beginning of 2016. In our view, this suggests that the majority of the Executive Board, despite its medium-term systemic concerns, is now more willing to accept a moderate upturn in house prices and debt levels given the benign inflation outlook. That said, we look forward to gaining significantly more colour on the how the different Board members view this issue in the forthcoming minutes (published 29 April). We hold on to our base case of All in all, we hold on to our base case scenario that policy rates will be left unchanged at the unchanged policy rates for July meeting. However, we believe the probability of a 25bp cut to 0.75% is a finely balanced now, but believe that the call. We do fear that high frequency survey data will weaken (align with recent moves probability of a 25bp cut is just among Sweden’s main training partners) going into the summer, but we still feel shy of 50% unconvinced about the ultimate magnitude of the slowdown. We also continue to see downside risks to the RB inflation forecast, but we do not believe that the RB will follow up with another substantial downward revision in July. The key driver for our base case call will therefore likely be whether the forthcoming minutes validate our impression that the majority of the Board is now willing to tolerate moderate near-term gains in house prices and debt levels, despite concerns with medium-term systemic risks. In this context it will obviously also be highly interesting to see who will replace Deputy Governor Wickman- Parak (retires in May) and whether LEO Svensson’s mandate will be extended. In addition, we will obviously also continue to monitor SEK developments, especially on the back of an ECB rate cut. 18 April 2013 49 Barclays | Global Rates Weekly INFLATION-LINKED MARKETS: EURO AREA Viva l’Italia… again! Khrishnamoorthy Sooben The recent BTP Italia issuance does not seem to have prompted significant switches out +44 (0)20 7773 7514 of BTP€is, unlike the placement in October. We therefore expect 3-5y BTP€is to correct khrishnamoorthy.sooben@ some of their recent cheapening. That sector also appears attractive in breakeven. barclays.com The placement of the latest BTP Italia (maturing April 2017) once again drew very impressive demand; at slightly above €17bn, the volume is only about €1bn less than the size of the October deal. The press release containing the details of the placement mentions a large participation by retail investors, with a significant role of retail portfolio managers and private banking. The high participation of retail investors is evidenced by the fact that more than 80% of the number of contracts amounted to €50k or less. However, the presence of institutional investors is also highlighted, with foreign ones accounting for about 12% of the total amount issued. While it is difficult to properly gauge the participation of institutional investors from the placement statistics, it again appears likely that the BTP Italia has attracted interest well beyond the primary investor base – the domestic retail one – that it supposedly targets. One similarity between the BTP Italia April 2017 and the previous issue is the apparent high participation of institutional investors. As well, both bonds were issued when a BTP was being redeemed, which suggests that both benefited from reinvestment flows, given that BTP holdings by Italian retail investors are non-negligible. One striking difference, however, is that, unlike at the October supply, the latest issue does not appear to have prompted significant switches out of BTP€is or BTPs (see Viva l’Italia, 18 October 2012). In October, there was a general consensus that the issue was being offered at a very cheap level (based on the minimum coupon) relative to the BTP€i16 or 4y sector BTPs. The performance of the BTP€i16 during the placement period in October indicated strongly that there were switches out of it. While it does not appear to have been the case for the recent BTP Italia, the BTP€i market seemed to have braced itself for that possibility before the opening of the issuance window, as evidenced by the cheapening in 3-5y BTP€is before the placement period. With no indication of significant switches ex-post, we expect a correction of this “pre-emptive” cheapening. FIGURE 1 FIGURE 2 BTP€i issuance dwarfed by the BTP Italia Some corrective cheapening in 5y sector ASW in 2013 30 Gross BTP€i issuance BTP Italia issuance 800 BTP€i19 z-spread asw 250 BTP€i19 relative z-spread asw 700 25 200 600 20 500 150 15 400 300 100 10 200 5 50 100 0 0 0 03 04 05 06 07 08 09 10 11 12 13 Apr-10 Dec-10 Aug-11 Apr-12 Dec-12 Note:* 2013 data are year-to-date. Source: Italian Treasury, Barclays Research Source: Barclays Research 18 April 2013 50 Barclays | Global Rates Weekly The bulk of interest for BTP€is over the past year or so has been from domestic investors and mainly in asset swap. For instance, demand from Italian bank treasuries for their liquidity portfolios accounts for a notable share of activity. Interest is driven mainly by the extent to which BTP€is trade at a discount in asset swap versus BTPs. Such demand tends to materialise mainly in sub-10y issues, given the nature of the portfolios involved. With 3- 5y BTP€is having cheapened in absolute and relative asset swap to some extent this year, we expect asset swap interest in that sector to continue. From a breakeven valuation perspective, the 3-5y BTP€is also appear very cheap. Based on our economists’ euro HICPx forecasts, their forward breakevens to mid-September (when there is no seasonality bias in valuations) are less than 1%. This may not appear strikingly cheap compared with the distressed levels reached in 2011 or early 2012, but as we explain in This time it should different, 28 February 2013, those distressed levels were caused by specific factors that are no longer at play. In the 10y area, BTP€is offer notably more value in real yield or asset swap (absolute and relative to nominals) and are therefore attractive for carry or yield-enhancement positions. However, as we highlighted above, long maturities do not suit the portfolio maturity profiles of the main asset swap buyers as much as shorter- dated issues. One obvious question that arises following two consecutive sizeable BTP Italia placements is whether the Tesoro will be driven to reduce BTP€i issuance. We note also that Reuters reported comments (in Italian) from the head of the Italian debt management office, Maria Cannata, that Italy is investigating potential issuance of an Italian linker indexed to Italian inflation and aimed at institutional investors but copying the format of BTP€is rather than the BTP Italia. When the first BTP Italia was launched last year, Tesoro officials indicated that the new programme provides scope to reduce BOT issuance, rather than BTP€is. In the 2013 Guidelines for Public Debt Management, the commitment to the BTP€i market was reiterated, and this commitment to supply was evident even after the BTP Italia drew more than €18bn in demand last October. However, the Tesoro also stated that BTP€i issuance choices will have to consider overall exposure to inflation risk, with the BTP Italia taken into account. In our view, this does not necessarily indicate that BTP€i issuance will be cut. However, we believe the Tesoro may elect to reduce BTP€i issuance in the 4y sector if it feels this sector is being crowded by BTP Italia issuance, especially if the latter continues to capture significant institutional demand. FIGURE 3 FIGURE 4 BTP€i relative value most attractive in 10y sector 3-5y BTP€is cheap in breakeven 80 BTP€i relative z-spread asw curve 1.8 70 1.6 1.4 60 1.2 50 1.0 40 0.8 BTP€i spot breakeven curve 30 Forwards to mid-Sep 13 implied by 0.6 Barclays' Euro HICPx forecasts 20 0.4 2014 2019 2024 2029 2034 2039 2014 2019 2024 2029 2034 2039 Source: Barclays Research Source: Barclays Research 18 April 2013 51 Barclays | Global Rates Weekly INFLATION-LINKED MARKETS: UNITED KINGDOM Supply me to the moon Henry Skeoch UK linker supply is unusually concentrated in the next three weeks. The 15y sector is +44 (0)20 7773 7917 cheap on the curve, but the IL27 offers more value than the auctioned IL29. We expect email@example.com the IL62 mini-tender to draw good demand and favour IL22/62 breakeven steepeners. The UK linker market faces heavy supply over the next three weeks, which we estimate will supply roughly £9 mn/bp of risk. First up is a £1.3bn notional auction of the IL29 on Wednesday, April 24. In cash terms, this is worth £1.5bn and so is slightly lower than the £1.6bn average linker auction size implied by the FY13/14 gilt remit. We think the slightly lower-than-average auction size reflects the concentration of supply in the coming weeks, and also the weakness in linker breakevens since the start of April. The 5y sector in particular came under selling pressure in recent days, but once adjusting for carry it has outperformed the 10y sector this month. The IL29 has been one of the better performing sub-20y issues recently, showing little sign of specific pre-auction concession. The IL27 offers markedly better value on the real yield curve and we continue to recommend buying the issue versus the IL22+IL32 real yield barbell. We see risks of the IL29 reversing its recent outperformance into its auction, which we estimate will shorten all- linker and +5y indices by 0.01y and 0.02y respectively. Long end of curve offers most After Wednesday, focus shifts to the 30 April IL62 mini-tender then to the scheduled IL44 value on a relative basis auction on 8 May. The mini-tender was announced on 12 April and was likely no great surprise, given the lack of other ultra-long linker supply this quarter. The size of the tender will be announced on Friday, 19 April – we expect a £400-500mn size, consistent with the usual sizing of mini-tenders as half that of a regular auction. In risk terms this would be worth a modest £2.5- 3.0mn/bp, and should be easily absorbed. We see value in both real yield flatteners and breakeven steepeners, given the shapes of both respective curves (Figure 2), with recent flows in shorter-dated linkers skewed towards selling recently. We think the underperformance of the 10y sector represents mainly profit-taking on long positions. Also, the lack of major MPC remit changes may have led some to reconsider the risks of inflation persistence. In this vein, the latest MPC minutes noted that the majority of the committee are wary of rising medium-term inflation expectations. We think that 10y breakevens still offer structural value versus the MPC 2% CPI target as inflation is likely to remain uncomfortably sticky; but for now, we see much better value in the long end of the curve where the demand base is more structural, albeit recently dormant. FIGURE 1 FIGURE 2 IL27 cheap on curve, IL29 less obvious value IL62 vs IL22 BEI steepeners, RY flatteners attractive 25 160 90 IL27 vs IL22+32 real yield barbell IL29 vs new IL24+34 real yield barbell 140 80 20 70 120 60 15 100 50 80 40 10 60 30 40 5 20 IL62 vs IL22 real yield slope 20 10 IL62 vs IL22 breakeven slope (rhs) 0 0 0 Apr-10 Dec-10 Aug-11 Apr-12 Dec-12 Jan-12 Apr-12 Jul-12 Oct-12 Jan-13 Apr-13 Source: Barclays Research Source: Barclays Research 18 April 2013 52 Barclays | Global Rates Weekly EUROPE: VOLATILITY Conditionally hedged Piyush Goyal Buy low-strike EUR 1y SL 30y CMS floor spread vs low-strike 1y*10y receiver spread (bull +1 212 412 6793 flattener) to hedge a eurozone risk flare. It takes advantage of relative cheapness of vol on firstname.lastname@example.org 30y tails, steepness of rate curve and high convexity adjustment for the current rate levels. Buy EUR 1y*30y versus sell EUR 1y*10y Hitendra Rohra +44 (0)20 7773 4817 Implied vol on mid-expiry 30y tails has come off over the past few months and is now at its email@example.com lowest levels since late 2008 (Figure 1). Lower concerns about a eurozone risk flare, weaker growth expectations and a more benign inflation outlook have all contributed to this sell-off in vol. While these factors still hold, the current low levels offer a great way to position for a hedge against medium-term risks to the eurozone through long vol trades on 30y tails. However, as Figure 2 shows, implied vol on 30y tails is at a premium to realised vol and could still fall somewhat in the near term if rates remain range-bound. Therefore, instead of taking outright positions, we prefer buying vol on 30y tails relative to 10y tails. Specifically, we like initiating EUR 10y-30y conditional bull flatteners (ie, buying 1y*30y receivers versus 1y*10y receivers) as a hedge against a medium-term risk flare. The trade is attractive from both rate and vol perspectives. EUR 1y*30y is currently cheap From a vol perspective, this is attractive for two reasons: 1) implied-to-realised vol is higher compared with EUR 1y*10y on for 10y tails than for 30y tails (Figure 2); 2) even on an absolute basis, 1y*30y is trading an implied-to-realised basis almost at parity to 1y*10y vol. In our view, EUR 1y*30y should trade at a premium to compensate for its tendency to spike more sharply during times of stress (Figure 1). While a part of the justification for higher vol on the 30y compared with the 10y is that the 30y rate has more room to decline in a risk flare, a lot of it is also due to the threat of CVA hedging. The latter, currently on the back-burner, can become an issue if there is a significant risk flare in the eurozone (see CVA Hedging: Here to stay?, 20 July 2012). At current low rates, the curve From a rates perspective, the curve is close to its historically steep levels (Figure 3), implying would bull flatten in a sharp good entry levels for the bull flattener. As the figure shows, earlier when 10y rates were rally in rates high, the level of 10y rates and the 10y-30y curve were inversely related: a rate rally would accompany a curve steepening and vice versa. However, for the past year-and-a-half, 10y rates have been low enough that any violent move has been led by the long end (ie, 30y rates), implying flattening in rate rallies. This is exactly what happened during the risk flare FIGURE 1 FIGURE 2 EUR 1y*30y is at its lowest since late 2008 EUR 1y*10y is extremely rich compared with realised vol 170 1.3 Imp Vol 2 Yr 5 Yr 10 Yr 30 Yr 3m 39 50 56 62 150 1.2 6m 40 52 59 64 130 1y 44 58 65 67 20d rlzd vol 2 Yr 5 Yr 10 Yr 30 Yr 110 1.1 3m 39 39 39 50 90 6m 40 39 40 51 1 1y 40 43 43 53 70 Imp/rlzd 2 Yr 5 Yr 10 Yr 30 Yr 50 0.9 3m 0.98 1.30 1.41 1.24 Apr-08 Apr-09 Apr-10 Apr-11 Apr-12 Apr-13 6m 1.02 1.33 1.47 1.26 EUR 1y*30y (bp/y) EUR 1y*30y/1y*10y (RHS) 1y 1.10 1.35 1.52 1.26 Note: As of 17 April 2013. Source: Barclays Research Note: As 17 April 2013. Source: Barclays Research 18 April 2013 53 Barclays | Global Rates Weekly of mid-2012, when the 10-30y curve flattened sharply as the rates rally was led by the long- end. Consequently, we expect the bull flattener to benefit in a significant move lower in rates and expire without a pay-off if rates move higher. Initiate a low-strike EUR 1y SL Admittedly, the curve is already flatter in forward space, worsening the carry characteristics 30y CMS floor spread vs EUR of the trade. For example, the spot 10-30y is c.69bp, while the 6m forward and 1y forward 1y*10y receiver spread are 60bp and 52bp, respectively. To improve the carry characteristics, we propose striking the two option legs at lower rate levels and replacing the EUR 1y*30y receiver with a EUR 1y SL 30y CMS floor. Specifically, as: • Buy €935mn 1y SL 30y CMS floor spread struck at 2.10 versus 1.60% • Sell €100mn 1y*10y receiver spread struck 1.55% (ATM – 27bp) versus 1.05% As of 18 April 2013, the trade at initiation requires no premium outlay (mid-levels). Net, the trade can be struck at 55bp, which is actually steeper than the forward curve (by 3bp). The trade is carry-neutral In this fashion, the trade is carry-neutral. Therefore, if the level of rate and curve remain unchanged in the next one year, the P&L on the trade is negligible. In contrast, an ATM version of the trade would incur a negative carry of -20cts in three months and -95cts in one year. At the same time, the lower strikes do not hamper the potential gains on the trade. This is because in a risk flare, when the curve actually bull flattens, the 10y and 30y rates will likely rally much below the current spot levels anyway. Using CMS floors instead of It is also important to note how the convexity adjustment in CMS floors improves the trade. receivers on 30y tails improves Essentially, the EUR 1yf 30y CMS rate is higher than the vanilla swap rate by about 6bp due the strike on the trade to convexity adjustment. This means that for a given premium outlay, higher-strike 1y SL 30y CMS floors can be purchased compared with 1y*30y vanilla receiver swaptions. As a result, a premium-neutral 1y expiry 10y-30y bull flattener structure can be struck at a steeper curve level using CMS floors instead of receivers on 30y tails. In the above structure, a trade via vanilla swaptions would be struck at 50bp, which is 5bp worse than the current strike. Therefore, the spot 10-30y curve needs to flatten by a lesser extent to start making profits than would have been required if the trade were done via vanilla swaptions. The chief risk to the trade comes from a sharp bull steepening of the curve. However, as shown in Figure 3, over the past year, the 10-30y curve has steepened to its all-time highs, while the EUR 10y rate has rallied sharply. In other words, the curve has been bull steepening for a while now, and there is limited scope for this dynamic to continue. Besides, the trade being limited-loss should not incur runaway losses even if the curve bull steepens sharply. FIGURE 3 FIGURE 4 EUR 10-30y curve has steepened considerably Receiver skew (100bp wide) is richer for EUR 1y*30y 70 20 5 15 50 4 10 30 3 5 10 0 2 -10 -5 Risk Flare in mid 2012 1 -30 -10 Apr-08 Apr-09 Apr-10 Apr-11 Apr-12 Apr-13 Apr-08 Apr-09 Apr-10 Apr-11 Apr-12 Apr-13 EUR 10y (%) EUR 10y-30y (RHS, bp) 1y*10y 1y*30y Note: As of 17 April 2013. Source: Barclays Research Note: Shows the difference between implied vol 50bp low-strike receiver vs 50bp high-strike payer for EUR 1y*10y and EUR 1y*30y. As of 17 April 2013. Source: Barclays Research 18 April 2013 54 Barclays | Global Rates Weekly JAPAN: RATES STRATEGY How the BoJ can achieve a ¥175trn current account balance Chotaro Morita The BoJ plans to increase its current account balance to ¥175trn by end-2014. Domestic +81 3 4530 1717 banks will have to cut their ¥160trn in JGB holdings to around ¥70trn. For this to be firstname.lastname@example.org accomplished smoothly, yields might have to decline somewhat further. Last week we considered whether life insurers would drastically alter their asset allocation in Reiko Tokukatsu, CFA the wake of the BoJ’s latest easing measures. We concluded that although their activity +81 3 4530 1532 might be affected in some way by the added absorption of over ¥8trn in super-long JGBs on email@example.com the BoJ’s balance sheet, this would not lead necessarily to a commensurate shift to foreign bonds. We argued that they would respond to the decline in JGB yields not through Noriatsu Tanji allocations but by refraining from portfolio duration extension. +81 3 4530 1346 firstname.lastname@example.org Now we turn to the banks. Banks, unlike life insurers, will be directly affected by the increase in the BoJ’s monetary base (ie, the increase in its current account), the main policy target of this month’s easing. The bank decided on 4 April to raise its current account to ¥175trn by Unlike life insurers, banks will end-2014. That would represent an increase of ¥128trn in two years from end-December be directly affected by the BoJ’s 2012. Domestic banks account for nearly 40% of the BoJ’s current account deposits, foreign decision to increase its current banks around 15%, and other reserve deposit holders around 40%. Assuming the increase account in the current account centers on the major domestic banks, the figure for the domestic banking sector would grow by ¥90trn, suggesting a change in the sector’s main asset categories as in Figure 1. We assume the current level of growth continues for deposits, lending and non-JGB securities. Under this assumption, domestic banks would sell ¥90trn in domestic bonds, meeting precisely the current account increase, by end-2014. Domestic banks currently hold ¥160trn in JGBs including TBs, so the calculations imply an average cutback of ¥4trn per month. There may be some question over whether it is appropriate to assume that domestic banks alone will represent ¥90trn of the ¥128trn current account increase. At the same time, we feel it is unrealistic to assume a ¥128trn rise across all formats at the same proportion of present ownership, since it is not clear that foreign banks and other reserve deposit holders have the financial leeway to boost their reserves. FIGURE 1 Projected change in assets at domestic banks JPY trn JPY trn 180 520 Forecasts 160 500 140 480 120 460 100 440 80 420 60 400 40 380 20 360 0 340 05 06 07 08 09 10 11 12 13 14 JGBs (LHS) Non-JGB securities (LHS) Reserves and other deposits (LHS) Loans (RHS) Source: BoJ, Barclays Research 18 April 2013 55 Barclays | Global Rates Weekly If the yield curve does not fall What impact would a ¥90trn cut in JGB holdings have on domestic banks? Assuming a further, the BoJ would be portfolio yield of 0.5%, income gains would drop by ¥360bn annually if the bonds are unable to achieve its current transferred to the BoJ current account, which pays 0.1%. The banks would also earn capital account goal gains on bond sales, of course. Supposing a premium of 0.1% from the BoJ’s expanded JGB purchasing and a duration of around 3y on the bonds sold, the additional capital gains would come to some ¥270bn. Capital gains at that level would hardly be an incentive to sell in order to make up for a constant drop of income gains. If the yield curve does not fall further, the BoJ would be unable to achieve its current account goal. Bank balance sheets should experience a clear change as the BoJ’s operations proceed. However, we believe that this change will require a corresponding drop in yields. At yield levels (10y 0.7%) prior to March, when the markets began pricing in a monetary easing in earnest, we suspect that yields ultimately will have to drop below 0.5%. JGB 10s20s flatteners not reliant on today’s strong auction Today’s strong 20y JGB auction looks to have stopped the rebound steepening of the JGB 10s20s curve after its dive to 65bp following the BoJ’s April MPM. We see more room for the JGB 10s20s curve to flatten. At the same time, we do not think that flattening of 10s20s necessarily needs strong momentum in the 20y segment, as partly provided by today’s auction. We discuss below the reasons why we think a 10s20s flattener position is still attractive to enter around the 90bp level. 1) Fundamentals: Yield curve fundamentals argue that lower front-end rates mean a steeper curve. Indeed, we saw an extreme steepening of the JGB 10s20s early this year, as the 2y JGB yield fell to 4bp (Figure 2) and the front end of the JGB curve inverted as the market started to price in the possibility of a deposit rate cut in addition to tight supply- demand conditions. As the interest on excess reserves (IOER) rate was not changed in the April MPM, the short end of the JGB curve returned to a more normal shape with a positive slope, similar to its profile about a year ago (Figure 3). Based on the current shape of the curve, we think 10s20s can flatten to around 80bp. This view is also related to our thinking that the 10y belly is likely to cheapen, as a retracement takes place from 2012 curve dynamics, when the BoJ’s JGB purchases were concentrated below 3y and the yield curve steepened with the belly richening. 2) OTR premium in 10y: Currently, the JGB 10y is rich largely, which is thought due to the fact that the BoJ’s new rinban operation will allocate its largest share of purchases to the 5- FIGURE 2 FIGURE 3 Recent JGB 10s20s and JGB 2y Short end of JGB curve (%) 105 0.03 0.35 04/17/13 (10s20s=93bp) 100 0.05 0.30 04/10/12 (10s20s=79bp) 95 02/06/13 (10s20s=100bp) 0.25 90 0.07 85 0.20 0.09 80 0.15 75 0.11 0.10 70 0.13 65 0.05 60 0.15 0.00 3-Sep 3-Oct 3-Nov 3-Dec 3-Jan 3-Feb 3-Mar 3-Apr 0.25 1 2 3 4 5 JGB 10s20s (bp, LHS) JGB 2Y (%, RHS) Maturity (years) Source: Barclays Research Source: Barclays Research 18 April 2013 56 Barclays | Global Rates Weekly FIGURE 4 FIGURE 5 Current(OTR) premium JGB 10y relative to 5s10s curve (bp) 10s20s box (bp) 4.0 80 18 3.5 70 16 14 3.0 60 12 2.5 50 10 2.0 40 8 1.5 30 6 3m spread for current vs. ex current 4 1.0 20 when new maturity JB is issued (bp, LHS) 2 0.5 JGB 5s10s (bp, RHS) 10 0 0.0 0 13-Dec 13-Jan 13-Feb 13-Mar 13-Apr May-11 Sep-11 Jan-12 May-12 Sep-12 Jan-13 10s20s box (bp) Source: Barclays Research Source: Barclays Research 10y sector. But richness in the 10y sector appears mainly driven by an increase in the current premium, not by overall richness of the 5-10y sector, in our view. Figure 4 shows that the 3m spread between the on-the-run (OTR) JGB 10y versus the last-on-the-run issue has tightened to 0.5bp. This looks tight, even after considering that JGB 5s10s has also flattened over this time. In terms of the swap spread, the OTR JB328 is 1.6bp richer than the ex-OTR JB327. We admit that the repo is tight for JB328. However, if the repo rate is tighter by 50bp for one month, it can explain the richness in the swap spread by 0.5bp (0.5%/12/0.098 = 0.5bp, 0.098 is the dv01 for 10y JGB). Therefore, we think that if the current premium returns to a more normal level, then the 10y richness is likely to decrease. FIGURE 6 Trade recommendation updates (bp) Year Current end/ Level (incl Weekly Risk Target Entry Entry at last carry) P&L (DV01, (includin date level report or closed (JPY mn) JPY mn) g carry) Stop Horizon Action JGB 10s20s flattener 18-Apr 90.0 n/a 90.5 -2.5 5.0 80.0 100.0 1-3m New 1y OIS pay 13-Jul 6.0 5.5 5.5 0.8 100bn face 8.0 5.0 1y Hold Swap 5s10s steepener 11-Apr 35.0 34.0 37.0 10.0 5.0 50.0 30.0 3m Hold 5x5-10x10 flattener 10-Apr 120.0 116.0 128.0 -40.0 5.0 100.0 130.0 3m Hold Swap 20s30s fly (JL142/JX38) 13-Mar 10.50 10.0 10.0 0.0 6 11.0 30.0 1-3m Remaining spread 20s30s leg Short 1mx20y (atm+20bp) vs 28-Mar -9.0 50.0 22.0 -28.0 10bn face 0 -15 1m Hold Swaption long 1mx10y (atm+11bp) (for 10y) Pay 1yx1y 10-Nov -44.0 -40.0 -40.0 0.0 10 -30.0 -80.0 medium- Hold long Xccy basis Pay 1yx1y 11-Apr -46.0 -47.0 -46.0 5.0 5 -30.0 -80.0 long Hold Pay 4y 7-Mar -51.0 -56.0 -54.0 16.0 5 -40.0 -70.0 medium- Hold long Weekly P&L =-38.7; total P&L since 2013: 344.3; balance sheet=5.9 Note: Current levels based on the absolute maturity to capture rolldown correctly; therefore, it is different from the constant-maturity spread. Source: Barclays Research 18 April 2013 57 Barclays | Global Rates Weekly TRADE PORTFOLIO UPDATE Trade portfolio update Piyush Goyal Since the previous publication (April 11, 2013), the portfolio has lost $0.2mn. It has +1 212 412 6793 increased $10.8mn year-to-date and $59mn since inception.15 email@example.com FIGURE 1 Vivek Shukla Mark-to-market performance of the portfolio – cumulative P&L, $mn +1 212 412 2532 mn firstname.lastname@example.org 70 $59.0 60 50 40 30 20 10 0 Jan-09 Jul-09 Jan-10 Jul-10 Jan-11 Jul-11 Jan-12 Jul-12 Jan-13 Note: As of April 18, 2013. Portfolio stop loss = $10mn. Given this total loss allowed, we allocate $500k as the stop- loss for high-conviction trades and less for low-conviction trades. Source: Barclays Research Total equity = $100mn, stop-loss = $10mn We estimate an initial and variation margin for each derivative trade and a haircut for cash trades. The total of all such margins and haircuts is less than $100mn. In other words, the portfolio is assumed to have $100mn of equity. Thus, all returns are computed on a base of $100mn. Any unused equity is invested in fed funds and assumed to earn the daily funds rate. 15 Since January 2009. 18 April 2013 58 Barclays | Global Rates Weekly TRADE PORTFOLIO New Trades Inception Weights/Notional Levels @ Current Net Change (Gain Total Stop Initial Variation Total Theme Trade Horizon Date Amount Inception Level (+) /Loss (-)) Loss (bp) Margin Margin Margin US TIPS Long 5y5y breakeven 4/18/2013 Dovish Fed $25k dv01 266bp 264bp ($50,000) ($500,000) 1m $350,000 $50,000 $400,000 (Apr 17-Jul22) fwd BE US Treasury Curve looks too steep given 4/18/2013 7s30s curve flattener $50k dv01 172.2 172.3 ($5,000) ($500,000) 1m $500,000 $5,000 $505,000 inflation expectations JGB End of extreme low short end 4/18/2013 JGB 10s20s flattener $40k dv01 90 90.5 ($20,000) ($300,000) 3m $832,000 $20,000 $852,000 means flatter curve US Swaps / Futures Cross mkt Sell 10y France vs eq 4/18/2013 $50k dv01 64.8 64.3 ($25,000) ($300,000) 3m $1,500,000 $25,000 $1,525,000 relative value wgt US and Japan US Options Long 6/14/13-30y receiver 2.725% vs short 4/18/2013 Bull-flattener +20mn: (500 each) 0 -15000 ($20,000) ($500,000) Expiry $1,000,000 $20,000 $1,020,000 3EM3C 99 and 34EMC 98.375 Note: All prices as of April 18, 2013. Source: Barclays Research 18 April 2013 59 Barclays | Global Rates Weekly Trades Outstanding Inception Weights/Notional Levels @ Net Gain (+) Variation Balance Sheet Theme Trade Current Level Stop Loss (bp) Horizon Initial Margin Date Amount Inception /Loss (-) Margin Used US TIPS $6,570,000 Front-end Asset 1/12/2012 Long Jan '14 TIPS ASW $500mn Libor - 13bp Libor -7bp $1,300,000 ($250,000) 1y $2,500,000 ($1,300,000) Swap Tightener Long Core Sell 1% 2y CPI Floors vs. Long CL4 Puts 8/3/2012 ($100mn): +50 ($376,500) ($176,500) $200,000 ($200,000) 1y $800,000 ($200,000) Inflation 65 10/5/2012 Carry Long Jan '15 ASW relative to nominals $100mn 14bp 9bp $120,000 ($300,000) 1y $1,000,000 ($120,000) 10/18/2012 Supply unwind Long Feb 40 Relative ASW $10mn 22bp 31bp ($130,000) ($400,000) 1y $400,000 $130,000 Apr16-Jan22 rel ASWs flattener (20K 3/8/2013 Supply 20k DV01 11bp 11bp ($10,000) ($200,000) 3m $1,250,000 $10,000 DV01) Low realized CPI 3/27/2013 Sell 3y y/y CPI strangle, energy hedged $100mn 1%-3.5% ($1,400,000) ($1,500,000) ($100,000) ($500,000) 3m $2,000,000 $100,000 vol 2/14/2013 Curve trade Long OBL€i18 vs OAT€i40 BE $50k dv01 95bp 81bp ($100,000) ($500,000) 4m $39,000 $100,000 US Treasury $1,545,000 Improvement in 3/14/2013 Long OTR3y versus OIS $100k DV01 8.9 8.94 $26,000 ($500,000) 3m $1,000,000 ($26,000) funding 4/11/2013 Low front end 5s10s Flattener (65% dv01 on 10s) $50k dv01 43.5 39.9 $179,000 ($500,000) Unwound $750,000 ($179,000) risk premium Eurozone Sovereign debt $555,000 10/5/2012 UFR positioning Receive 5y5y/5y10y/5y15y fwd $15k dv01 102bp 99bp $45,000 ($375,000) 3-6m $600,000 ($45,000) US Swaps / Futures $1,420,000 1y1y Libor-OIS tightener against 1y1y 3/8/2013 RV $100k DV01 13.6 12.8 $80,000 ($500,000) 1m $1,500,000 ($80,000) 3s1s widener JPY Swaps $5,402,000 7/13/2012 Short front-end Pay 1y OIS $160k dv01 6bp 6.55bp $88,000 ($350,000) 1y $1,600,000 ($88,000) Swap spread 3/13/2013 term structure is Swap spread 10s20s30s long usd 60 k dv01 23 24 ($60,000) ($450,000) 1y $1,500,000 $60,000 distorted Volatility is high Short 1mx20y (ATM+20bp) payers vs. 3/28/2013 for 20y tail vs. 100mn -5 20 $250,000 ($200,000) 6m $1,200,000 ($250,000) long 1mx10y (ATM+15bp) payers 10y tail 4/11/2013 Swap 5s10s Swap 5s10s steepner $60k dv01 35 37 $120,000 ($450,000) 3m $1,500,000 ($120,000) steepner US Options $10,165,000 7/19/2012 Short vol Short 1y*10y straddles ($20mn) ($1,220,000) ($470,000) $750,000 6m $880,000 ($750,000) 9/7/2012 Short vol Short 1y*10y straddles ($20mn) ($1,242,000) ($617,000) $625,000 6m $880,000 ($625,000) 9/27/2012 Short vol Short 1y*10y straddles ($10mn) ($596,000) ($336,000) $260,000 ($1,000,000) 6m $880,000 ($260,000) 10/5/2012 Short vol Short 1y*10y straddles ($10mn) ($597,000) ($367,000) $230,000 6m $880,000 ($230,000) 11/15/2012 Short vol Short 1y*10y straddles ($10mn) ($570,000) ($390,000) $180,000 Expiry $880,000 ($180,000) Long 1x11 cap -flr straddles 2% vs 6/14/2012 Relative Value $20mn: ($20mn) $1,964,000 $3,104,000 $1,140,000 ($500,000) 1y $880,000 ($1,140,000) 1y*10y straddles 2% Long risk- Long 1y*30y 100bp wide risk-reversal 8/3/2012 $100mn ($450,000) $1,430,000 $1,880,000 ($500,000) 6m $725,000 ($1,880,000) reversal (long recr), delta hedged Long 3y*1y 25bp low-strike recr vs 45bp 8/9/2012 Tactical $200mn:($200mn) 0 $540,000 $540,000 ($500,000) 1y $800,000 ($540,000) high-strike payer Short belly of 2y5y - 2y10y - 2y30y payer +$94mn : ($100mn): 8/16/2012 Relative Value ($300,000) ($20,000) $280,000 ($500,000) 1y $2,200,000 ($280,000) fly +$22mn 1/4/2013 Short vol sell 2y*10y straddles @ 2.65% ($50mn) ($4,105,000) ($4,015,000) $90,000 ($500,000) 6m $2,200,000 ($90,000) 1/17/2013 Sell skew Sell 3y*5y 2.5% payer, delta hedged ($100mn):($40mn) ($2,115,000) ($2,220,000) ($35,000) ($500,000) 1y $1,200,000 $35,000 1/24/2013 long 2ez3 p 99 vs short 4ez3 p 97.75 cheap flattener +2000: (2000) ($650,000) ($575,000) $75,000 ($500,000) Expiry $500,000 ($75,000) short TYM3P 128 vs matched expiry 2/1/2013 put-payer (1000): $129mn ($200,000) ($20,000) $180,000 ($250,000) Expiry $500,000 ($180,000) payer 2/14/2013 Calendar spread long 5y30y p 3.6 vs 1y30y p 3.15 +$50mn: ($50mn) $3,320,000 $3,700,000 $380,000 ($500,000) 1y $1,000,000 ($380,000) Long 1y30y recr, Short 1y30y payer, Pay $100mn: $100mn: 2/28/2013 Fade skew ($430,000) ($440,000) ($10,000) ($500,000) 6m $725,000 $10,000 fix 1y30y $50mn 4/11/2013 Steeper Vol Long 3x13 cfs vs 3y10y swaption +$50mn: -$50mn $4,650,000 $4,590,000 ($500,000) 1y $1,600,000 $0 $0 Surface straddle 18 April 2013 60 Barclays | Global Rates Weekly Trades Outstanding (continued) Inception Weights/Notional Levels @ Net Gain (+) Variation Balance Sheet Theme Trade Current Level Stop Loss (bp) Horizon Initial Margin Date Amount Inception /Loss (-) Margin Used EUR Options $4,191,000 6/1/2012 Short vol EUR 1x2 1y5y 1.25 vs 1 recr ladder EUR 20mn: (40mn) 0 $50,000 $50,000 ($250,000) 6m $744,000 ($50,000) EUR 100mn: 9/7/2012 Short vol EUR 1x2 1y5y 1.15 vs 0.9 recr ladder 0 $250,000 $250,000 ($250,000) 6m $1,692,000 ($250,000) (200mn) Long EUR 1y*10y 1x2 payer spread (2.2 (EUR 50mn): EUR 10/5/2012 Long vol 0 ($90,000) ($90,000) ($500,000) 6m $550,000 $90,000 vs 2.6) 100mn EUR 100mn: 2/1/2013 Low for long 6m5y 1x2x1 1.3 - 0.95 - 0.7 recr fly $650,000 $1,065,000 $415,000 ($250,000) Expiry $500,000 ($415,000) (200mn): 100mn Short 3m*4y GBP straddles vs 30bp high- +GBP 100mn: gbp 3/8/2013 RV ($1,040,000) ($1,170,000) ($130,000) ($500,000) Expiry $1,200,000 $130,000 strike 3m*4y payer 100mn Cross-currency $980,000 10/11/2012 Carry Pay 1yx1y Xccy basis $40k dv01 -53.5 -40 $540,000 ($400,000) 1y $400,000 ($540,000) Paying demand 3/8/2013 is larger in 4-5y pay 4y Xccy basis usd 40 k dv01 -51 -54 ($120,000) ($400,000) 3m-6m $1,000,000 $120,000 basis US BMA $2,130,000 Long 3m1y BMA ratio vs short 3y1y Sell Front-end 1/12/2012 ratio; 3m1y matured on 4/12 at 1y ratio $200mn : ($200mn) 54, 84 50, 69 $150,000 ($250,000) 1y $800,000 ($150,000) Ratios = 50, implying p&l -$42k Sell Front-end 5/10/2012 Short 3y ratio $200mn 65.375 60 $80,000 ($250,000) 1y $800,000 ($80,000) Ratios Sell Front-end 6/7/2012 Short 3y ratio $200mn 66.75 62 $40,000 ($250,000) 1y $800,000 ($40,000) Ratios Cash 4/11/2013 4/18/2013 Cash Used as Collateral/ Haircut $34,225,000 $37,399,000 Fed Funds (residual cash) $84,764,185 $81,430,657 Return on Fed Funds $104,369 $106,841 Return on trades $18,722,816 Total $118,829,657 Note: All prices as of April 18, 2013. Source: Barclays Research 18 April 2013 61 Barclays | Global Rates Weekly Trades Unwound Inception Unwound Weights/Notional Levels @ Levels @ Net Change (Gain Total Stop Date Date Theme Trade Amount Inception Unwind (+) /Loss (-)) Loss (bp) Horizon US TIPS 9/29/2011 1/12/2012 Front-end Asset Swap Tightener Long Jan '12 TIPS ASW $500mn Libor - Libor - 34bp $170,000 ($250,000) Unwound 15.75bp 1/6/2012 1/19/2012 Supply Trade Sell Apr '16 - Jul '21 - Apr '28 $25k dv01 20bp 21bp $25,000 ($500,000) Unwound 10/20/2011 2/23/2012 Relative Value 10y-30y breakeven steepener $20k dv01 21.5bp 5bp ($260,000) ($250,000) Stop-out 3/9/2012 3/28/2012 Relative Value Long Apr '14 - Apr '15 breakeven $30k dv01 165bp 185bp $500,000 $200,000 Unwound 1/27/2012 4/6/2012 Dovish Fed Long 5y5y breakevens $20k dv01 232bp 251bp $250,000 $300,000 Unwound 6/16/2011 4/19/2012 Eurozone contagion Long the belly Jan '16 - Apr '16 - July '16 real $20k dv01 5.5bp 7bp ($20,000) ($75,000) Unwound yield fly 8/5/2011 5/24/2012 Long TIPS Long July'12 TIPS energy hedged; bought 20 12k dv01, Short 40 -60.5bp, -253bp, 288 $865,000 $0 Unwound XBM2 on 4/6/12 for 327.66 XBH2 (XBM2) 267.76 3/29/2012 6/1/2012 Relative Value Long TII Jan 22 Relative Asset Swap 45k dv01 33bp 31bp $10,000 ($500,000) Unwound 4/19/2012 7/6/2012 Eurozone contagion Long Apr '17 vs Jan '17 $50k dv01 6bp -5bp $550,000 ($500,000) Unwound 6/1/2012 7/19/2012 Inflation risk premium Long 10y10y vs 5y5y BE $15k dv01 -19bp -8bp $80,000 ($150,000) Unwound 6/28/2012 7/19/2012 Supply Short belly of Jan '18 - Jan '21 - Jan 26 real $20k dv01 -12bp -9bp $90,000 ($250,000) Unwound yield fly 7/19/2012 8/30/2012 Flattener Jan '14 - Apr '17 breakeven flattener ; energy $15k dv01; (30 67bp, 94.44 52bp, 96.35 ($100,000) ($250,000) Unwound hedged contracts) 8/16/2012 10/18/2012 Supply Trade Apr '17 - Feb '42 breakeven flattener $15k dv01 39bp 35bp $12,000 ($250,000) Unwound 11/2/2011 10/25/2012 Normalization Long Apr' 13 TIPS vs. sell 2% CPI cap and sell $30k dv01; 45 -103bp, - -107bp, 4bp, $1,100,000 $700,000 Unwound XBH3 contracts 36bp, 249.83 275 9/20/2012 11/8/2012 Dovish Fed Receive 2y forward 2y break-even (Jan '14 vs $20k dv01 228bp 215bp ($250,000) ($500,000) Unwound Jan '16) 8/17/2012 11/15/2012 Sell Deflation Floor Long Jan' 17 vs. Apr '17 $50k dv01 0bp -1bp ($50,000) ($100,000) Unwound 9/7/2012 11/29/2012 Carry trade Buying German I/L Apr 2016, asset swapped $100mn 32.3bp 16bp $690,000 ($500,000) Unwound into USD 11/8/2012 1/4/2013 Fiscal Cliff Short Jul17 Breakeven 25k/$56mn 215bp 228bp ($292,344) ($500,000) Unwound 11/29/2012 1/4/2013 Supply Fly: 5s Sell the belly of Jan15-Jan17-Jan19 RY Fly $15K DV01 -21 -18 $60,000 ($200,000) Unwound 11/29/2012 1/4/2013 Risk Premium 10s30s BE steepener ahead of Dec. FOMC 15K DV01 2bp 6.2bp $70,000 ($250,000) Unwound 1/4/2013 2/7/2013 Concession unwind Long Jan15-Apr17-Jan19 RY Fly $15k dv01 -10 0 ($175,000) ($150,000) Stop-out 2/21/2013 1/10/2013 Front-end Underpriced Long $250mn TIIJan14s hedged with sell 100 $250mn -134bp, -226bp, 93.5 $1,003,000 ($600,000) Unwound CLZ4 91.74 1/17/2013 3/14/2013 Long the belly and fade the roll Jan19-Jul22-Jan25 real yield fly $20k dv01 32bp 37bp ($10,000) ($160,000) Unwound UK Inflation 6/1/2012 6/14/2012 Macro Long IL20 vs. pay match-maturity swap GBP 7mn 280bp 262bp ($515,000) ($500,000) Stop-out 6/22/2012 7/6/2012 Macro Buy IL29 Breakeven (vs. UKT 4.75%) GBP 13mn: (GBP 262bp 263bp $35,700 ($500,000) Unwound 13.3mn)/ $35.7k dv01 6/29/2012 7/26/2012 Real Yield curve flattener Long IL32 vs IL17 $45k dv01 126bp 145bp ($720,000) ($500,000) Stop-out 7/13/2012 9/28/2012 Relative Value Sell IL17 vs IL16, IL22 real yield fly $37.5k dv01 -24bp -24bp $150,000 ($300,000) Unwound 9/14/2012 10/18/2012 Cheap forward real rates Receive 20y20y forward RPI real rate swap $17.2k dv01 36bp 42bp ($103,000) ($500,000) Unwound 9/28/2012 11/29/2012 Real yield flattener IL22 - IL42 real yield flattener $25k dv01 97bp 100bp ($175,000) ($500,000) Unwound 11/16/2012 1/10/2013 UK Linker RV Sell IL22 vs IL17+32 RY barbell 62.5k DV01 0bp 1bp $125,000 ($500,000) Unwound 12/12/2012 1/10/2013 10y UK breakevens rich Sell IL22 Breakeven $25k DV01 250 293 ($1,075,000) ($1,000,000) Stop-out 1/17/2013 2/7/2013 UK 5y5y fwd real yield rich Sell IL22 into IL17 cash for cash $30k DV01 -20 -15 $150,000 ($500,000) 3m 2/21/2013 2/1/2013 UK Linker RV IL24 vs IL17+IL34 real yield barbell $125k dv01 32 37 ($500,000) ($500,000) Stop-out 2/14/2013 3/8/2013 Real Yield IL32-IL55 real yield flatteners $50k dv01 26bp 33bp ($500,000) ($500,000) Stop-out Source: Barclays Research 18 April 2013 62 Barclays | Global Rates Weekly Trades Unwound (continued) Inception Unwound Weights/Notional Levels @ Levels @ Net Change (Gain Total Stop Date Date Theme Trade Amount Inception Unwind (+) /Loss (-)) Loss (bp) Horizon EUR Inflation 6/1/2012 6/22/2012 Macro Receive 5y Euro HICPx Inflation EUR 25mn 1.35% 1.48% $390,000 $390,000 Stop-out 6/29/2012 7/26/2012 Relative Value Sell Bund i23 vs. OBL i18 breakeven $27.2k dv01 47bp 50bp ($106,000) ($300,000) Unwound 8/10/2012 9/7/2012 Short inflation Sell 10y FRCPIx ZC Inflation $37k dv01 2.32% 2.4% $50,000 $100,000 Stop-out 9/14/2012 9/28/2012 Relative Value Sell 10y FRCPIx vs. Euro HICPx $39.5k dv01 27bp 37bp ($388,000) ($400,000) Unwound 11/29/2012 12/14/2012 Real Yield Short BTPi16/19 fwd real yield $5.4k DV01 346bp 360bp $75,000 ($250,000) Stop-out 1/17/2013 2/28/2013 Curve Buy BTP€i41 versus BTP€i21 EUR 10mn 21s vs EUR 77bp 36bp $411,000 $75,000 Unwound 4.5mn 41s ($11k dv01) 1/4/2013 3/14/2013 EUR21.3mn 23s vs Unwound Relative value Buy OATi23 versus OATi22 3.5bp 1bp $105,000 ($75,000) 25mn 22s ($35k dv01) US Treasury 1/12/2012 1/26/2012 Fed-on-hold Long 2y-5y-10y treasury fly $50kdv01 -49.5bp -66.5bp $850,000 ($500,000) Unwound 1/19/2012 1/26/2012 Dovish Fed 10y-30y tsy curve steepener $50k dv01 106.25 bp 117.25 bp $550,000 ($500,000) Unwound 2/9/2012 2/23/2012 Unwind of auction concession 7y-30y tsy curve flattener $50k dv01 177.75bp 174.75bp $150,000 ($500,000) Unwound 3/1/2012 3/9/2012 Bond auction concession 10y-30y tsy curve steepener $50k dv01 111.5bp 116.25bp $237,500 ($500,000) Unwound 3/29/2012 4/6/2012 Bond auction concession 10y-30y tsy curve steepener $50k dv01 111.5bp 117.5bp $305,000 ($500,000) Unwound 3/1/2012 4/19/2012 Increase in odds of QE3 Long 5y-10y-30y fly $50k dv01 1.75bp -1.7bp $172,500 ($500,000) Unwound 3/1/2012 4/19/2012 Fading 7yr Short 5yr - 7yr - 10yr $50k dv01 -4.5bp -11.8bp ($365,000) ($500,000) Unwound 3/16/2012 4/19/2012 Dovish Fed Long ct2 75k dv01 36.9bp 26.7bp $865,000 ($500,000) Unwound 4/20/2012 5/17/2012 Low front-end term premium Vol weighted 2y - 3y steepener $80k dv01: ($50k -2bp -7.5bp ($270,000) ($500,000) Unwound dv01) 4/26/2012 5/17/2012 Bond auction concession 10y-30y tsy curve steepener $50k dv01 117.2bp 112.2bp ($250,000) ($500,000) Unwound 5/10/2012 6/1/2012 Fading 7yr Short 5yr - 7yr - 10yr $50k dv01 -12.2bp -18.2bp ($325,000) ($300,000) Stop-out 5/10/2012 6/14/2012 Relative Value Short HC Nov '15 Ps vs. HC Feb '15 Ps $50k dv01 6.25bp 8bp $87,500 ($300,000) Unwound 6/7/2012 7/6/2012 Fading Operation Twist Long 3% Sep '16 vs. OIS $50k dv01 16.1bp 13.5bp $130,000 ($250,000) Unwound 6/28/2012 7/19/2012 Dovish Fed/ Relative Value Long 9.25% Feb '16s $40k dv01 0.525% 0.40% $500,000 ($400,000) Unwound 6/14/2012 7/26/2012 Macro Long 10y treasury $25k dv01 1.635% 1.425% $570,000 ($300,000) Unwound 7/12/2012 7/26/2012 Flattener 10y-30y tsy curve flattener $50k dv01 108.4bp 106bp $115,000 ($300,000) Unwound 7/26/2012 8/9/2012 Macro Long 7y tsy $50k dv01 0.94bp 1.045bp ($525,000) ($500,000) Stop-out 9/7/2012 9/13/2012 Steepener 7y-30y steepener $50k dv01 171.5bp 184bp $625,000 ($500,000) Unwound 8/10/2012 9/20/2012 Fed on hold Long 5y $25k dv01 0.694% 0.68% $50,000 ($500,000) Unwound 10/11/2012 11/29/2012 Flattener 7y-30y flattener $50k dv01 177bp 178bp ($45,000) ($500,000) Unwound 9/20/2012 1/10/2013 Low funding rates Long 3y $50k dv01 35.5bp 33.2bp $219,000 ($500,000) Unwound 11/15/2012 1/10/2013 Liquidity Premium Long OTR 10s vs Old 10s $100K DV01 3.8bp 4.7p ($89,000) ($500,000) Unwound 12/13/2012 1/10/2013 Long end Fed purchases 7s30s Tsy curve flattener 25k dv01 175.8 175.7 $5,000 ($500,000) Unwound 2/14/2013 1/24/2013 bond refunding steepener 10s30s $50k dv01 119.7 120.3 $28,000 ($500,000) Unwound 1/10/2013 2/28/2013 Long front end Long 4y $50k DV01 59.4bp 53.3bp $378,000 ($500,000) Unwound 1/17/2013 2/28/2013 Lower growth, higher fiscal risk Long 5s10s30s fly $50k DV01 -9.6bp -12.5bp $194,370 ($500,000) Unwound premium 2/1/2013 2/28/2013 Easy Fed, Fading richness of 7s Long 2s5s7s gross fly $50k DV01 8.6bp 3.6bp $243,700 ($500,000) Unwound 2/14/2013 3/14/2013 Not enough liq premium in OTR Long Feb23s vs Nov22s $100k dv01 2.95bp 2.76 $53,000 ($500,000) Unwound 9/20/2012 3/27/2013 QE underpriced Long 10y $25k dv01 1.78% 1.76% $260,000 ($500,000) Unwound 2/14/2013 3/27/2013 sequester/post auction dynamics 10s30s curve flattener $50k dv01 116.65 124.2 ($414,000) ($500,000) Unwound 2/28/2013 4/5/2013 Fade the curve steepness 7s10s flattener $100k DV01 62.8 61.39 $175,000 ($500,000) Unwound 3/27/2013 4/5/2013 Risk aversion 7s30s flattener $50k DV01 186 175.9 $530,000 ($500,000) Unwound Source: Barclays Research 18 April 2013 63 Barclays | Global Rates Weekly Trades Unwound (continued) Inception Unwound Weights/Notional Levels @ Levels @ Net Change (Gain Total Stop Date Date Theme Trade Amount Inception Unwind (+) /Loss (-)) Loss (bp) Horizon JGB 6/8/2012 6/21/2012 Fading monetary policy easing 5y-10y flattener $125k dv01 66bp 61.8bp $525,000 ($250,000) Unwound 6/8/2012 6/21/2012 relative value 10y-20y flattener $125k dv01 81bp 84.4bp ($425,000) ($375,000) Stop-out 10/17/2012 10/25/2012 Bear flattener JGB 10y-20y bear flattener $50k dv01 90.5bp 91.5 ($50,000) ($150,000) Unwound 11/8/2012 12/13/2012 Auction Concession Unwind JGB 30s40s flattener 60k DV01 19 16.4 $156,000 ($150,000) Unwound 1/29/2013 Tactical JGB 10s12s flattener $60k dv01 30.2 29.8 $24,000 ($150,000) Unwound 2/21/2013 3/7/2013 JGB 20s30s steepener JGB 20s30s steepener $120K 17.5 14 ($400,000) ($400,000) Stop-out 3/28/2013 Still flattening room for JGB 3/21/2013 JGB 10s20s flattener 60 95 91 $240,000 ($150,000) Unwound 10s20s Eurozone Sovereign debt 4/19/2012 5/17/2012 Eurozone contagion Short FRTR Apr '20 vs. 50% RFGB Apr '20 and $35k dv01 -42bp -58.5bp $542,500 ($250,000) Unwound 50% RAGB Jul '20 4/27/2012 5/25/2012 Italy vs. Spain Short BTPS 3.75% Mar 21 vs SPGB 5.5% Apr $12.5k dv01 -37bp -70bp ($425,000) ($350,000) Stop-out 21 9/7/2012 11/21/2012 Front end periphery convergence Long SPGB 4.75% Jul 14 vs BTP 4.25% Jul 14 $10k dv01 60bp 102bp ($420,000) ($350,000) Stop-out 6/15/2012 2/28/2013 Relative Value Short Bobl ASW vs. EONIA long Schatz ASW $15k dv01 -48bp -24bp ($360,000) ($350,000) Stop-out vs. libor 1/10/2013 3/8/2013 Fundamental cheapness Long 5y5y fwd EUR ASW $30k dv01 -4 15 $570,000 ($450,000) Unwound US Swaps / Futures 1/19/2012 2/23/2012 Calendar roll USH2 invoice spread widener $100k dv01 0.45bp -2.4bp ($285,000) ($500,000) Unwound 1/6/2012 3/15/2012 Eurozone Contagion Short EDU2 Long EDU4 2000 contracts 47.75bp 54.75bp ($525,000) ($500,000) Stop-out 3/16/2012 4/6/2012 Spread widener FV invoice spread widener $50k dv01 17.75bp 28.25bp $525,000 ($500,000) Unwound 4/19/2012 5/24/2012 Relative Value Sell TYM2 Invoice spread vs. 1/3rd dv01 1y1y $50k DV01 12.15bp 19.7bp ($377,500) ($500,000) Unwound libor-OIS 5/18/2012 5/30/2012 Calendar Roll Long TUU2 Short TUM2 2000:(2000) 2.875 (ticks) 1.75 (ticks) $70,313 ($250,000) Unwound 5/18/2012 5/30/2012 Calendar Roll Long TYU2 Short TYM2 2000:(2000) 31.75 (ticks) 30 (ticks) $109,375 ($250,000) Unwound 1/6/2012 6/21/2012 Issuance Sell 30y spreads $50k dv01 -31bp -24bp ($350,000) ($250,000) Stop-out 4/6/2012 7/12/2012 Front-end spd widener March '14 FRA-ois (USFOSC8) widener $50k dv01 38.5bp 33.5bp ($250,000) ($250,000) Stop-out 7/19/2012 8/3/2012 Flattener 4y1y vs 1y1y flattener $50k dv01 109.5bp 120bp ($525,000) ($500,000) Stop-out 6/7/2012 8/30/2012 Relative Value Short 5y - US - 30y spread $50k dv01 13.15bp 8.4bp $237,500 ($250,000) Unwound 10/11/2012 10/18/2012 Long duration Receive 3y1y $50k dv01 103.25bp 116.9bp ($500,000) ($500,000) Stop-out 9/13/2012 11/23/2012 Relative Value Long 11/23/12 -> Aug '19 vs short TYZ2 C "$130mn: (1000) 0 $684,000 $684,000 ($500,000) Expired 133 9/7/2012 1/4/2013 Spread Curve Flattener 5y - 10y spread curve flattener $100k dv01 -7.2bp 263bp ($130,000) ($500,000) Unwound 9/13/2012 1/4/2013 Macro 30y Spread widener $50k dv01 -22.5bp -20.5bp $100,000 ($500,000) Unwound 10/11/2012 1/4/2013 Flattener Pay 3y1y Rec 5y9y $50k dv01 237bp 263bp ($500,000) ($500,000) Stop-out 2/14/2013 2/1/2013 7y auction concession unwind Long dv01 weighted OTR7y TYH3 basis $100mn/820 TYH3 cts -260.1 -260.5 ($9,000) ($500,000) Unwound 2/14/2013 4/5/2013 UK v US steepener UK 10s-30s steepener against US flattener 100K DV01 6bp 16bp $1,000,000 ($500,000) Unwound Source: Barclays Research 18 April 2013 64 Barclays | Global Rates Weekly Trades Unwound (continued) Inception Unwound Weights/Notional Levels @ Levels @ Net Change (Gain Total Stop Date Date Theme Trade Amount Inception Unwind (+) /Loss (-)) Loss (bp) Horizon JPY Swaps 2/14/2013 1/24/2013 Pause in one-way steepening in Long 1yx10y straddle vs. 1yx20y straddle $120mn -60 -43 $204,000 ($250,000) Unwound long end 2/5/2013 Volatility and rates are both high Short 1yx8y ATM+10bp payers USD60mn -92 -90 $12,000 ($150,000) Unwound 5/18/2012 5/25/2012 Fade excessive bull-flattening 6x2-8x2 steepener $120k dv01 48.75bp 50.5bp $210,000 ($450,000) Unwound 5/24/2012 6/7/2012 Calendar Roll Short calendar spread (JBM2-JBU2) 100 contracts 20 ticks 18 ticks $26,000 ($12,500) Unwound 5/18/2012 6/21/2012 Good carry trade with stable front 6mx1y pay $60k dv01 -53bp -44bp $540,000 ($420,000) Unwound end. 6/14/2012 6/21/2012 Tactical 5y5y-10y10y flattener $60k dv01 99.5bp 104bp ($270,000) ($120,000) Stop-out 6/7/2012 7/6/2012 Long spreads Long 30y swap spread $30k dv01 19bp 18.8bp ($6,000) ($120,000) Unwound 5/22/2012 7/20/2012 Long spreads Long 20Y swap spread $60kdv01 11.4 12.3 ($54,000) ($400,000) Unwound 7/13/2012 7/20/2012 Long spreads Reestablish 30Y swap spread $30k dv01 21.2bp 20bp $36,000 ($54,000) Unwound 7/19/2012 7/25/2012 Long spreads Short 20y swap spread vs 10y10y (1:3 ) $60kdv01 84.4bp 86.9bp $150,000 ($200,000) Unwound 6/14/2012 7/26/2012 Tactical Short 7Y (future) swap spread $60k dv01 -9.1bp -8.5bp $36,000 ($144,000) Unwound 7/6/2012 7/26/2012 Carry Long 6Y swap spread $120kdv01 -16.1bp -12.7bp ($408,000) ($300,000) Stop-out 8/3/2012 8/16/2012 Tactical pay 1y1y swap $75k dv01 -26.1bp -29bp $218,000 ($300,000) Unwound 8/9/2012 8/16/2012 Relative Value long 30y swap spread and receive 20yx10y at $100k dv01 123.13 123 $13,000 ($300,000) Unwound 2:1 7/20/2012 9/14/2012 Carry Receive USD/JPY Xccy basis 6yx2y vs. 4yx1y $60kdv01 7 -11.7 $1,120,000 ($360,000) Unwound 8/3/2012 9/28/2012 Tactical long 8y swap spread $120k dv01 -5.75bp -5.2bp ($62,000) ($300,000) Unwound 8/31/2012 10/5/2012 Relative Value short swap 7s10s20s $100k dv01 -42.125 -44.5 ($238,000) ($300,000) Unwound 9/14/2012 10/5/2012 Tactical pay 3yx5y Xccy basis $40k dv01 -83 -57 $1,040,000 ($400,000) Unwound 9/21/2012 10/11/2012 Limited rally in 10s Sell 1mx10y receiver spread $40mn -28cts -32cts ($16,000) ($100,000) Unwound 10/5/2012 11/8/2012 Tactical 10y10y - 20y10y steepener $60k dv01 -26bp -18bp $48,000 ($50,000) Unwound 11/7/2012 11/15/2012 Relative Value Swaps 5s10s flattener 120k DV01 44.75 42 $33,000 ($250,000) Unwound 8/31/2012 11/29/2012 Macro long 30y swap spread $60k dv01 21 15.3 $342,000 ($200,000) Unwound 11/7/2012 12/21/2012 Rate Directionality Pay 5y vs 3m Libor 60k DV01 21.5 21 ($3,000) ($150,000) Unwound 9/28/2012 12/28/2012 hedge to overall portfolio Long receiver fly 3mx10y (0.72-0.77-0.82%) $40mn 8cts 17cts $36,000 ($50,000) Expired 11/1/2012 1/7/2013 Carry Trade Sell 1y1y ATM receiver USD40mn -6 -8 ($8,000) ($100,000) Unwound 12/7/2012 1/7/2013 Hedge to short 1y1y long 30y swap spread $60k dv01 14.3 -1 $918,000 ($200,000) Unwound 1/7/2013 1/17/2013 5s swap spread too rich 5s7s box spread $120k dv01 7.5 6.9 ($60,000) ($400,000) Unwound 1/23/2013 2/28/2013 Pause in easing speculaion Swap 20s30s box (20y long) $80k dv01 -7 -4.6 $232,000 ($500,000) Unwound 1/17/2013 3/8/2013 Carry short 5x5, long 10x10, short 20x10 $60k dv01 -149 -140 $360,000 ($250,000) Unwound 2/5/2013 3/8/2013 Volatility Short 1yx8y ATM+10bp payers $60mn -92 -62 $180,000 ($150,000) Unwound 10/18/2012 3/21/2013 Widener 2y Tibor 6v1 widener $120k dv01 19 19.2 $24,000 ($400,000) Unwound 2/7/2013 3/28/2013 Volatility and Rates Short 1yx8y ATM straddle with 5x2 long $30mn -235 -202 $99,000 ($150,000) Unwound Source: Barclays Research 18 April 2013 65 Barclays | Global Rates Weekly Trades Unwound (continued) Inception Unwound Weights/Notional Levels @ Levels @ Net Change (Gain Total Stop Date Date Theme Trade Amount Inception Unwind (+) /Loss (-)) Loss (bp) Horizon US Options 10/7/2011 1/9/2012 Sell GBP Gamma Sell 3m5y straddles GBP 25mn ($650,000) ($491,000) $159,000 ($500,000) Expired 10/7/2011 1/9/2012 Sell EUR Gamma Sell EUR 3m2y straddles EUR 25mn ($225,000) ($175,000) $50,000 ($100,000) Expired 10/20/2011 1/20/2012 Sell US Gamma Sell 3m*10y straddles $10mn ($420,000) ($285,000) $135,000 ($125,000) Expired 1/19/2012 1/26/2012 Steepener Long 4m30y payer spread (1.8 vs 2.1) and $100mn: (2400) ($100,000) $700,000 $800,000 ($500,000) Unwound short TYM2 puts @ 128.5 11/18/2010 2/3/2012 Fed on hold long 1y1y collar $300mm:($300mm) $1,546,000 $2,069,000 $523,000 ($500,000) Unwound 3/10/2011 2/3/2012 Fed on hold 1y5y covered call $10mn ($150,000) $108,000 $258,000 ($250,000) Unwound 5/19/2011 2/3/2012 Hedge to Fed on hold Long 1y*2y payer spread (atm vs 100bp high- $240mn: ($100mn) 0 $100,000 $100,000 ($250,000) Unwound strike) and sell high-strike 1y*5y payer; unwound the long 1y2y payer on 9/2/11 7/8/2011 2/3/2012 GBP options 1y1y vs 1y5y bear flat; unwound the long 1y1y 470mn: (100mn) ($125,000) $517,000 $642,000 ($250,000) Unwound payer on Sep 22 '11 11/4/2011 2/6/2012 Sell US Gamma Sell 3m*10y straddles $10mn ($431,000) ($233,000) $198,000 ($250,000) Expired 2/4/2011 2/9/2012 Fed on hold Rec 1y1y and sell 25bp low 1y*1y recr $100mn ($225,000) $249,000 $474,000 ($250,000) Expired 11/17/2011 2/17/2012 Sell US Gamma Sell 3m*10y straddles $10mn ($440,000) ($134,000) $306,000 ($250,000) Expired 12/15/2011 2/24/2012 Sell US Gamma Sell TYH2 straddles 100 ($284,375) ($64,375) $220,000 ($100,000) Expired 12/2/2011 3/2/2012 Sell US Gamma Sell 3m*10y straddles $10mn ($393,000) ($259,000) $134,000 ($250,000) Expired 11/10/2011 3/9/2012 Eurozone contagion 6m 10y-30y CMS Bull Flattener $50k dv01 $225,000 ($350,000) ($575,000) ($500,000) Stop-out 2/3/2012 3/22/2012 Steepener 5y*2y vs 5y*30y bear steepener ($200mn):$20mn ($90,000) ($630,000) ($540,000) ($500,000) Stop-out 2/16/2012 3/22/2012 Steepener 2y*10y vs 2y*30y bear steepener ($112.5.mn): $50mn ($20,000) ($560,000) ($540,000) ($500,000) Stop-out 2/9/2012 3/23/2012 Sell US Gamma Sell TYJ2 straddles 100 ($201,000) ($150,000) $51,000 ($100,000) Expired 2/23/2012 4/20/2012 Sell US Gamma Sell TYK2 straddles 100 ($232,813) ($67,813) $165,000 ($100,000) Expired 3/1/2012 4/20/2012 Sell US Gamma Sell TYK2 straddles 100 ($212,500) ($117,500) $95,000 ($100,000) Expired 11/4/2011 5/4/2012 Eurozone contagion Long 6m1y payer spread vs. short 6m5y payer $485mn: ($100mn) 0 0 $0 ($250,000) Expired spread 2/9/2012 5/9/2012 Eurozone contagion Long 3m1y payer spread vs. short 3m7y payer $490mn: ($100mn) ($110,000) $270,000 $380,000 ($500,000) Expired spread 11/17/2011 5/11/2012 Eurozone Contagion Buy 6m2y payr spd vs 6m10y payr spd EUR 225mn: 0 0 $0 ($500,000) Unwound (EUR50mn) 2/23/2012 5/11/2012 Higher rates Buy 3m*10y payer 2.25% KO 2.75% $100mn $570,000 $40,000 ($530,000) ($500,000) Unwound 3/9/2012 5/17/2012 Sell US Gamma Sell TYM2 straddles 100 ($245,313) ($320,313) ($75,000) ($100,000) Unwound 3/22/2012 5/17/2012 Rangebound rates Long 1y30y @ 3.1% vs short 3m30y @ 3.1% $20mn:($20mn) $1,540,000 $1,035,000 ($505,000) ($500,000) Stop-out 3/29/2012 5/17/2012 Sell US Gamma Sell TYM2 straddles 200 ($440,625) ($750,625) ($310,000) ($100,000) Unwound 4/6/2012 5/17/2012 Sell US Gamma Sell TYM2 straddles 100 ($190,625) ($280,625) ($90,000) ($100,000) Unwound 4/12/2012 5/17/2012 Sell US Gamma Sell TYM2 straddles 100 ($193,750) ($273,750) ($80,000) ($100,000) Unwound 4/19/2012 5/17/2012 Sell US Gamma Sell TYN2 straddles 100 ($221,875) ($266,875) ($45,000) ($100,000) Unwound 4/26/2012 5/17/2012 Sell US Gamma Sell TYN2 straddles 100 ($192,188) ($192,188) ($30,000) ($100,000) Unwound 1/26/2012 5/25/2012 Cross -currency Long EUR 4m*7y vs TYM2 straddles EUR 10mn: (100) ($20,000) $160,000 $180,000 ($500,000) Expired 4/13/2012 6/1/2012 Higher rates Buy EUR 3m*5y payer 1.55% KO 2.05% EUR 100mn $560,000 $50,000 ($510,000) ($500,000) Stop-out 1/6/2012 6/21/2012 Eurozone Contagion Long 6m1y payer spread vs. short 6m7y payer $490mn: ($100mn) ($340,000) 0 $340,000 ($500,000) Unwound spread 2/17/2011 7/12/2012 Relative Value Buy 3y*10y payer @ 5% Sell 3y SL 10y CMS $50mn: ($350mn) ($100,000) 0 $100,000 ($500,000) Unwound Cap @ 5% 8/18/2011 7/12/2012 Eurozone contagion Long 1y2y payer spread vs 1y10y payer spread $90mn: $20mn ($130,000) $10,000 $140,000 ($500,000) Unwound 2/3/2012 7/12/2012 Eurozone contagion Long 1y1y payer spread (0.55% vs 1.05%) $100mn $105,000 $55,000 ($50,000) ($50,000) Unwound 6/15/2012 7/12/2012 Eurozone contagion 1m*10y vs 1m*30y bull flattener ($116.5mn):$50mn 0 ($510,000) ($510,000) ($500,000) Stop-out 4/20/2012 7/26/2012 Long EUR vs. US gamma Long EUR 4m*7y std vs. TYU2 std EUR 50mn: (750) ($945,000) ($1,495,000) ($550,000) ($500,000) Stop-out 4/26/2012 7/26/2012 Relative Value Long 1y5y straddles vs 3EM2 straddles; sold $100mn:(2000) ($250,000) ($10,000) $240,000 ($500,000) Unwound 3EU2 straddles for $1.75mn 6/28/2012 7/26/2012 Short vol Short 1y*10y straddles ($20mn) ($1,260,000) ($1,560,000) ($300,000) ($250,000) Stop-out 5/18/2012 8/3/2012 Eurozone contagion Buy 1y*30y flr 2.25% vs 3m*30y flr 2.25% $200mn: ($200mn) $350,000 $1,090,000 $740,000 ($250,000) Unwound Source: Barclays Research 18 April 2013 66 Barclays | Global Rates Weekly Trades Unwound (continued) Inception Unwound Weights/Notional Levels @ Levels @ Net Change (Gain Total Stop Date Date Theme Trade Amount Inception Unwind (+) /Loss (-)) Loss (bp) Horizon 7/12/2012 8/3/2012 Eurozone contagion Long USU2 156-157 Call spread (digital floor) 3000 contracts $515,625 $195,625 ($320,000) ($250,000) Stop-out 8/9/2012 9/13/2012 Tactical 3m*10y vs 3m*30y bull steepener $117mn: ($50mn) 0 $725,000 $725,000 ($500,000) Unwound 9/7/2012 9/20/2012 Steepener 1y*7y vs 1y*30y bull steepener $63mn: ($20mn) 0 $90,000 $90,000 ($500,000) Unwound 9/20/2012 11/23/2012 Relative Value Short TYZ2 straddles vs 11/23-12 -> 7y +$100mn: (820 ($250,000) ($70,000) $180,000 ($500,000) Expired swaption straddles contracts) 10/23/2012 11/23/2012 Election Long FVZ2 straddles: strike 124 1000 contracts 671,875 781,875 $110,000 ($200,000) Expired 5/24/2012 11/29/2012 Short vol Short 1y*10y straddles vs. long 1y1y payer ($20mn): $200mn ($1,160,000) ($1,015,000) $145,000 Unwound spread 1- 1.25 6/7/2012 11/29/2012 Short vol Short 1y*10y straddles ($50mn) ($3,250,000) ($2,350,000) $900,000 Unwound 11/15/2012 12/11/2012 unwind fvz2 std Short FVZ2 to hedge the long FVZ2 124 1000 124-25 124-25+ $0 Unwound straddles 1/15/2009 12/13/2012 Longer Rates could Rise Buy 5y*10yr Payr Spd (ATM vs 100 bp high), $100mm: - $100mm ($1,924,000) ($1,910,000) $14,000 ($250,000) Unwound added the short CMS cap @ 5% on 7/2/10 ; sell 6w*10y payer @ 2.75% on Nov 10 '11 2/9/2012 12/13/2012 Hike expectations Long 1y*1y - 1y*3y - 1y*5y payer fly ($300mn): $200mn: ($30,000) ($3,000) $27,000 ($500,000) Unwound ($60mn) 5/25/2012 12/13/2012 Short vol Short 2y*10y straddles ($20mn) ($1,830,000) ($1,465,000) $365,000 Unwound 6/21/2012 12/13/2012 Short vol Short 2y*10y straddles ($20mn) ($1,800,000) ($1,445,000) $355,000 ($1,000,000) Unwound 8/16/2012 12/13/2012 Short vol Short 2y*10y straddles ($10mn) ($935,000) ($805,000) $130,000 Unwound 10/23/2012 1/4/2013 Election volatility Long 220mn 2m*5y ATM payer vs. 50mn 220mn:(50mn) ($580,000) ($725,000) ($145,000) ($200,000) Expired 2m*30y ATM payer 1/4/2013 4/5/2013 Tactical 3m*7y vs 3m*30y bear flattener +$75mn:($25mn) ($110,000) $0 $110,000 ($500,000) Expired -491bp: -330bp: $300,000 3/27/2013 4/11/2013 Front end looks cheap Long Jul13s energy hedged $200mn:65 XBM3 ($500,000) Unwound 308.12 284.39 EUR Options 3/9/2012 12/13/2012 Capped steepener 2y5y vs 2y30y bear steepener, short 2y SL 5y- (EUR 90mn): EUR ($1,597,000) ($2,127,000) ($530,000) ($500,000) Stop-out 30y curve cap @ 75bp 20mn: (EUR 400mn) JPY Options 5/30/2012 6/29/2012 Higher rates Sell 1m*20y receiver spread $27mn 41cts 20cts $110,700 ($225,000) Unwound 6/28/2012 7/20/2012 Tactical 1m*10y risk reversal (long payers) $80mn notional 0 cts -38 cts ($304,000) ($160,000) Stop-out 7/6/2012 10/5/2012 Relative Value Long 6mx10y 1x2 payers $80mn notional -26cts 1.8cts $222,000 ($200,000) Unwound 10/25/2012 11/15/2012 Tactical Conditional 10s20s flattener 80mn -8 0 $64,000 ($200,000) Unwound 7/13/2012 12/7/2012 Long vol Long JPY 5y5y OTM recr , delta hedged with $80mn notional 80cts 121cts $328,000 ($160,000) Unwound 10y swap 11/29/2012 1/17/2013 Bear flattener ATM+10bp 6mx5y payer long vs. 6mx10y $100mn -10 -7 $30,000 ($200,000) Unwound payer short 1/10/2013 1/24/2013 Low vol sell 1mx10y receiver (0.8% at 20 sen) $80mn -20 -43 ($24,000) ($200,000) Stop-out 1/10/2013 1/24/2013 Low vol sell 3mx20y receiver (1.65% at 74 sen) $50mn -74 120 ($15,000) ($200,000) Stop-out 11/28/2012 3/28/2013 Volsurface is too flat Long 10yx10y straddle vs. 5yx5y straddle $120mm 88 96 $96,000 ($250,000) Unwound Cross-currency 5/4/2012 6/28/2012 Tactical Long EUR 6m*2y payer vs GBP 6m*2y payer EUR 100mn 0 $135,000 $135,000 ($250,000) Unwound :(GBP82mn) 8/17/2012 10/25/2012 Tactical Receive USD/JPY Xccy basis 20yx10y $30k dv01 53bp 29bp $720,000 ($200,000) Unwound 9/21/2012 11/23/2012 US vs EUR gamma Long RXZ2 std vs. TYZ2 std +500: (860) $300,000 $663,000 $363,000 ($500,000) Expired 10/15/2009 12/13/2012 Cross -currency Short 5x10 US caps @ 8% vs Long 5x10 EUR ($75mm): EUR 50mm ($200,000) $65,000 $265,000 ($500,000) Unwound caps @ 5% 2/3/2012 12/13/2012 Cross -currency Long EUR 3y10y P @ 4% vs USD 3y10y P @ EUR 10mn: (13mn) ($90,000) ($80,000) $10,000 ($500,000) Unwound 4% 11/29/2012 2/28/2013 EUR vs US Long RXH3 std 144.5 vs short TYH3 std 133.5 +400: (740) $100,000 ($720,000) ($820,000) ($1,000,000) Stop-out 12/21/2012 3/7/2012 USD funding supplied via MoF Pay 8y Xccy basis $40k dv01 -66.5 -55 $460,000 ($400,000) Unwound program 2/28/2013 3/7/2013 JPY long end is too rich vs. USD short 10yx10y JPY swap vs. USD (beta = 0.4) usd 40 k dv01 107 97 ($400,000) ($400,000) Stop-out Source: Barclays Research 18 April 2013 67 Barclays | Global Rates Weekly GLOBAL SUPPLY CALENDAR Country Bond Coupon Maturity Size - bn Euro Area 24-Apr-13 Italy BTPei Linker Auction 1.00 24-Apr-13 Germany 30y Bund Auction 2.50% 04-Jul-44 2.00 29-Apr-13 Italy 5yr BTP Auction 3.50% 01-Jun-18 3.00 29-Apr-13 Italy 10yr BTP Auction 4.50% 01-May-23 3.00 29-Apr-13 Belgium 5y BGB Auction 1.00% 04-Mar-18 1.00 29-Apr-13 Belgium 10y BGB Auction 28-Sep-23 1.25 29-Apr-13 Belgium 15y BGB Auction 5.50% 28-Mar-28 1.00 May-13 Belgium New 30y 4.00 02-May-13 France 10y OAT 2.25% 25-Oct-22 4.00 02-May-13 France 15y OAT 2.75% 25-Oct-27 2.00 02-May-13 France 30y OAT 01-May-44 1.50 03-May-13 Spain Bond Announcement 07-May-13 Austria 5y RAGB 15-Apr-18 0.83 07-May-13 Austria 10y RAGB 3.40% 22-Nov-22 0.83 08-May-13 Germany New 5y OBL Auction Apr-18 5.00 08-May-13 Italy Nominal Bond & Size Announcement 09-May-13 Spain 2y SPGB 2.75% 31-Mar-15 1.50 09-May-13 Spain 3y SPGB 31-Mar-16 1.50 09-May-13 Spain 15y SPGB 5.90% 30-Jul-26 1.00 13-May-13 Italy 3yr BTP Auction 01-May-16 3.00 13-May-13 Italy 5yr CCT Auction FRN 15-Jun-17 1.00 13-May-13 Italy 15y BTP 4.75% 01-Sep-28 2.00 14-May-13 Holland 5y DSL Tap (1.5-2.5bn) 15-Jan-18 2.50 14-May-13 Ireland 5y IRISH 5.50% 18-Oct-17 0.25 14-May-13 Ireland 10y IRISH 4.25% 20-Mar-23 0.25 15-May-13 Germany New 2y Schatz Auction Jun-15 5.00 16-May-13 France 5yr BTAN 1.00% 25-May-18 3.50 16-May-13 France 2yr BTAN 0.25% 01-Nov-15 3.00 16-May-13 France 2yr BTAN 1.00% 25-Oct-16 1.50 16-May-13 France OATi/ei Auctions 1.50 17-May-13 Spain Bond Announcement 21-May-13 Belgium Bond Announcement 22-May-13 Germany 10y Bund Auction May-23 5.00 23-May-13 Spain 5y SPGB 4.50% 31-Jan-18 23-May-13 Spain 10y SPGB 5.40% 31-Jan-23 2.00 23-May-13 Spain 20y SPGB 5.75% 30-Jul-32 Japan 23-Apr-13 Japan Liquidity Enhancement Auction 300 25-Apr-13 Japan 2y JGB Auction 2700.00 01-May-13 Japan 10y JGB Auction 2500.00 08-May-13 Japan Liquidity Enhancement Auction 300.00 14-May-13 Japan 30y JGB Auction 700.00 16-May-13 Japan 5y JGB Auction 2600 21-May-13 Japan 40y JGB Auction 400 23-May-13 Japan Liquidity Enhancement Auction 300 UK 24-Apr-13 UK UKT linker 2029 0.125% 22-Mar-29 1.5 08-May-13 UK UKT linker 2044 0.125% 22-Mar-44 1.2 14-May-13 UK UKT 2018 1.250% 22-Jul-18 4.5 16-May-13 UK UKT 2044 3.250% 22-Jan-44 2.5 US 23-Apr-13 US 2y Note Auction 35 24-Apr-13 US 5y Note Auction 35 25-Apr-13 US 7y Note Auction 29 07-May-13 US 3y Note Auction 32 08-May-13 US 10y Note Auction 24 09-May-13 US 30y Bond Auction 16 23-May-13 US 10y TIPs Auction 13 Unconfirmed Barclays Capital Estimate Rich Cheap Source: Barclays Research 18 April 2013 68 Barclays | Global Rates Weekly GLOBAL BOND YIELD FORECASTS US Treasuries US swap spreads Fed funds 3m Libor 2y 5y 10y 30y 10y RY 2y 5y 10y 30y Q2 13 0.00-0.25 0.25 0.20 0.75 1.80 3.00 -0.75 Q2 13 15 15 10 -15 Q3 13 0.00-0.25 0.25 0.20 0.75 1.80 3.00 -0.80 Q3 13 15 15 10 -10 Q4 13 0.00-0.25 0.20 0.25 0.80 2.00 3.15 -0.65 Q4 13 15 15 10 -5 Q1 14 0.00-0.25 0.20 0.30 0.90 2.00 3.15 -0.75 Q1 14 15 15 10 -5 Euro government (Germany) Euro area swap spreads Refi rate 3m 2y 5y 10y 30y 10y RY 2y 5y 10y 30y Q2 13 0.75 0.23 0.10 0.55 1.60 2.35 -0.15 Q2 13 40 40 30 5 Q3 13 0.75 0.25 0.15 0.65 1.75 2.45 -0.10 Q3 13 40 40 30 5 Q4 13 0.75 0.25 0.20 0.70 1.80 2.50 -0.10 Q4 13 40 40 30 5 Q1 14 0.75 0.25 0.30 0.85 1.95 2.60 0 Q1 14 40 40 30 5 UK government UK swap spreads Bank rate 3m 2y 5y 10y 30y 10y RY 2y 5y 10y 30y Q2 13 0.50 0.50 0.30 1.00 2.00 3.15 -1.35 Q2 13 40 15 10 -15 Q3 13 0.50 0.50 0.35 1.10 2.10 3.20 -1.30 Q3 13 40 15 10 -15 Q4 13 0.50 0.50 0.40 1.20 2.20 3.30 -1.25 Q4 13 35 10 5 -10 Q1 14 0.50 0.50 0.45 1.30 2.30 3.35 -1.15 Q4 14 35 10 0 -10 Japan government Japan swap spreads Official rate 3m 2y 5y 10y 30y 10y RY 2y 5y 10y 30y Q2 13 0.10 0. 20 0.09 0.18 0.50 1.25 0.00 Q2 13 15 12 1 -4 Q3 13 0.10 0. 20 0.09 0.18 0.50 1.25 0.00 Q3 13 15 13 2 -10 Q4 13 0.10 0.20 0.09 0.18 0.45 1.15 0.10 Q4 13 15 13 3 -10 Q1 14 0.10 0.20 0.09 0.18 0.45 1.15 0.10 Q1 14 15 13 3 -10 Source: Barclays Research 18 April 2013 69 Barclays | Global Rates Weekly GLOBAL RATES RESEARCH Global Ajay Rajadhyaksha Head of Rates and Securitised Products Strategy +1 212 412 7669 email@example.com US Joseph Abate Piyush Goyal James Ma Chirag Mirani Fixed Income Strategy Fixed Income Strategy Fixed Income Strategy Fixed Income Strategy +1 212 412 6810 +1 212 412 6793 +1 212 412 2563 +1 212 412 6819 firstname.lastname@example.org email@example.com firstname.lastname@example.org email@example.com Amrut Nashikkar Michael Pond Anshul Pradhan Rajiv Setia Fixed Income Strategy Head of Global Inflation-Linked Treasury Strategy Head of US Rates Research +1 212 412 1848 Research +1 212 412 3681 +1 212 412 5507 firstname.lastname@example.org +1 212 412 5051 email@example.com firstname.lastname@example.org email@example.com Vivek Shukla Fixed Income Strategy +1 212 412 2532 firstname.lastname@example.org Europe Laurent Fransolet Cagdas Aksu Fritz Engelhard Jussi Harju Head of European Fixed Income, European Strategy German Head of Strategy European Strategy Commodities and Credit Research +44 (0)20 7773 5788 +49 69-7161 1725 +49 69 7161 1781 +44 (0)20 7773 8385 email@example.com firstname.lastname@example.org email@example.com firstname.lastname@example.org Moyeen Islam Sreekala Kochugovindan Giuseppe Maraffino Amy Mignosi Fixed Income Strategy Asset Allocation Strategy Fixed Income Strategy Barclays Live +44 (0)20 777 34675 +44 (0)20 7773 2234 +44 (0)20 313 49938 + 44 (0)20 3134 9774 email@example.com sreekala.kochugovindan@ firstname.lastname@example.org email@example.com barclays.com Mikael Nilsson Hitendra Rohra Michaela Seimen Henry Skeoch Fixed Income Strategy Fixed Income Strategy SSA & Covered Bond Strategy Inflation-Linked Strategy +44 (0)20 7773 6057 +44 (0)20 7773 4817 +44 (0) 20 3134 0134 +44 (0)20 777 37917 firstname.lastname@example.org email@example.com firstname.lastname@example.org email@example.com Khrishnamoorthy Sooben Marcus Widen Huw Worthington Inflation-Linked Strategy Fixed Income Strategy European Strategy +44 (0)20 7773 7514 +44 (0)20 3134 5632 +44 (0)20 7773 1307 khrishnamoorthy.sooben@ firstname.lastname@example.org email@example.com barclays.com Asia Pacific Igor Arsenin Chotaro Morita Reiko Tokukatsu Head of Fixed Income Strategy Head of Fixed Income Strategy Senior Fixed Income Strategist, Research, Emerging Asia Research, Japan Japan +65 6308 2801 +81 3 4530 1717 +81 3 4530 1532 firstname.lastname@example.org email@example.com firstname.lastname@example.org For any questions about Barclays Live for Rates, please contact: Jason Howell Amy Mignosi +1 212 412 6706 +44 (0)20 3134 9774 email@example.com firstname.lastname@example.org 18 April 2013 70 Analyst Certification We, Laurent Fransolet, Amrut Nashikkar, Rajiv Setia, Hitendra Rohra, Vivek Shukla, Anshul Pradhan, James Ma, Piyush Goyal, Joseph Abate, Cagdas Aksu, Huw Worthington, Giuseppe Maraffino, Moyeen Islam, Mikael Nilsson Rosell, Khrishnamoorthy Sooben, Henry Skeoch, Chotaro Morita, Noriatsu Tanji, Reiko Tokukatsu, CFA and Chirag Mirani, hereby certify (1) that the views expressed in this research report accurately reflect our personal views about any or all of the subject securities or issuers referred to in this research report and (2) no part of our compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this research report. Each research report excerpted herein was certified under Reg AC by the analyst primarily responsible for such report as follows: I hereby certify that: 1) the views expressed in this research report accurately reflect my personal views about any or all of the subject securities referred to in this report and; 2) no part of my compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report. Important Disclosures: Barclays Research is a part of the Corporate and Investment Banking division of Barclays Bank PLC and its affiliates (collectively and each individually, "Barclays"). For current important disclosures regarding companies that are the subject of this research report, please send a written request to: Barclays Research Compliance, 745 Seventh Avenue, 17th Floor, New York, NY 10019 or refer to http://publicresearch.barclays.com or call 212-526-1072. Barclays Capital Inc. and/or one of its affiliates does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that Barclays may have a conflict of interest that could affect the objectivity of this report. Barclays Capital Inc. and/or one of its affiliates regularly trades, generally deals as principal and generally provides liquidity (as market maker or otherwise) in the debt securities that are the subject of this research report (and related derivatives thereof). Barclays trading desks may have either a long and / or short position in such securities and / or derivative instruments, which may pose a conflict with the interests of investing customers. Where permitted and subject to appropriate information barrier restrictions, Barclays fixed income research analyst(s) regularly interact with its trading desk personnel to determine current prices of fixed income securities. Barclays fixed income research analyst(s) receive compensation based on various factors including, but not limited to, the quality of their work, the overall performance of the firm (including the profitability of the investment banking department), the profitability and revenues of the Fixed Income, Currencies and Commodities Division ("FICC") and the outstanding principal amount and trading value of, the profitability of, and the potential interest of the firms investing clients in research with respect to, the asset class covered by the analyst. To the extent that any historical pricing information was obtained from Barclays trading desks, the firm makes no representation that it is accurate or complete. All levels, prices and spreads are historical and do not represent current market levels, prices or spreads, some or all of which may have changed since the publication of this document. The Corporate and Investment Banking division of Barclays produces a variety of research products including, but not limited to, fundamental analysis, equity-linked analysis, quantitative analysis, and trade ideas. Recommendations contained in one type of research product may differ from recommendations contained in other types of research products, whether as a result of differing time horizons, methodologies, or otherwise. In order to access Barclays Statement regarding Research Dissemination Policies and Procedures, please refer to https://live.barcap.com/publiccp/RSR/nyfipubs/disclaimer/disclaimer-research- dissemination.html. Disclaimer: This publication has been prepared by the Corporate and Investment Banking division of Barclays Bank PLC and/or one or more of its affiliates (collectively and each individually, "Barclays"). It has been issued by one or more Barclays legal entities within its Corporate and Investment Banking division as provided below. It is provided to our clients for information purposes only, and Barclays makes no express or implied warranties, and expressly disclaims all warranties of merchantability or fitness for a particular purpose or use with respect to any data included in this publication. Barclays will not treat unauthorized recipients of this report as its clients. Prices shown are indicative and Barclays is not offering to buy or sell or soliciting offers to buy or sell any financial instrument. Without limiting any of the foregoing and to the extent permitted by law, in no event shall Barclays, nor any affiliate, nor any of their respective officers, directors, partners, or employees have any liability for (a) any special, punitive, indirect, or consequential damages; or (b) any lost profits, lost revenue, loss of anticipated savings or loss of opportunity or other financial loss, even if notified of the possibility of such damages, arising from any use of this publication or its contents. Other than disclosures relating to Barclays, the information contained in this publication has been obtained from sources that Barclays Research believes to be reliable, but Barclays does not represent or warrant that it is accurate or complete. Barclays is not responsible for, and makes no warranties whatsoever as to, the content of any third-party web site accessed via a hyperlink in this publication and such information is not incorporated by reference. The views in this publication are those of the author(s) and are subject to change, and Barclays has no obligation to update its opinions or the information in this publication. The analyst recommendations in this publication reflect solely and exclusively those of the author(s), and such opinions were prepared independently of any other interests, including those of Barclays and/or its affiliates. This publication does not constitute personal investment advice or take into account the individual financial circumstances or objectives of the clients who receive it. The securities discussed herein may not be suitable for all investors. Barclays recommends that investors independently evaluate each issuer, security or instrument discussed herein and consult any independent advisors they believe necessary. The value of and income from any investment may fluctuate from day to day as a result of changes in relevant economic markets (including changes in market liquidity). The information herein is not intended to predict actual results, which may differ substantially from those reflected. Past performance is not necessarily indicative of future results. This communication is being made available in the UK and Europe primarily to persons who are investment professionals as that term is defined in Article 19 of the Financial Services and Markets Act 2000 (Financial Promotion Order) 2005. It is directed at, and therefore should only be relied upon by, persons who have professional experience in matters relating to investments. The investments to which it relates are available only to such persons and will be entered into only with such persons. Barclays Bank PLC is authorised and regulated by the Financial Services Authority ("FSA") and a member of the London Stock Exchange. The Corporate and Investment Banking division of Barclays undertakes U.S. securities business in the name of its wholly owned subsidiary Barclays Capital Inc., a FINRA and SIPC member. Barclays Capital Inc., a U.S. registered broker/dealer, is distributing this material in the United States and, in connection therewith accepts responsibility for its contents. Any U.S. person wishing to effect a transaction in any security discussed herein should do so only by contacting a representative of Barclays Capital Inc. in the U.S. at 745 Seventh Avenue, New York, New York 10019. Non-U.S. persons should contact and execute transactions through a Barclays Bank PLC branch or affiliate in their home jurisdiction unless local regulations permit otherwise. Barclays Bank PLC, Paris Branch (registered in France under Paris RCS number 381 066 281) is regulated by the Autorité des marchés financiers and the Autorité de contrôle prudentiel. Registered office 34/36 Avenue de Friedland 75008 Paris. This material is distributed in Canada by Barclays Capital Canada Inc., a registered investment dealer and member of IIROC (www.iiroc.ca). Subject to the conditions of this publication as set out above, Absa Capital, the Investment Banking Division of Absa Bank Limited, an authorised financial services provider (Registration No.: 1986/004794/06. Registered Credit Provider Reg No NCRCP7), is distributing this material in South Africa. Absa Bank Limited is regulated by the South African Reserve Bank. This publication is not, nor is it intended to be, advice as defined and/or contemplated in the (South African) Financial Advisory and Intermediary Services Act, 37 of 2002, or any other financial, investment, trading, tax, legal, accounting, retirement, actuarial or other professional advice or service whatsoever. Any South African person or entity wishing to effect a transaction in any security discussed herein should do so only by contacting a representative of Absa Capital in South Africa, 15 Alice Lane, Sandton, Johannesburg, Gauteng 2196. Absa Capital is an affiliate of Barclays. In Japan, foreign exchange research reports are prepared and distributed by Barclays Bank PLC Tokyo Branch. Other research reports are distributed to institutional investors in Japan by Barclays Securities Japan Limited. Barclays Securities Japan Limited is a joint-stock company incorporated in Japan with registered office of 6-10-1 Roppongi, Minato-ku, Tokyo 106-6131, Japan. It is a subsidiary of Barclays Bank PLC and a registered financial instruments firm regulated by the Financial Services Agency of Japan. Registered Number: Kanto Zaimukyokucho (kinsho) No. 143. Barclays Bank PLC, Hong Kong Branch is distributing this material in Hong Kong as an authorised institution regulated by the Hong Kong Monetary Authority. Registered Office: 41/F, Cheung Kong Center, 2 Queen's Road Central, Hong Kong. This material is issued in Taiwan by Barclays Capital Securities Taiwan Limited. This material on securities not traded in Taiwan is not to be construed as 'recommendation' in Taiwan. Barclays Capital Securities Taiwan Limited does not accept orders from clients to trade in such securities. This material may not be distributed to the public media or used by the public media without prior written consent of Barclays. This material is distributed in South Korea by Barclays Capital Securities Limited, Seoul Branch. All equity research material is distributed in India by Barclays Securities (India) Private Limited (SEBI Registration No: INB/INF 231292732 (NSE), INB/INF 011292738 (BSE), Registered Office: 208 | Ceejay House | Dr. Annie Besant Road | Shivsagar Estate | Worli | Mumbai - 400 018 | India, Phone: + 91 22 67196363). Other research reports are distributed in India by Barclays Bank PLC, India Branch. Barclays Bank PLC Frankfurt Branch distributes this material in Germany under the supervision of Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin). This material is distributed in Brazil by Banco Barclays S.A. This material is distributed in Mexico by Barclays Bank Mexico, S.A. Barclays Bank PLC in the Dubai International Financial Centre (Registered No. 0060) is regulated by the Dubai Financial Services Authority (DFSA). Principal place of business in the Dubai International Financial Centre: The Gate Village, Building 4, Level 4, PO Box 506504, Dubai, United Arab Emirates. Barclays Bank PLC-DIFC Branch, may only undertake the financial services activities that fall within the scope of its existing DFSA licence. Related financial products or services are only available to Professional Clients, as defined by the Dubai Financial Services Authority. Barclays Bank PLC in the UAE is regulated by the Central Bank of the UAE and is licensed to conduct business activities as a branch of a commercial bank incorporated outside the UAE in Dubai (Licence No.: 13/1844/2008, Registered Office: Building No. 6, Burj Dubai Business Hub, Sheikh Zayed Road, Dubai City) and Abu Dhabi (Licence No.: 13/952/2008, Registered Office: Al Jazira Towers, Hamdan Street, PO Box 2734, Abu Dhabi). Barclays Bank PLC in the Qatar Financial Centre (Registered No. 00018) is authorised by the Qatar Financial Centre Regulatory Authority (QFCRA). Barclays Bank PLC-QFC Branch may only undertake the regulated activities that fall within the scope of its existing QFCRA licence. Principal place of business in Qatar: Qatar Financial Centre, Office 1002, 10th Floor, QFC Tower, Diplomatic Area, West Bay, PO Box 15891, Doha, Qatar. Related financial products or services are only available to Business Customers as defined by the Qatar Financial Centre Regulatory Authority. This material is distributed in the UAE (including the Dubai International Financial Centre) and Qatar by Barclays Bank PLC. This material is distributed in Saudi Arabia by Barclays Saudi Arabia ('BSA'). It is not the intention of the publication to be used or deemed as recommendation, option or advice for any action (s) that may take place in future. Barclays Saudi Arabia is a Closed Joint Stock Company, (CMA License No. 09141-37). Registered office Al Faisaliah Tower, Level 18, Riyadh 11311, Kingdom of Saudi Arabia. Authorised and regulated by the Capital Market Authority, Commercial Registration Number: 1010283024. This material is distributed in Russia by OOO Barclays Capital, affiliated company of Barclays Bank PLC, registered and regulated in Russia by the FSFM. Broker License #177-11850-100000; Dealer License #177-11855-010000. Registered address in Russia: 125047 Moscow, 1st Tverskaya-Yamskaya str. 21. This material is distributed in Singapore by the Singapore branch of Barclays Bank PLC, a bank licensed in Singapore by the Monetary Authority of Singapore. For matters in connection with this report, recipients in Singapore may contact the Singapore branch of Barclays Bank PLC, whose registered address is One Raffles Quay Level 28, South Tower, Singapore 048583. Barclays Bank PLC, Australia Branch (ARBN 062 449 585, AFSL 246617) is distributing this material in Australia. It is directed at 'wholesale clients' as defined by Australian Corporations Act 2001. IRS Circular 230 Prepared Materials Disclaimer: Barclays does not provide tax advice and nothing contained herein should be construed to be tax advice. Please be advised that any discussion of U.S. tax matters contained herein (including any attachments) (i) is not intended or written to be used, and cannot be used, by you for the purpose of avoiding U.S. tax-related penalties; and (ii) was written to support the promotion or marketing of the transactions or other matters addressed herein. Accordingly, you should seek advice based on your particular circumstances from an independent tax advisor. © Copyright Barclays Bank PLC (2013). All rights reserved. No part of this publication may be reproduced in any manner without the prior written permission of Barclays. Barclays Bank PLC is registered in England No. 1026167. Registered office 1 Churchill Place, London, E14 5HP. Additional information regarding this publication will be furnished upon request.
Pages to are hidden for
"Barclays Capital - Summertime blues"Please download to view full document