Deutsche Bank

Deutsche Bank Markets Research Global Foreign Exchange FX Spot FX Blueprint Thin end of the wedge Theme #1: Holler for dollars: Buy USD vs. NZD, CHF, SGD, buy EUR/USD risk reversal Theme #2: Play it again, san: Buy USD/JPY Theme #3: Home Counties trump Mounties: Buy GBP/CAD Theme #4: Swiss -Out: Sell CHF/NOK Theme #5: Dingo unchained: Buy AUD/NZD Theme #6: SEK to score with thaw: Sell EUR/SEK, buy EUR/NOK put Theme #7: Trend not bitter end: Follow trend in JPY and CAD Theme #8: Balti not faulty, China still finer: Buy USD/INR put, buy USD/SGD, sell USD/CNH, sell JPY/KRW Theme #9: Pole dances, TRY trips: Sell EURIPLN, sell USD/ILS, sell EUR/HUF, sell EUR/RUB, buy USD/TRY Theme #10: Pesos no prickly pair. Buy MXN vs. USD, CLP, RUB, buy CLP/COP Theme #11: Vol to roll: Buy EUR/USD FVAs, USD/JPY vol swaps and risk reversals, AUD/USD puts r4 9 January 2014 'Research Team London Bilel Hafeez James Malcolm Henrik Gullberg George Sarevelos Siddhanh Kapoor Oliver Harvey Nicholas Weng New York Alan Ruskin Daniel Brehon Dreusio GiacomoIli Guilherme Marone Singapore Sameer Goel Ma'like Sadie!eye Perry Kojodjojo Sydney Adam Boyton Tokyo Teisuke Tanaka Head of FX Strategy Bilel Hafeez Deutsche Bank AG/London DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1. MICA(P) 054/04/2013. EFTA00610276 9 January 2014 FX Blueprint: Thin and of the wedge Overview Sticking to regime change; dollar uptrend 2013 marked a fundamental regime change from the crisis-prone 2008-2012 period. The dollar's correlation to equities flipped, the euro-area avoided a crisis and the Fed announced a rolling back, rather than an expansion, of QE. If there was a locus of crisis it was in emerging markets, which felt the shock of Fed taper. This could hint that the 1990s dynamic of first half dollar weakness and developed market crises and second half dollar strength and emerging market crises could be repeating itself. We therefore remain committed dollar bulls. If last year was all about the US long-end being re-priced on taper, 2014 will mark the re-pricing of the US short-end. December's Fed decision therefore represents only the thin end of the wedge for US interest rate normalization and its effect on markets. This should allow the USD to strengthen against the core European hold-outs to dollar strength, the euro, Swiss franc and pound. The equity flow picture should finally move in favour of the US as slow-moving capital adjusts to the new DM regime. Our favourite expression of dollar strength would be to buy it against the three most over-valued currencies in the world, the New Zealand Dollar, Swiss franc and Singaporean dollar. Yen trend still down While we are looking for a reversal in core European currency trends, on the yen we remain firmly in the bearish camp and look for a trend extension from last year. What adds to our confidence is that major yen turns tend to see the yen move by 43% on average,and we're nowhere near such a move yet. On fundamentals, the BoJ is also conspicuous amongst the major central banks in ramping up QE, the basic balance of payments is heavily negative and foreigners have yet to unwind their safe-haven inflows to Japan that were accumulated in the crisis years. Rest of G10 We underestimated UK growth in 2013, but for 2014 we intend not to miss the changes in the UK economy. The starkest one will likely be the pick-up in inflation, which will only add to expectations of a more hawkish Bank of England. The pipeline for FDI into the UK also looks good. The main weakness for the pound remains the current account deficit, so as a FX trade we like to buy the pound against another current account deficit currency, the Canadian dollar. Helping the bearish CAD case is that expectations of a hawkish Bank of Canada appear overdone given the disconnect between the US and Canadian economy, the likely reversal of the surge of bond inflows seen since 2008 and a turn lower in commodity prices. A neat way of playing the lead-lag between different segments of the market to the normalization in developed markets is to buy the Norwegian krone and sell the Swiss franc. The former saw a large unwind of post-crisis safe-haven inflows last year, while safe- haven flows to Switzerland have yet to be unwound. We should start to see this happen in 2014. Elsewhere in Europe, the Swedish krona should do well as the Swedish economy finally catches up to German and US economic strength. Asia-Pac winners and losers In the Asia-Pacific region, one of the largest cross moves in 2013 was AUD/NZD, but we expect a major reversal this year. Aside from attractive valuations, the rates markets will likely price a more hawkish RBA compared to an already aggressively priced RBNZ. We'd look for the Korean won to outperform the Japanese yen on an improving current account, a pick- up in global growth and a robust domestic financial system. The Singaporean dollar will struggle as valuations are stretched, household debt is elevated and the currency is closely tied to the overall dollar trend. Finally, we'd still buy CNH as the current account, inflation and likely capital inflows are supportive, though we remain wary of the carry unwind dynamics seen in the currency. Fragile EM; strong EM The Indian rupee is the only 'fragile five' currency we like to be long. Current account improvement, portfolio inflows after last year's reduction and beneficial policy action adds up to a bullish case. By contrast, the Turkish lira and South African rand should continue to struggle as their current account dynamics are poor. While both Indonesian rupiah and the Brazilian real also suffer from rickety current accounts and domestic dynamics, better valuations and high carry may prevent excessive weakness. Not all EM is bad. We like the Polish zloty, Israeli shekel and Mexican peso. The first on growth, the second on commodities and third on expected FDI and cyclical pick-up. Last year's Blueprint's Trades Our trades from the last Blueprint were mixed. Our best trade was going long MXN/BRL (+7.1%) while our worst was being short TRYIZAR (-3.6%). Overall, 6 of the themes made money while 4 lost money. Overall, our 10 themes made a 0.47% average retum. Bi!al Hafeez, London Page 2 Deutsche Bank AG/London EFTA00610277 9 January 2014 FX Blueprint: Thin end of the wedge Theme #1: Holler for dollars ▪ We expect the Fed and a recovery in US equity inflows to be the two main pillars of support for the dollar in 2014. Buy USD vs. NZD, CHF and SGD on cycle and extreme valuation. e The Euro-area current account remains supportive for the euro, but tight liquidity is fully priced, the risk of negative rates is material, portfolio inflows are peaking and we are reaching the 20% FX over- valuation bound. Buy 12m EUR/USD 1.4011.31 risk reversals for zero cost. Don't Rely on Fed Dovishness Last year was all about pricing out QE, even though tapering is just beginning. The overwhelming message is that the market front-runs major events, and that the timing of re-pricings is very unpredictable. Just as 2013 was about QE unwinds and higher long-end yields, we think this year will be about the re-pricing of "low for long" and higher short-end yields. First, US short-end expectations are exceptionally benign. The market is not pricing the first rate hike until Q3 2015, just in line with FONT projections, and by which time a simple linear extrapolation of the US unemployment rate takes us well below 6%. Second, the US yield curve is close to all-time steepness extremes. On the one hand, this means that the risks are skewed towards flattening, historically one of the most supportive yield curve environments (chart 1). On the other hand, the forwards are extremely high, suggesting that even if these are realized, the US dollar will climb up the carry ladder and drop-out of the bottom-3 yield ranking by year-end. Foreigners to Come Late to the Party The second building block to our bullish USD view is our positive outlook on growth and by extension US equity inflows. 2013 stood out for large dollar cash accumulation, on the back of UST liquidation and an adjustment of USD hedge ratios (chart 2). The year has also stood out for record outflows from US equities as Americans have invested large amounts offshore and foreigners have refused to engage in the S&P 500 rally. But looking at relative valuations, outflows have overshot what is a relatively benign valuation picture for US stocks. Using the average P/E ratio of Hong Kong and UK stocks as a global proxy, we find that valuations are close to the medium-term average (chart 3). The odds therefore seem skewed towards higher equity inflows into the US, which combined with a flatter US curve should see an improvement in portfolio flows that was lagging this year. 'Curve Flattening Very Bullish USD ChingelnDAYerounicifferent moves in 2yr-10/rUSyleidcufve 20ra 19804orisy 2096 15% 10% 6% 0% -10% 'Growth' •15% -2096 -25% I I Beer Twist Flatten linen Sourer Ocuadre Sank iMf But Flatten 'Etetbtlorr Beer ai TNSt Steepen Steepen Steepen 'UST Selling Last Year Offset By Cash and Hedging 2000 1800 1600 1400 1200 1000 800 600 400 200 C..sourative US short-term ponfoho flows Immlo & ST loans, Do USD) 94 96 98 00 02 04 06 08 10 12 SOU/014 Ditatitho Hunk abentre RAMC* LP, ONIStehat Equity Valuations Not Extreme, Inflows Should Recover —Relative P/E ratio, US vs world•, 12m lead Ohs 1.2 Net equity inflows Ohs, % gross flowsl f 120% 1.t 1.0 0.9 0.8 0.7 0.6 70% 20% -30% -80% 0.5 I 130% 96 97 98 99 00 01 03 04 05 06 07 08 10 11 12 13 14 Swear thvescne Bonk lleaming Mime t0 'we pan *odd by the ME end *op Sang I Deutsche Bank AG/London Page 3 EFTA00610278 9 January 2014 FX Blueprint: Thin end of the wedge Hard to See EUR Story Get Any Better The EUR was the star performer in 2013, despite a strong dollar elsewhere. The Euro-area's large current account surplus has helped, which has pushed the basic balance into positive territory and in the past has been associated with broad-based EUR-appreciation. It is for this reason that we expect the EUR to continue to hold up relatively well against many G10 FX. Looking at drivers more relevant to EUR/USD however, the positive factors that have driven strength versus the dollar have peaked. First, interest rate differentials should turn lower over 2014. Looking across different tenors as well as bond versus implied forward yields, we find that the euro is most sensitive to short-dated forward-implied yields. Last year short-end European yields moved higher not only on the back of ECB LTRO liquidity withdrawal, but as the cross-currency basis also moved back to flat for the first time since 2008 (chart 1). For this year, the risks are skewed the other way. There is less than 200bn EUR of excess liquidity left, EONIA is back to the refi rate and cross-currency basis is flat, so there is no room left for higher short-end European yields. In contrast, the ECB retains a strong easing bias and negative rates or additional liquidity injections remain a strong possibility. On the flow side, the best is behind us as well. Portfolio inflows into the Euro-area have been dominated by equity, but cumulative purchases are now back to trend and on a relative valuation basis Euro-area equities are at a 10-year high. On the outflows side, European offshore investment remains very pro-cyclical, so an improving cycle should lead to a pick-up in Euro- area outflows (chart 2). Add to that the peak in the current account surplus on the back of recovering domestic demand and the risk of additional ECB easing, and we like buying a 1.42p/1.34c EUR/USD risk reversal for zero cost. (a) Other Dollar Crosses to Short Looking outside of EUR/USD, NZD, SGD, and CHF are our top shorts. The Swiss franc is a higher beta version of EUR/USD, with valuations more stretched and greater potential for capital outflows. NZD and SGD are the most over-valued currencies in the world, having lagged all other FX in the USD appreciation that has materialized so far. We therefore like buying USD vs. NZD, CHF and SGD. This basket has a steady —80% correlation with both the narrow and broad USD trade- weighted indices over the last ten years. George Saravelos, London, Bilal Hafeez, London, I EUR/USD Still Tracking Rate Differentials 1.0 EUR/USD Ohs) mplied 2yr E R/USD yields (rhs) 1.38 1.36 1.30 1.32 1.3 1.28 1.26 1.24 1.22 1.2 Jun-12 Sep-12 Dec-12 Mar-13 Jun-13 Sep-13 Dec-13 0.1 Sono Damen, &trot Consenws economy, tr ancectormegotions 0 -0.1 -0.2 -0.3 -0.4 -0.5 IPortfolio Inflows Into Euro-Area Peaked 400 300 200 100 0 -100 -200 300 -400 -500 Euro - World Business Surveys (Ihs) Net Portfolio flow, inveretd outho 08 08 10 11 12 13 &am basset* (WA Scoarteg fa no, 0' -20 -15 -10 -5 0 5 10 15 20 25 30 I NZD, SGD, CHF Most Expensive FX in the World 30 PPP valuation adjusted for productivity I20 -. and terms of trade (BEER) 10 JIllrrr -10 - -20 -30 00 -50 ou_occ<noomm-zleceoavo>-zei.oiln-a.ler4r4ticca ,..aaDD..cE,cr.dpeori Ea-ni optuE-lozo zvmw ev, r2v mo mgvo.mucen vz,cd Somas OwitstAo Byµ libonits7 Fiume fit Page 4 Deutsche Bank AG/London EFTA00610279 9 January 2014 FX Blueprint: Thin end of the wedge Theme #2: Play it again, san We always suspected 2013 would be special, and in this publication a year ago urged investors not to wait for a dip in USD/JPY to buy. We highlighted the ongoing deterioration in Japan's balance of payments and huge pent-up energy in risk-averse short-term capital flows post-GFC. We noted how sentiment was shifting but investors seemed overly cautious, and pointed out that trend turns are typically worth 20.25% on a first-year basis - basically what ensued. It didn't matter that we couldn't anticipate the BoJ's radical departure or mixed Fed signals on QE. The associated swings in risk appetite and positioning created some bumps in the road. But a single-minded focus and conviction ultimately paid off big time. Last year's gain now appears to be this year's pain. The yen posted its largest percentage point loss in 34 years and the Nikkei its biggest rally since 1972. Both go into the New Year on their highs, even as WY JGB yields are within 5bp of their end-2012 level. Few savor the prospect of chasing or fighting such markets. The trend is yet young Still, outsized (10%+) moves in USD/JPY have extended in the following year twice as often as they retraced: 12 times versus 6 since 1971 (top chart).1 We think this is a multi-year-trend, and historically such moved have cumulatively been worth 43% (log terms) with a standard deviation of 20%, typically lasting for just over three years (see chart 2). Mechanically speaking, that would take spot to 116 in December, and well through most conceptions of 'fair value.' One should not shy from forecasting such things, because overshoot is very much the norm for FX (chart 3). Base money differentials matter, will grow As far as the eye can see policy settings also imply it. Deputy Govemor Iwata argued in October that past experience both in the US and Japan show QE works mainly via changes in the money stock. (Hence the importance of the BoJ's commitment to doubling the monetary base; he asserts 'the current level of the monetary base is irrelevant.') Time lags in transmission mean its impact - including on the yen (top chart over page) -- will build even as the initial expectations boost from 'shock and awe' ebbs. Consequently, markets need not fear the absence of further easing by the Bank if, as seems likely, the consumption tax hike in Q2 passed smoothly. Governor Kuroda has also made clear that Japan's massive money expansion will continue until the 2% core CPI target is not only achieved but deemed secure on a sustainable (medium-term) basis. xFor the Nikkei the same stets using a 20% threshold are 5 and 10, i.e., only halt as many large moves continue. 'What happens in year after big" USD/JPY move? 6 Trend reverses s - 4 2 • 0 Trend continues -25%-301-15%-10%-5% 0% 5% 10% 15% 20% 25% 50% 35% Scum.: &vow,. Bent Boor nn Faience UP. • &.c. N/ ass POSt Sotkg ego ILong-term trends in USDJPY Date Rate 13/01/1971 358.44 Log chg Months 14/03/1973 254.45 -34% 26.0 08/12/1975 306.84 19% 32.8 30/10/1978 177.05 -55% 34.7 04/11/1982 277.65 45% 48.1 25/11/1988 121.10 -83% 72.7 17/04/1990 160.20 28% 16.7 19/04/1995 79.75 -70% 60.1 11/08/1998 147.66 62% 39.7 26/11/1999 101.25 -38% 15.5 31/01/2002 135.15 29% 26.2 17/01/2005 101.69 -28% 35.6 22/06/2007 124.14 20% 29.2 31/10/2011 75.35 -50% 52.3 Abs avg 43% 37.7 Stdev 20% 16.7 20/12/2014 116.00 43% 38.6 San Ammo. Gant &bombe. Memo UP IFair value contestable, overshoot is not 160 150 140 130 120 110 100 90 80 70 • 60 • REER (2005*100) — — — Polynomial trend (3) — Linear trend 71 74 77 80 83 86 89 92 95 98 01 04 07 10 13 Scurry Oeinschr Bret 8bombeep 'mane LIP Deutsche Bank AG/London Page 5 EFTA00610280 9 January 2014 FX Blueprint: Thin end of the wedge External accounts highlight major vulnerability Japan's balance of payments still promises a brisk tailwind for yen weakness. The trade account has deteriorated further, against most expectations of stabilization. This is overwhelmingly an oil and gas story, though lackluster external shipments and strengthening domestic demand are supplementary issues. Net FDI outflows a similar-sized drag that is only likely to get bigger even if the current account gradually improves. That reflects a massive cost-of- funding advantage spurring catch-up after Japanese corporates have been risk averse and lagged their competitors in overseas expansion over the last two decades. Overall it leaves Japan's 'narrow basic balance' in a pretty neutral state (middle chart). Cross-border portfolio flows have also had little impact on the yen thus far. Foreign equity inflows seem to have been largely hedged and new bond outflows were limited to banks' offshore treasury activity. Henceforth, real money inflows to Japanese stocks should pick up, but will probably be balanced by growing Japanese outflows into foreign bonds and stocks - the latter encouraged by the new NISA scheme and more aggressive GPIF portfolio adjustment away from JGBS. That would leave the financial account's short-term loans balance as largest swing factor again. It captures carry trades and foreign asset hedging activity and responds to risk aversion and investor's perception of long-term interest rate differentials. It is several orders of magnitude larger than other BoP components and seen in stock terms retains scope for several hundred billion dollars of yen-selling unwinds (lower chart). Abenostics abound Conventional market tops are characterized by hubris in the mainstream which creates unquestioning acceptance of a 'new status quo.' By contrast, most forecasts for Japanese asset prices strike us as intensely conservative as there is tremendous skepticism that the country's long-term prospects have really been changed by the advent of Abenomics. 2 Inevitably there will be bumps along the way, and nobody should expect a free lunch. But until the basic tenets of what remains a uniquely favorable backdrop of fundamentals and potential flows are challenged or overshadowed (geopolitics, anyone?), encourage investors to replay last year's template as this year's basic game plan. We expect FX-equity correlations to remain extraordinarily high and volatility to stay elevated. Our end-2014 and 2015 USD/JPY forecasts of 115 and 120 are reiterated with upside risks. 2 For a convincing exposition of this assertion, see Kuroda's recent speech: https://www.boi.or.iaten/announcements/pressAoen 2013/clata/ko131225 alacif 'Relative base money growth will only point higher 30 20 10 0 -10 -20 -30 -no —ease money Pp (%YoY.1N-LtS) - USCUPY (RHS) 00 01 02 03 04 OS 06 07 08 09 10 11 12 13 14 Santa Oaercaraant abaft." &Awe CP 140 130 120 110 100 90 80 70 60 'Narrow basis balance still under pressure JOY aril, 3mma 3.0 ! 2.0 - 1.0 0.0 -1.0 -2.0 3.0 liTransfe s ncome i' Goods & services II —1 m(6nvina) 04 06 08 10 12 02 Same' Dana* awe Seam:Keg "'nano, CP 'Financial account's 'other investment balance' is key 30 20 10 0 •10 • -20 —Short-term loans Ise foreign asset (WY aril, commutative) hedge demand -USD1PY (RHS, inverted) Carry trade extends. -30 150 96 98 00 02 04 06 08 10 12 14 and unwinds &Knee' Deutsche aset 8boftwg 'nine.> CP 60 70 80 90 100 110 120 190 140 James Malcolm, London, (+44)20754 50884 Taisuke Tanaka, Tokyo Page 6 Deutsche Bank AG/London EFTA00610281 9 January 2014 FX Blueprint: Thin end of the wedge Theme #3: - Home Counties trump Mounties • Inflation risks and potential FDI inflows should keep sterling supported this year, but a widening current account deficit and stronger USD will constrain gains versus the euro and dollar. We are moderately bearish EUR/GBP and GBP/USD, forecasting 80p and 1.54 by end-year respectively. • As an outright trade we prefer buying GBP versus CAD, on which we are more bearish. It remains vulnerable to an equally-large current account deficit, QE-flow unwinds and unwinding internal imbalances. The UK domestic demand cycle has kicked in quicker, stronger and more sustainably than anticipated. We see inflation and FDI inflows as being the major source of upside risk for GBP next year. Inflation, not deflation to be theme in 2014 Last year was all about UK growth expectations. This year, the risk is attention turns to prices. First, even though recent inflation prints have surprised to the downside, pipeline pressure is building. Productivity is not recovering as quickly as the BoE is expecting, leading to a much faster than expected drop in the unemployment rate. In turn, wage inflation could accelerate sharply in 2014 (chart 1). Second, the starting point of inflation is much higher in the UK than the rest of G10. Any turn in the trend will focus minds much more quickly than elsewhere. With the first BoE rate hike still only priced for mid-2015 (a bit earlier than the Fed), there is plenty of potential for near-term yield support as price pressure builds. Outside of monetary policy, FDI is another source of support. Excluding the Verizone-Vodafone deal, UK inbound has been muted by pre-crisis standards. But UK deal-flow is pro-cyclical, tracking the broad trends in equities and global transactions well. Our equity analysts are positive on both this year. Add to that the UK government's increasing dedication to attracting foreign investment - particularly into the publicly-owned banks - (chart 2) and FDI stands out as a potential additional source of support for GBP in 2014. Current Account Deficit Will Constrain Gains vs. EUR On the flipside, the UK recovery is happening for the "wrong" reasons. Domestic demand, not exports are driving the cycle. The current account deficit is deteriorating and stands in contrast to the US and Euro-area. This goes a long way to explain the lag in EUR/GBP versus cyclical indicators (chart 3). This is reflected in our conservative EUR/GBP forecasts: we see a slow grind lower to 0.80 by the end of the year, with GBP/USD revisiting the low 1.50s on the back of a strong USD. IInflation Could Be Here Sooner Than Market Expects 58.0 56.0 54.0 52.0 50.0 48.0 46.0 44.0 42.0 40.0 ,Emplyt PMI, 9m lead —GBP AWE reg pay, priv Jan-01 Mar-03 May-05 Jul-07 Sep-09 Nov-11 San' Saba f Afars Itpreway. 1W I FDI Upside Risk To GBP Next Year so so 40 0Budget ler UK Trees end wwwww. hived IndtritMent. MAO GBP 30 20 10 0 — 2008 2000 2010 1 Souses Asses &int Doemeerd 'Mang LIP 1 2011 2012 2013 I EUR/GBP Holding Up Better on Back of Flow EUR/GBP (6m lag, Ms/ —RICS House Price Balance (rhs, inverted) 0.9 0.8 0.7 0.6 0.5 94 96 98 00 02 04 06 08 10 12 Soutar Disney Gant ow-vow guano, LIP 1 90 130 30 -20 12( Deutsche Bank AG/London Page 7 EFTA00610282 9 January 2014 FX Blueprint: Thin end of the wedge CAD Flow Picture Very Negative We are more bearish CAD. Speculative shorts are building, but on the portfolio flow side, the balance of payments is characterized by a large overhang of fixed income inflow positions as a by-product of Fed QE (chart 4). Canada benefitted from close to 280bn of "excess" inflows over the 2010.2013, which have plugged a sharply wider current account deficit similar in size to the UK. The UK, in contrast, has suffered from a lack of inflows in recent years (chart 4). While FDI and underweight fixed income positions have the potential to "plug" the UK deficit, the risks appear skewed the other way in Canada. On the monetary policy side, Canada doesn't appear well placed to benefit from an improving US cycle either. Rate expectations are already running ahead of the Fed and BoE by a quarter, and given the BoC's renewed dovishness there is limited potential of these happening sooner. Indeed, the correlation between US and Canadian data surprises has dropped off sharply over the last few years, pointing to the divergent trends in local housing markets and domestic leverage. To boot, Canadian terms of trade have clearly peaked (chart 6), with the currency moderately expensive versus non-energy commodity prices and supply glut in North America oil production providing an additional headwind. Big picture, GBP/CAD is a dollar-neutral cross, with correlations with other USD crosses all lying below 50% over 1-3 year time horizons and zero correlation with the broad USD TWI since 1995. The relative performance between UK financials and Canadian stocks does a decent job of explaining big picture turns since the early 1990s. The cross remains more than 20% below its pre-crisis average suggesting plenty of potential for upside as the respective domestic stories play out. George Saravelos, London, Oliver Harvey, London, IGBP/CAD a Beta Neutral Cross 2.8 i —GBP/CAD Ohs) —UK fina cials/Toromo stock index (rhs) 2.6 2.4 2.2 2 1.8 1.6 1.4 Sewn Amuse-ftBIM 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 IUK Inflows Could Improve, Opposite of Canada 1100 Cumulative bond inflows, 1000 local currency, bn 900 800 700 600 500 400 300 200 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 Sown. a* it. r later Rea &Ye.: Nig -750bn (UK) 1+280bn (Canada) change relative to trend 2500 2000 1500 1000 0 !GBP/CAD 'Current Account Neutral' Cross 10 Current account deficits., (my bn 90 91 92 93 95 96 97 98 00 01 02 03 05 06 07 08 10 11 12 Source' °math, eant atiOntteep Aveiro UP IUSD/CAD Following Commodity Prices Lower 95 97 99 01 03 05 07 09 11 13 0.85 0.95 1.05 1.15 1.25 1.35 1.45 1.55 1.65 45( 40( 35( 30( 25( 20( — USD/CAD (Ihs, inverted) — BoC Non-energy Commodity Price Index irhs 15( Soutar Akan» &Mc &bombers Fran. UP Page 8 Deutsche Bank AG/London EFTA00610283 9 January 2014 FX Blueprint: Thin end of the wedge Theme #4: - Swiss out We can think of three reasons to go short CHF/NOK. 1. The relative cycle is supportive of rate differentials. Inflation is at a much higher starting point in Norway than in Switzerland, leaving the Norges Bank much less room to manoeuvre than the SNB in the event of upside surprises to import prices or stronger European growth. Market pricing in Norway has evolved rapidly from three months ago, with the first hike now expected in Q3 2015 rather than Q1 2014, more or less in line with the Norges' own projections. NOK is also much better placed to benefit from a stronger US cycle with one of the strongest correlations to US growth, and CHF one of the weakest. One risk is house prices, which have risen precipitately in Norway over the last five years, a sharp reversal of which could weigh on domestic demand and prompt Norges' dovishness. We think the risks of this are slim, however, (see theme #6), and moderate falls will be welcomed by the central bank as skimming froth from the market. 2. Swiss safe-haven unwind should follow Norway's. Like Switzerland, Norway experienced large-scale safe haven inflows as a consequence of the financial crisis, helping to pause customary current account surplus recycling. In the latter's case, these inflows have largely reversed, to the tune of NOK 190bn on a 2y/2y basis). This has been one of the primary recent drags on the krone, but appears to have largely run its course, in contrast to Switzerland where it has yet to begin. 3. CHF also more vulnerable to domestic outflows. As well as the foreign inflow, Norway and Switzerland both saw significant repatriation of domestic assets from abroad. Again, this flow has long turned in Norway but not yet in Switzerland. One possible catalyst will be stronger price pressures next year on the back of more robust growth. This should erode real returns Swiss domestics have enjoyed on already some of the most expensive assets in the world. Historically, Swiss capital flows have been counter-cyclically related to prices (chart 3). Indeed, CHF performance closely tracked changes in real yields last year. By contrast, while conventional Norwegian outflows have resumed, NOK is more protected by surplus savings being invested by the oil fund. Along with the above, CHF/NOK appears fundamentally misaligned with traditional drivers like oil/gold. Finally, the trade benefits from being USD and EUR/USD neutral (with a 15 year correlation of 16% and -11% respectively). Oliver Harvey, London, I NOK one of highest betas to US cycle, CHF lowest 35% FX correlations to quarterly US GDP Ekon 1990 30% ARom2000 25% 20% 15% la% 4% -10% AUD sec NOK 1420 GaP CAD EUA 01f JPY San O.41tithe Sent Scone., name, LP 'Norway flow reversal nearly done, Swiss yet to begin not 00,110P4 1yµ9. % GDP 5% 9,titualend. not path:. flows 0% 5% -36 .40% -30% J96.99 4000 M.042 OCI-03 May-0500006 Jul-013 Feb.m 500-11 San Dames bee. /900M00/0 rename LP IMore Swiss inflation means CHF weakness 40000 20000 0 -20.000 -40.000 40.000 40.000 -1000r= -120.000 .140.000 J0400 A.900 514102 04103 KUrf05 Occ.c5 Ja.ce Feb.10 Sip 1 1 San 0.34rti Ore Seat abaft.. name, 1.1 - S'et knitted NM Pt.ddio DAWNS —$wesCn yin. inverted ths .1 Ob 0 06 2 Deutsche Bank AG/London Page 9 EFTA00610284 9 January 2014 FX Blueprint: Thin end of the wedge Theme #5: Dingo unchained • 2013 saw the largest fall in AUD/NZD in over 28 years. Behind the fall in the cross was a significant move in interest rate differentials. • With the cross near record lows we think the risk / reward from here favors AUD/NZD upside. As a result we would be long AUD/NZD at current levels. As shown in Figure 1, 2013 marked the 'worst' year for the AUD/NZD cross since at least 1986. Interestingly, in prior years when the cross has fallen over 10% the subsequent year sees a significant bounce back. The 13.5% fall in 1987 to 1.0959 was followed by a bounce of 23.5% in 1988; while the 12.7% fall in 2002 (to 1.0727) saw a 6.7% bounce in the following year. With the cross ending 2013 at 1.0850 after a 14.0% decline, history would suggest some likelihood of a significant move higher in AUD/NZD over 2014. Of course there is much more to FX than history. From a more fundamental perspective we see a number of reasons to expect a move higher in the AUD/NZD cross through the course of 2014. The first is valuation. As Figure 2 shows, the AUD/NZD cross has traditionally found a base around 1.05 (with just below here therefore likely to serve as a good level to set any stop). Our valuation metrics (as published in Exchange Rate Perspectives) also find NZD the most over-valued of the G10 currencies on a PPP and BEER basis. The NZD is around 33% 'expensive' on a PPP basis, versus the AUD which is 26% 'expensive'. On a BEER basis the NZD is 22% 'expensive', versus the AUD which is only 5% above 'fair-value'. As far as 'big picture' drivers of the AUD/NZD cross are concerned, it is hard to go past interest rate differentials as shown in Figure 3. Looking a little more closely at the last few data points in that chart it also appears that AUD/NZD has 'overshot' interest rate differentials to the downside. Indeed, the current interest rate differential would appear to be more consistent with the cross trading around 1.13 versus its current level. We should not, of course, rule out interest rate differentials 'catching up' to the cross. That would require, however, markets to price even more tightening for the RBNZ, and/or easing from the RBA. On the outlook for the kiwi central bank our central view remains that we will see 75bps of policy tightening over the course of 2014. Market pricing is a little more aggressive than that, with a little over 100bps of hikes priced for 2014. On the RBA the market is essentially priced for no change in rates in Australia over the coming year - something consistent Figure 1: Large declines in AUD/NZD usually see bounces AG 200 160 lae bet 00 60 100 -160 -200 Annual Amp In AUD/NZI3 „Iri_ta 'II.- i 1 IS Iwo tOM WA 'Oct 2010 2014 Savertv Da en. amt lkentarg Fano, ty 'Figure 2: The cross is also at an extreme AUD/N213 106 PAA36 MAN MO93 WV UN 01 Mr OS Vs CO UM 13 &wet DivetAOLIS.M. 1160nodat Mann LP IFigure 3: Interest rate differentials are the key driver AUO/N20 ar4 interact rate offactit,ats ~4tcnlm 1e]1 --4in MAW. 41111eGnIar PPM 1.06 54006 50207 54409 54000 Sao 10 Sep 1 1 Sao 12 Sp10 Sep 1 J SWAY Data@ Sont Stowlarg foam CP 2.50 ZO2 1.50 I.00 050 0.00 020 Page 10 Deutsche Bank AG/London EFTA00610285 9 January 2014 FX Blueprint Thin end of the wedge with our own view. That said, the Australian rates market has a tendency to sell off considerably once it becomes clear than an easing cycle in Australia is over. (Figure 4). As to what might spark such a sell-off; an improvement in the labor market stands as one clear possibility. RBA action - and market pricing for the RBA - is often driven by conditions in the labor market. As Figure 5 shows, our tracking of 11 different monthly indicators of labor demand and sentiment suggests a pick-up in the pace of employment growth over coming months. All up, we therefore see the risks being skewed toward a narrowing of the front end interest rate differential between Australian and New Zealand, something that would be supportive of our long AUD/NZD stance. Finally, it would be remiss of us not to consider the outlook for China when discussing any view on AUD, either against the USD or crosses. Our house view remains quite constructive on China, with GDP growth expected by DB to continue its recovery towards 8.6% in 2014. Despite the modest decline in the December PMI reading, our local economists note that the PMI's Q4 average reached 51.3, O.5pts above the Q3 average which suggests that Q4 IP and GDP growth are unlikely to have decelerated from that seen in Q3. Looking forward, stronger external demand should see an acceleration of GDP growth over 2014. If this view on China is correct, then it should be supportive of the AUD given the tendency of the market to see the Aussie as a China proxy (see Figure 6). We are; however, a little cautious about overplaying any impact that stronger growth in China may have on the AUD, given that Australia's key commodity export to China (iron ore) is likely to be moving into an oversupply situation this year. We should note here that while many analysts focus heavily on relative commodity prices when considering the AUD/NZD cross; we are inclined to think that the true importance of commodity prices is captured through the influence on interest rate differentials. In other words, we take the view that the impact of divergent commodity price trends is likely to have already been captured by interest rate differentials. All up, some recovery following the sham fall in 2013, the favorable valuation backdrop, and the risk that the interest rate differential moves in the AUD's favor over coming months has us long AUD/NZD. Adam Boyton, Sydney, Figure 4: Once the RBA is 'done' easing, Australian rates usually (and often incorrectly) price a lot of hikes 112001 1100. 1003 sop St* 20, Jan 116 An 16 Jm01 Janie .1m01 Jor.00 MIPS JMO: Ante -Sill Jan', Mat* pdong anus the RBA cab ire AM.." Loh. one el 00 C. th< Jm to* to I •a Y, t Ww4 Spar' Dena &w* Scoateg Faroe:CP 'Figure 5: The labor market might be the catalyst Momay newspaper end 'cameo:poss.business kinnkis and consumes nnys canted to Australian amslayrnont growth Sown Macho 8sat, Stoomberg flan* f* .414 Ara M4& Oeltelt. W8C-Aff Figure 6: Stronger growth in China could also help AUD 12 —0-Ms nSBC PM13rnih *mg* 8 4 0 -4 2009 2011 2013 2005 2007 SMANV Dana* &rot Scoartarg Fnlnro ffr r 26 16 5 4 -15 -26 Deutsche Bank AG/London Page 11 EFTA00610286 9 January 2014 FX Blueprint: Thin end of the wedge Theme #6: SEK to score with thaw • The krona is not the main culprit of a lukewarm Swedish recovery. Rather the latter reflects a broader trend of DMs losing market share to EMs. ▪ But global recovery still positive for the high beta Swedish economy and SEK. Moreover, global growth is equity supportive which is bullish SEK. Swedish growth numbers have disappointed repeatedly over the past 6-9 months, causing market participants to scale down their expectations for growth and policy normalization. This has in turn resulted in a gradual weakening of the SEK from around 8.40 vs. the EUR back in March/April last year to current levels around 8.90. Disappointing growth numbers are a reflection of subdued industrial activity, which in a small open economy like Sweden has also been weighing on consumer sentiment, job growth and domestic activity. However, blaming the disappointing industrial activity on a strong krona is difficult when the currency is undervalued on all metrics (PPP, BEER, FEER), and the SEK REER not historically strong. Instead, the country breakdown of imports in Germany and Norway, Sweden's key export markets, shows developed markets continuing to lose market share to developing countries, suggesting the latter can now compete in major Swedish export areas like "machinery & transport equipment" where they previously have been small players. This is backed up when looking at hard data, with rising German imports mainly benefitting Central and Eastern Europe. The CE-4 market share of total German imports post-crisis has risen by 50% to around 12% of total. Meanwhile Norwegian imports are showing a similar but broader trend towards developing countries (including Asia) and away from the rest of the Nordics, Eurozone, US and UK. This of course very much goes against the popular view that DM imports might be decoupling from EM exports. Nonetheless, whilst losing market share means an ongoing global recovery will be less of a driver of Swedish GDP growth, it will still be growth supportive, and given the degree of openness in the Swedish economy, more so than in most of the rest of G10. This is SEK positive, partly because it will mean market participants will have to re-assess the policy outlook, but more because it should translate into similar growth in corporate revenues. With Swedish stock valuations only in line with the longer- term average, this should also translate into higher equity prices. Global growth and higher equity prices are typically associated with a stronger SEK, and are factors more important than [an orderly] tumaround in the interest rate cycle in the US. Target a gradual move to 8.55 in EUR/SEK, with a stop @ 9.15. In options a3m EUR/SEK 8.65 put costs an indicative 52bp, and can be largely financed by selling a 3m EUR/SEK call with a strike around the more than 2y highs @ 9.15. Figure 1: Albeit with a lag, the Swedish PMI is likely to follow USD and DEM PMIs higher 70 - 65 - 60 — 55 — 50 — 45 — 40 - 35 - 30 - 2005 2006 2007 2008 2009 '2010 I 2011 '2012 I San Detach. Bart &bombs" Rare LP IFigure 2: Exports of Goods & Services, Percent of GDP oe FUR 50 JK 130 20 10 P-IMCAD 4101 40 GB 0 2011 sows. amanita Bat &owners Ramos LP Figure 3: SEK trade-weighted (TCW) & the US ISM 2000 2002 2009 2M6 2008 20111 2012 2014 — US ISM mfg — Sweden TON Index. Thy (inverted site lower Sex = stronger SEK.I. ns awn. Gewalt S,.,& 8bomOtvg 'awe CP Page 12 Deutsche Bank AG/London EFTA00610287 9 January 2014 FX Blueprint Thin end of the wedge • Norges bank to continue to balance falling house prices vs. sticky CPI. • NOK cheap, but investors wary of jumping back in untess/until Norges Bank and/or crude provide support supportive. In Norway, the mainland economy has been growing around 2.0-2.5% YoY for most of the last couple of years, with the recent PMI and manufacturing output suggestive of continuing improvement going forward. Norges Bank's focus is balancing the declining housing market against sticky inflation at the consumer level. With house prices having slowed to largely flat on the year, representing a 3-3.5% drop from the peak in Q3 last year, and headline CPI back in line with the Bank's 2.5% target (from 3.2% YoY in August), Norges Bank is erring on the dovish side, arguing that the rise in [core] inflation is transitory. Indeed, according to the Bank's projections, core inflation is expected to drift back up over the next few months, reaching a peak just above the inflation target of 2.6% in April/May, before dropping down and remaining just below 2% up until the end of 2015. House prices meanwhile, are seen slowing further, to -2.5% to -3.0% YoY in H1 2014, before returning to positive YoY growth in early 2015. The risk to Norges Banks's finely balanced outlook for inflation and housing is twofold. First is a scenario in which past and current FX weakness feeds through to imported inflation, thus preventing core from moderating in line with Norges forecasts. If the Bank then feels compelled to hike rates at a time when house prices already are declining, that would exacerbate the decline and not be currency supportive. An alternative risk scenario is if the house price falls feeds on themselves. With policy rates already as low as 1.50%, and with core CPI projected at or above target over the next 3-6 months, there would be limited scope for policy to provide a stopgap. While the above scenarios are not our baseline, they will continue to be a key factor in the Norges Bank's decision-making process, with monetary policy likely to be stuck between a fear of adding to the decline in house prices on one hand and sticky inflation on the other. Meanwhile crude is likely to continue to flatline in the relatively tight $90 to $110 range of the past few years. Monetary policy and crude are therefore unlikely to provide much in terms of direction in the NOK. However, given recent depreciation cannot be explained by fundamentals, with the Norwegian unit arguably oversold even when taking into account the market's now very dovish outlook for Norges Bank policy and flat oil prices, we are cautiously constructive. On balance we anticipate very gradual downside in EUR/NOK from current levels. A 3m EUR/NOK put @ 8.25 costs an indicative 76bp. Alternatively, finance it by selling a 3m EUR/NOK 8.65 call. Figure 4: Norway, Real Estate Prices, All Residential Buildings, Total & YoY 20 -015 -I to # 5 0 -5 -10 2001 2003 2005 2007 2009 2011 2013 Swat LIPataltil0 Beak IMF 35000 30000 25000 200000 isocioa t0000 5000 0 20 5 'Figure 5: EUR/NOK and Crude oil lob 9.5 - 9.0 - 8,5 - 2 to - 7.5 - 7.0 - 2009 2010 2011 2012 2013 —010,04 Brent, Close, USD, its —EUR/NOIC actual. Ito &wore DeoYene Bork Eitiomewg Fount* CP 30 40 50n 60c 70 80c 90 7gi 100 1102. 120 130 Figure 6: Core inflation - actual and Norges Bank's projection 3.5 3.0 2.5 2.0 1.5 1.0 0.5 JMS/MSJstSJ IMMS ,IbiSJM 2038 2009 2010 2011 2012 2013 2014 We 2.5% Mete target - CPI-ATE (adjusted for taxes and exch.:Jog energy). YoY NO, GIGS BanKs paecbon &wow dissent &mit elocintlxvg finance to Henrik Gullberg, London, 194/ Deutsche Bank AG/London Page 13 EFTA00610288 9 January 2014 FX Blueprint: Thin end of the wedge Theme #7: Trend no bitter end In 2007 we replicated a RBA study that claimed non- commercial "profit seekers" made money on their IMM positions at the expense of commercial "liquidity seekers". 3 The striking conclusion is that currency speculators generally know the correct direction of G5 currencies and investors can profit from knowing their positions (although these signals are less useful in practice since IMM data is lagged three days). In a sense, money is "left on the table" by commercial foreign exchange users, and to a lesser extent by foreign bond and equity investors, which can be earned by non-commercial actors that provide liquidity to currency markets. These profit seekers collect FX risk premia in the same manner Keynes first identified when describing risk transfer in commodity markets.' Profit Seekers Made Money in Every Year since 2003 By our calculations, non-commercial positions made money in every year since data was first released in 1993, mainly at the expense of commercial users (dealers also made money).' In recent years profits have come from timing big EUR/USD moves (2010-11) and catching last year's USD/JPY rise. These MI numbers more closely resemble pre-crisis profits than the outsize gain in 2008 (reflecting higher FX volatility) and the nearly flat 2009 period (probably due to an unexpected GBP rebound). IS MI Eroded As Speculators Accumulate Positions? Our analysis rests on the crucial assumption that profit seekers accumulate positions over the course of the week at the average price. By contrast, live trading metrics such as the Parker Index of currency manager returns show a loss since 2011 as currency volatility has overwhelmed macro trends (with the exception of USD/JPY in 2013) even as the correlation between weekly IMM In and Parker returns remains positive. It is possible that intra-week volatility causes profit seekers (especially momentum traders) to "buy high and sell low" relative to WVAP and that IMM is eroded when currencies trade in a choppy range. Fortunately for investors FX volatility continues to fall and promising trends (JPY, CAD) have emerged. Daniel Brehon, New Yor; 3 See Kearns and Manners (2004), "The profitability of Speculators in Currency Futures Markets-. Reserve Bank of Australia: and Bilal Hafeez 120071, "Currency Markets: Is Money Leh On the Table?" 4 Keynes. M. 11930). ilise on Money". London: Macmillan, 5 We calculate weekly by determining the notional value of positions on Tuesday and assuming longs and shorts are accumulated (or squared) at the average price over the course of the week. Figure 1: Absolute = ($mil) of Profit-Seekers and Liquidity-Seekers Using G5 IMM Data (1993-2013) 18 - E 12" 4,, o.8 - O 0 4 -12 - USD mm (1993-2013. except euro 1999-2013) oNon-Commercial /Profit-seekers aCommercial/Liquidity-seekers Cs. 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 Soutar Daunts &M. lboomberg Feigned LIP Figure 2: Speculators had the correct EUR positions in 2 01 0-1 1 and a large JPY short in 2013 12 8 0 USD mm (1993-2013. except euro 1999-2013) •Non-commercial /Profit-seekers • Commercial / Liquidity-seekers AUD GBP CAD JPY CHF Source DArtne.he BanA. Stomotvg Foram UP EUR Figure 3: The Parker Index of currency manager returns has lagged IMM in recent years despite continued positive correlation between them Average IMM Cum P&L (5100m. 810 — Parker Index Uan-2003 = 0.1hs) — IMM P&L v. Parker (12m corr, rhs) 45 40 35 25 20 IS 10 5 -5 -' :•••••'*".. "-F 03 04 05 06 07 08 Sarre' Oeutsche Bre* 8bomberg Amino> ELP 09 10 11 12 13 100% - 90% 80% 70% 60% 50% 40% 30% - 20% 10% 0% Page 14 Deutsche Bank AG/London EFTA00610289 9 January 2014 FX Blueprint: Thin end of the wedge Theme #8: Balti not faulty, China yet finer 2013 saw Asian FX fracture into a North and South complex. Relative sensitivity to the yen, the EM debt bubble, and the developed world equity/growth cycle all served to discriminate the two intra-regional groups at different points during the year. Ultimately, the North emerged fairly unscathed, while the South cheapened significantly. Indeed, much of South Asia FX still faces a host of challenges from political instability (THB), offshore debt holding overhang and weak commodity prices (MYR, IDR). However we refrain here from adding to shorts against most of this group, given a mix of large valuation adjustments, expensive carry, and/or improving policy responses that is shifting the risk-reward. We concentrate instead on currencies like INR where fundamentals have improved sufficiently to go long, and on SGD, where persistent overvaluation and exposure to a broadening in USD strength out to low-yielders favors a short bias. In North Asia, we stay short USD/CNH, but are mindful of the large build up in positioning. We favor a short JPY/KRW position to express our positive view on Korean fundamentals. The rupee has had a makeover After spending three years as one of the world's worst performing currencies, the rup