Asset Allocation

Highlights Dear Client, This issue of BCA's Commodity & Energy Strategy features our 2017 Outlook for Bulks and Base Metals. The evolution of China's economy will, as always, be critical to these markets, given that country's outsized role in iron ore, steel and base metals. We are broadly neutral the complex, and, with the exception of the nickel market, see supply and demand relatively balanced. That said, the potential for price spikes - e.g., copper, where spare capacity is shrinking - and for monetary and fiscal policy errors to spill into these markets keeps downside price risk elevated. Next week, we will publish our 2017 Outlook for Energy Markets, with special attention to the oil market. As expected, OPEC and Russia agreed to cut production. As we went to press, WTI and Brent crude oil prices were up ~ 8.5% on the news. We will take profits today on our Long February 2017 Brent $50/bbl Calls vs. Short February 2017 $55/bbl Calls, which was up 73.6% basis Wednesday's close when we went to press. We remain long August 2017 WTI vs. Short November 2017 WTI futures in anticipation of a backwardated forward curve in 2017H2; as of Wednesday's close, this position returned 76.39% since November 3, when we recommended the exposure. Our 2017 Precious Metals and Agricultural outlooks will be published in the following weeks. We will finish with an outlook for commodities as an asset class in 2017 at year-end. We trust you will find these reports informative and useful for your investing and year-ahead planning. Kindest regards, Robert P. Ryan, Senior Vice President The monetary and fiscal stimulus that massively boosted China's housing market this year will wind down, bringing an end to the run-up in iron ore, steel and base metals prices. While we expect "reflationary" policies to continue going into the Communist Party Congress next fall, when new leadership roles will be announced, we do not expect anything along the lines of the surge in policy stimulus seen earlier this year: Unwinding and controlling property-market excesses and high debt levels will limit policymakers' desire to turbo-charge the housing market again, limiting the boost such policies provide. We are downgrading our tactically bullish view on iron ore to neutral. Our out-of-consensus bullish call was proven correct with a 43% rally in iron ore prices within the past eight weeks.1 Strategically, we retain a bearish bias, as rising iron ore supply may overwhelm the market again in 2017H2. We remain tactically neutral and strategically bearish steel. Low steel inventories and production disruptions caused by China's recently launched environmental inspection program likely will continue to support steel prices in the near term. However, persistently high steel output and falling demand from the Chinese property sector should eventually knock down prices in 2017H2. We remain neutral copper going into 2017, expecting Chinese reflationary stimulus to continue along with a concerted effort to slow the housing boom in that country. This will still support real demand for copper, but will reduce demand from new construction. Manufacturing will play a larger role on the demand side next year, while a stronger USD could limit price appreciation. We still believe nickel will outperform zinc over a one-year time horizon. We are bullish nickel prices, both tactically and strategically, as we expect a supply deficit to widen on rising stainless steel demand and falling nickel ore supply in 2017. For zinc, we remain tactically neutral and strategically bearish. We expect zinc supply to rise considerably in response to current high prices. For the global aluminum market, we remain tactically bullish and strategically neutral. Supply shortages will likely persist ex-China over the next three to six months. We have three investment strategies, including long iron ore/short steel futures, long nickel/short zinc futures, and buying aluminum on weaknesses. Feature Iron Ore & Steel: Limited Upside In 2017 A Quick Recap Back in early October, we wrote an in-depth report on global iron ore and steel markets in which we made an out-of-consensus tactically bullish call on iron ore, expecting the price to reach the April high of $68.70/MT in 2016Q4. Our prediction was realized, with iron ore prices surging 43% to a two-year high of $79.81/MT on November 11 (Chart 1, panel 1). Although the steel market has been much stronger than the assessment driving our tactically neutral stance indicated earlier in the quarter, our call that iron ore would outperform steel in the near term was correct: Steel prices rose 21% during the same period of time - only half of the iron ore price rally (Chart 1, panel 1). Over the past two months, the rally occurred in both futures and spot markets, and in the markets globally (Chart 1, panels 2 and 3). Chart 1Iron Ore: Downgrade To Tactically Neutral Chart 2Steel: Remain Tactically Neutral The 2017 Outlook First, we downgrade our tactically bullish view on iron ore to neutral, as China likely will import less iron ore in 2017Q1 (Chart 2, panel 1). China has imposed stricter environmental regulations on its domestic metals industry since 2014 to control pollution. The government currently is sending environmental inspection teams to major steel-producing provinces to check how well the steel producers are complying with state environment rules. Many steel-producing factories were closed this year, due to environmental violations. This will constrain growth in Chinese steel output in the near term (Chart 2, panel 2). Between 2011 - 15, the state-owned Xinhua news agency states Chinese steel capacity has been reduced by 90 million MT; authorities want to cut as much as 150 million MT by 2020, including 45 million MT this year.2 Chinese steel production generally falls in January and February as workers are celebrating the Chinese Spring Festival - the most important festival for the Chinese. Iron ore inventories at major Chinese ports are still high (Chart 2, panel 3). Given iron ore prices have already rallied more than 100% since last December and steel demand outlook remains uncertain next year, most steel producers likely will choose to push off purchases into 2017Q2 or later. While China may slow its iron ore purchases next year, global iron ore supply is set to increase in 2017 as many projects will come on stream. The world's biggest iron ore project, Vale's S11D, which has a capacity of 90 million metric tons (mmt) per year, is expected to ship its first ore in January 2017. Moreover, with iron ore prices above $70/MT, global top iron ore companies with low production costs can be expected to sell as much as they can to maximize their profit, given their all-in production costs for high-quality iron ore (62% Fe) typically are between $30 and $35/MT.3 That said, we are not bearish on iron ore prices in the near term. We prefer to be neutral. Iron ore prices will have pullbacks, but the downside may be also limited in 2017H1. Chinese domestic iron ore production is still in a deep contraction (Chart 2, panel 4). Plus, most steel producing companies prefer high-quality ore from overseas over the domestic low-quality ore. In addition, almost all steel companies in China are profitable at present, which means Chinese steel production will rise after the Spring Festival holidays. All of these factors will support iron ore prices. Chart 3Iron Ore & Steel: Strategically Bearish Second, we retain our tactically neutral view on steel. Chinese steel demand was lifted by China's expansionary monetary and fiscal policies this year - which we have dubbed China's "reflationary" policy - which included reductions in its central bank's policy rate and reserve requirement ratio, and implementation of additional infrastructure projects (Chart 3). This was the driving force for the sharp steel price rally this year. The big question is how sustainable Chinese steel demand growth will be? This will be highly dependent on the Chinese government's decisions and actions. More than a third of steel demand is accounted for by the property market, of which some 70% is residential property.4 Mortgages accounted for approximately 71% of all new loans in August of this year, down from 90% in July, according to Reuters.5 This loan growth powered the iron ore and steel markets this past 12 - 18 months and China's credit-to-GDP ratio to extremely high levels. The OECD recently observed, "The high pace of debt accumulation was sustained despite weaker domestic demand growth. This raises concerns about the underlying quality of new credit, disorderly corporate defaults and the possible extent to which it has been used to support financial asset prices. Residential property prices in some of the largest cities have risen by over 30% year-on-year, although price growth in smaller cities has been much more modest. The price gains have been partly driven by loose monetary policy and ample credit availability as well as reduced land supply."6 Based on our calculations, Chinese steel demand started showing positive yoy growth in July and, so far, had posted four consecutive months of positive yoy growth from July to October. In September and October, the growth was accelerated to 8.3% and 6.6%, respectively, a clear improvement from the 0.8% yoy growth registered in July. The growth may last another three to six months but could peak sooner, if there are no new stimulus plans announced by the government. In addition to the housing sector, China's auto industry also saw significant demand growth. As China cut the sale taxes on small passenger vehicles from 10% to 5% this year, Chinese car sales jumped 13.6% yoy for the first 10 months of 2016, a significant improvement from a 5.7% yoy contraction in the same period of last year. If the government lets the tax cut expire at year-end, Chinese auto production may decline in 2017, which will weaken Chinese steel demand. In the meantime, Chinese steel producers will keep boosting production next year, which likely will limit the upside for steel prices. That said, current steel inventories in China are still low. According to the China Iron and Steel Association (CISA), steel inventories at large and medium steel enterprises fell 9% from mid-September to late October. This probably will limit the downside for steel prices. Third, we retain a strategic bearish view on both iron ore and steel. If there is no additional reflationary stimulus deployed in 2017, we expect Chinese steel demand to weaken. In the meantime, Chinese steel producers will keep boosting their production. Let these two factors run nine to 12 months, and we believe they will be sufficient to knock down both steel and iron ore prices. Our research last year concluded the Chinese property sector is structurally down-trending.7 Given that the property market is the biggest end user of steel in China, accounting for about 35% of total steel demand, we are strategically bearish on steel and iron ore prices. How To Make Money In The Iron Ore & Steel Market? Chart 4Take Profit On Long ##br##Iron Ore/ShortSteel Rebar Trade We went long May/17 iron ore futures in Dalian Futures Exchange in China and short May/17 steel rebar futures in Shanghai Futures Exchange on October 6 (Chart 4). Both contracts are denominated in RMB. The relative trade gives us a return of 18.1% in two months. We are taking profits with this publication, but we may re-initiate this pair trade on pullbacks. Risks If China deploys additional fiscal and monetary stimulus next year, similar in scope to this year's stimulus, we will re-evaluate our view accordingly. If global iron ore production is less than the market expects we could see further rallies in iron ore prices. Should this occur, we will re-examine our market call, as well. Copper: Market Is Balanced; Little Flex On Supply Side The reflationary stimulus that powered China's property markets - and drove demand for iron ore and steel higher - also propelled copper prices to dizzying heights in 2016H2. We do not expect this juggernaut to continue, and instead expect copper to trade sideways next year as global supply and demand stay relatively balanced (Chart 5). China accounts for roughly half of global refined copper demand (Chart 6). Manufacturing activity has the greatest impact on prices: A 1% increase in China's PMI translates to a 1.8% increase in LME copper prices (Chart 7). Chart 5Copper Market Is In Balance Chart 6World Copper Markets Are Balanced Chart 7China Demand Will Remain Key For Copper China's property market accounts for about a third of global copper demand in used in construction, according to the CME Group, which trades copper on its COMEX exchange. A 1% increase floor-space started in China leads to a 0.3% increase in LME copper prices (Chart 8). The surge in demand from the housing market lifted China's copper demand over the past 12 - 18 months, as credit creation in the form of home-mortgage loans expanded at a rapid clip (Chart 9). We expect the Chinese government to continue to try to rein in a booming property market, which has seen mortgage-loan growth of 90% p.a. recently. If the government is successful, this will limit price gains for copper next year. If not, the bubble will continue to expand in large tier-1 and -2 cities in China, making the copper rally's fundamental support tenous to say the least. Chart 8China PMIs and USD TWI Drive LME Prices Chart 9Mortgage Growth Likely Slows in 2017 This drives our expectation that the real economic activity in China - chiefly manufacturing - will be the dominant fundamental on the demand side for copper next year. On the supply side, we expect 2.65% yoy growth in refined copper production, just slightly above the International Copper Study Group's 2% estimate. Company and press reports cite a reduced mine capacity additions, lower ore content in mined output, and labor unrest as reasons supply side growth is slowing. Our balances reflect a convergence of supply and demand for next year, and also highlight the reduced flexibility in the system to respond to unplanned outages. For this reason, the global copper market could be prone to upside price risk in the event of a major unplanned production outage. Watch Out For USD Strength Copper, like all of the base metals, is sensitive to the path taken by the USD. We continue to expect the Fed to lift rates next month and a couple of times next year. This most likely will lift the USD 10% or so over the next 12 months. This would be bearish for base metals, particularly copper, since 92% of global demand for the red metal occurs outside the U.S. Our modeling indicates a 1% increase in the broad USD trade-weighted index leads to a 3.5% decrease in LME copper prices. A stronger USD will raise the local-currency cost of commodities ex-U.S. EM demand would suffer, which would slow the principal source of growth for base metals. Metals producers' ex-U.S. with little or no exposure to USD debt-service obligations would see local-currency operating costs fall. At the margin, this will lead to increased supply. These effects would combine to push commodity prices lower, producing a deflationary blowback to the U.S. Nickel & Zinc: Going Different Ways In 2017? Zinc has outperformed nickel significantly for the past six years. This year alone, zinc prices have shot up over 90% since January, almost doubling the 50% rally in nickel prices for the same period of time (Chart 10, panel 1). The nickel/zinc price ratio has declined to its lowest level since 1998 (Chart 10, panel 2). Will nickel continue underperforming zinc into 2017? Or will the trend reverse next year? We believe the latter has a higher probability. Tactically, we are bullish nickel and neutral zinc. Strategically, we are bullish nickel and bearish zinc.8 Zinc's bull story has been well-known for the past several years, and nickel's oversupplied bear story also has been commented on in the news. However, both markets' fundamentals are changing. Based on World Bureau of Metal Statistics (WBMS) data, for the first nine months of this year, the supply deficit in the global nickel market was at its highest level since 1996. Meanwhile, the global zinc market was already in balance (Chart 10, panels 3 and 4). Chart 10Nickel Likely To Outperform Zinc In 2017 Chart 11Nickel Has More Positive Fundamentals Than Zinc Both nickel and zinc markets are experiencing ore shortages (Chart 11, panels 1 and 2). For the nickel market, the ore shortage was mainly due to the Indonesian ore export ban, and Philippines' suspension of nickel miners for violating that country's environmental laws. For the zinc market, the ore shortage arose because of several big mines' depletion, years of underinvestment, and mine suspensions due to low prices late last year. The nickel ore shortage will become acute as the Indonesian ban remains in place and the Philippines' government becomes stricter on domestic mining operations. However, for zinc, most of the output loss occurred last year, and actually may be restored to the market in the near future. Zinc prices reached $2,811/MT last year as the market was adjusting to lost supply - the highest level since March 2008. In terms of demand, nickel exhibits much stronger demand growth versus zinc (Chart 11, panels 3 and 4). In addition, China's auto sales tax-cut policy will expire at year-end, which may cause Chinese auto production to fall in 2017. This will affect zinc much more than nickel, as less galvanized steel will be needed next year if Chinese car production falls. Investment Strategies We sold Dec/17 zinc at $2,400/MT on November 3, and the trade was stopped out at $2,500/MT with a 4% loss (Chart 12, panel 1). Zinc prices jumped 11.5% in four trading days in late November, which we believe was mainly driven by speculative buying. Nonetheless, in the near term, global zinc supply is still on the tight side, and zinc inventories are low (Chart 12, panel 2). Zinc prices could rally more in the near term. We were looking to go Long Dec/17 LME nickel vs. Short Dec/17 LME zinc if the ratio drops to 4.3 since mid-November (Chart 13, panel 1). We also suggested that if the order gets filled, put a stop-loss for the ratio at 4.15. Chart 12Zinc: Stay Tactically Neutral Chart 13Risks To Long Nickel/Short Zinc On November 25, the order was filled at the closing price ratio of 4.17. But unfortunately the ratio declined to 4.08 on the next trading day (November 28), based on the closing price ratio, which triggered our predefined stop-loss level with a 2.2% loss. The ratio was trading at 4.17 again as of November 29. As the market is so volatile, we recommend initiating this relative trade if it drops below 4.05 to compensate the risk. If the order gets filled, we suggest putting a 5% stop-loss level for the relative trade. After all, nickel prices could still have pullbacks, as global nickel inventories still are elevated (Chart 13, panel 2). Risks Our strategically bearish view on zinc will be wrong if global zinc ore supply does not increase as much as we expect, or global zinc demand still has robust growth in 2017. Our strategically bullish view on nickel will be wrong if Indonesian refined nickel output increases quickly, resulting in a smaller supply deficit than the market expects. However, due to power shortages, poor infrastructure and funding problems, development on many of the smelters and stainless steel plants once envisioned for the nickel market have been delayed. We believe these problems will continue to be headwinds for Indonesian nickel output growth, and will continue to restrict supply growth going forward. Aluminum: Cautiously Bullish In 2017 Chart 14Aluminum: Remain Tactically Bullish ##br## And Strategically Neutral Sharp supply cuts combined with tight inventories have pushed aluminum prices higher this year. Prices in China have rallied more than 50% so far this year, which was more than double the 20% rise in the global aluminum market (Chart 14, panel 1). This probably indicates a tighter Chinese domestic market than the global (ex-China) market. Looking forward, we remain tactically bullish on LME aluminum prices and neutral on SHFE aluminum prices.9 The supply shortage will likely persist ex-China over next three to six months. Global aluminum production has declined faster than demand so far this year. Based on the WBMS data, global aluminum output was still in a deep contraction in September (Chart 14, panel 2). Even though China's operating capacity has been rising every month so far this year, Chinese total aluminum output for the first 10 months was still 1.1% less than the same period last year. In addition, considering the possible output loss due to the Spring Festival in late January, we believe it will take another three to six months for China to meet its own domestic demand and inventory restocking. Extremely tight domestic inventories should limit the downside of SHFE aluminum prices (Chart 14, panel 3) as the market adjusts on the supply side. We think there is more upside for LME aluminum prices, as the supply shortage will likely persist ex-China over next three to six months. Currently, Chinese aluminum prices are about 18% higher than the LME prices (both are in USD terms), which will likely limit the supply coming from China's exports to the rest of world. Strategically, we are neutral LME aluminum prices and bearish on SHFE aluminum prices. Currently, about 85% of the China's aluminum operating capacity is making money. With new low-cost capacity and more idled capacity coming back on line, profitable Chinese smelters will continue boosting their aluminum production to maximize profits. This, over a longer term like nine months to one year, should eventually spill over to the global market. Investment strategy Chart 15Still Look To Buy Aluminum We recommended buying the Mar/17 LME aluminum contract (Chart 15) if it falls to $1,640/MT (current: $1,721/MT). We expect the contract price to rise to $1,900/MT over the next three to five months. If our order is filled, we suggest a 5% stop-loss. Risks Prices at both the SHFE and LME may come under intense pressure if aluminum producers in China increases their output quickly, even at a small loss, in order to create jobs and revenue for local governments. If global aluminum demand falters in 2017 while supply is rising, we will revisit our strategically neutral view on LME aluminum prices. Ellen JingYuan He, Editor/Strategist ellenj@bcaresearch.com Robert P. Ryan, Senior Vice President rryan@bcaresearch.com 1 Please see Commodity & Energy Strategy Special Report for iron ore and steel "Global Iron Ore And Steel Markets: Is The Rally Over?," dated October 6, 2016, available at ces.bcaresearch.com. In this report, we are using Metal Bulletin iron ore price delivered to Qingdao port in China as our iron ore reference price. 2 Please see "N. China city cuts 32 mln tonnes of steel capacity" published October 30, 2016, by Xinhua's online service, xinhuanet.com. 3 Please see "CHART: The breakeven iron ore prices for major miners in 2016," published June 7, 2016, by Business Insider Australia. 4 Please see "China Resources Quarterly, Southern spring ~ Northern autumn 2016," published by the Australian Department of Industry, Innovation and Science and Westpac, particularly this discussion on p. 4, "The real estate sector." 5 Please see "China August new loans well above expectations on mortgage boom," published by Reuters September 14, 2016. 6 Please see the OECD Economic Outlook, Volume 2016 Issue 2, Chapter 1, entitled "General Assessment of the Macroeconomic Situation," p. 44, under the sub-head "Rapid debt accumulation risks instability in EMEs." The IMF also expressed concern over rising debt levels supporting the real-estate boom in China, particularly in the larger cities, noting, "Credit and financial sector leverage continue to rise faster than GDP, and state-owned enterprises in sectors with excess capacity and real estate continue to absorb a major share of credit flow. The deviation of credit growth from its long-term trend, the so-called credit overhang--a key cross-country indicator of potential crisis--is estimated somewhere in the range of 22-27 percent of GDP..., which is very high by international comparison." Please see the IMF's Global Financial Stability Report for October 2016, "Fostering Stability in a Low-Growth, Low-Rate Era," p. 35, under the sub-heading "China: Growing Credit and Complexities." 7 Please see Commodity & Energy Strategy Special Report "Chinese Property Market: A Structural Downtrend Just Started," dated June 4, 2015 and "China Property Market Q&As," dated July 2, 2015, available at ces.bcaresearch.com 8 Please see Commodity & Energy Strategy Weekly Report "Oil Production Cut, Trump Election Will Stoke Inflation Expectations," dated November 17, 2016 and "The Lithium Battery Supply Chain: Efficient Exposure To Electric-Vehicle Market," dated October 27, 2016, available at ces.bcaresearch.com 9 Please see Commodity & Energy Strategy Weekly Report "Market Saturation Likely In Asia, If KSA - Russia Fail To Curb Oil Production," dated November 10, 2016, available at ces.bcaresearch.com Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Closed Trades

Highlights An Italian referendum 'no' is not really revolting. Some people are voting no for no change to the current constitution's vital checks and balances. Lean against any knee-jerk widening of the Italian sovereign yield spread versus France that followed a no vote. Lean against any knee-jerk rally in Italian banks that followed a yes vote. A 50bps spike in the JPM Global Government Bond Yield in just 3 months is normally a bad omen for risk-asset performance. Retain a cautious stance to risk-assets on a 3-month horizon. Feature After shock victories for Brexit and Donald Trump at the polls, a 'no' vote in Italy's December 4 referendum on constitutional reform would be the next worrying sign of a growing grassroots revolt against the establishment. Or would it? An Italian 'No' Is Not Really Revolting The votes for Brexit or Donald Trump were clearly votes for change. At first glance, an Italian no would also look like a revolt, with the potential to trigger political uncertainty and instability in the euro area's third largest economy. Chart of the WeekItalian Banks Are Tracking Japanese 'Zombie' Banks The truth is more nuanced. Clearly, some Italians are voting no to reject Prime Minister Renzi. But others - including former Prime Minister Mario Monti - are voting no for no change. These voters want to leave in place the current constitution's vital checks and balances. If Italians vote yes to constitutional reform, the upper house of parliament - the Senate - would be relegated to an advisory chamber. Meanwhile, an already approved new electoral law for the lower house of parliament - the Chamber of Deputies - hands an automatic 55 percent majority of seats to the largest party. Some people fear that this combination would amount to excessive executive power. So they are voting no to mitigate the danger. Granted, a no vote might also force Renzi to resign, but this would not necessarily trigger new elections. President Sergio Mattarella would likely explore options for a new government - perhaps a technocratic government - which the parties in the current governing coalition have a strong incentive to support until the next elections are due in 2018. Even if there were early elections, it is improbable that they would result in a government led by the populist 5 Star Movement. If 5 Star was the largest party, it would hold a 55 percent majority of seats in the lower house, but only 30 percent in the upper house, in proportion to its popular vote share (Chart I-2). Therefore, it could not form a government. Under the current constitution, the government needs the support of both houses. The irony is that a yes vote - by giving the executive excessive powers - would make it more likely for a populist party like 5 Star to form a government in 2018 or beyond. Still, even this might prove a tall order. Italy's constitutional court is reviewing the electoral law change that gives 55 percent of lower house seats to the largest party. The court will likely demand more proportionality, making it hard for any one party to win an outright majority. This means more coalition governments, which 5 Star rejects. Hence, an Italian no will not be the equivalent of the Brexit vote or U.S. election of Donald Trump. Fears that it will unleash a dangerous phase of populism and political instability in Italy are overblown. Yet in the last three months, the Italian sovereign yield spread has widened sharply versus France (Chart I-3). Note also that the 65-day fractal dimension of the Italy versus France sovereign bond performance is close to its technical limit, indicating excessive pessimistic groupthink. Chart I-2The 5 Star Movement Could Not Form A ##br##Government Under The Current Constitution Chart I-3Italy's Political Risk Premium Has ##br##Increased, But Is It Justified? If December 4 brings a no vote in the Italian referendum combined with the election of a far-right President of Austria - whose role is largely ceremonial - the knee-jerk market response might still be fright. In which case, a further widening in the Italy/France yield spread would be a tactical entry opportunity, given that political risk is overstated. Fixing Italian Banks Needs A 'Deep-V' Or A 'Long-L' The real question in Italy is not about an imminent populist backlash. The real question is what does the cure for Italy's banking malaise look like? The answer is either a 'deep-V', meaning a banking crisis forces a quick workout; or a 'long-L', meaning no banking crisis but a very long struggle back to normal health. As an investor, neither seems particularly appealing. Italy's banking malaise has built up stealthily, generating frequent financial tremors but without an outright crisis. In contrast, the housing-related credit booms in Ireland, Spain, the U.K. and the U.S. did eventually cause housing busts and full-blown financial crises - requiring urgent government-led and central bank-led bailouts. Today, Italian banks' non-performing loans (NPLs) account for 18% of gross lending, and NPLs net of provisions equal 85% of equity capital. A few years ago, Irish banks looked even worse. Irish NPLs peaked at 25% of gross lending in 2013 and net NPLs peaked at 100% of equity capital. Following government bailouts Irish banks have recovered well (Chart I-4 and Chart I-5). Likewise, the Spanish government created a 'bad bank' in 2012 to offload bank NPLs. Subsequently, Spanish banks' NPLs as a share of gross lending has almost halved. Chart I-4Ireland Looked Worse Than Italy##br## For NPLs As A Share Of Loans Chart I-5Ireland Looked Worse Than Italy ##br##For NPLs As A Share Of Capital Compared to Ireland and Spain, Italy's avoidance of outright crisis (thus far) appears a blessing. Unfortunately, it is now a curse. In waiting so long, Italy cannot follow Ireland, Spain, the U.K. and the U.S. in their escapes from their banking woes. The EU Bank Recovery and Resolution Directive (BRRD), which came into full force on January 1 2016, has blocked the bailout escape route. The BRRD does allow state intervention in a banking crisis. But the overarching aim is to protect banks' critical functions and stakeholders, specifically: payment systems, taxpayers and depositors. Therefore, in a banking crisis "other parts may be allowed to fail in the normal way... after shares in full... then evenly on holders of subordinated bonds and then evenly on senior bondholders." For bank investors, this would constitute the 'deep-V' cure: likely intense pain up-front albeit with much better long-term prospects thereafter. Alternatively, without a crisis, the process to recognise and expunge NPLs would be largely up to the private sector and markets. But a long chain of events from the repossession of assets under bankruptcy law, to valuation, to full divestment from the banks' balance sheets could take years. Indeed, the Chart of the Week shows a striking parallel between Italian bank profits and Japan's 'zombie' bank profits, if we lag the Japanese experience by 17 years. Japan perfectly illustrates this alternative 'long-L' cure: no outright crisis, just a long and seemingly never-ending struggle back to normal health. Either way, absent any further information, we would lean against any knee-jerk rally in Italian banks that followed a yes vote on December 4. What Happens When Bond Yields Spike? Turning to the broader financial markets, a bigger concern is the impact that sharply higher bond yields will have on growth and/or on risk-asset valuations. Higher long-term borrowing costs depress credit growth as captured in the credit impulse (Chart I-6 and Chart I-7). A depressed credit impulse then almost always drags down subsequent GDP growth. The recent spike in U.S. 15-year and 30-year mortgage rates has already caused mortgage refinancing applications to plunge by 40% since July (Chart I-8). Chart I-6Higher Bond Yields Depress##br## Credit Growth In Europe... Chart I-7...And In ##br##The U.S. Chart I-8Mortgage Applications##br## Have Plunged Prior to the current incidence, a 50bps rise in the JPM Global Government Bond Yield in just 3 months has occurred only eight times this century (Chart I-9). Table I-1 lists those eight occasions and the subsequent 3-month performance of the equity market. On three out of the eight occasions, the equity market rose modestly, but on the other five it fell. Chart I-9The Bond Yield Has Spiked Table I-1What Happens When Bond Yields Spike? But perhaps the most interesting finding is that on all eight occasions, the equity market's subsequent 3-month performance consistently deteriorated, on average by -7%, compared to the preceding 3-month performance. For reference, today's preceding 3-month performance is just 0.7%. Given this evidence, it is prudent to retain a cautious stance to risk-assets on a 3-month horizon. Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com Fractal Trading Model* The Italy versus France sovereign bond underperformance indicates excessive pessimistic groupthink. However, in this instance we would wait until after Italy's December 4 referendum on constitutional reform before initiating the countertrend trade. For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-10 * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields Interest Rate Chart II-5Indicators To Watch##br## - Interest Rate Expectations Chart II-6Indicators To Watch ##br##- Interest Rate Expectations Chart II-7Indicators To Watch##br## - Interest Rate Expectations Chart II-8Indicators To Watch ##br##- Interest Rate Expectations

My colleague Dhaval Joshi, Senior Vice President of BCA’s European Investment Strategy, has penned an excellent update on the upcoming Italian constitutional referendum. Dhaval argues that the market is mispricing risks emanating from the referendum. Not all voters who reject the plebiscite are Euroskeptic. In fact, many will vote against the referendum precisely because it removes checks and balances and increases the odds of an anti-establishment party forming a government. Geopolitical Strategy group agrees with Dhaval and has made a similar point in our November Monthly Report. Our September Special Report also posited that Italy cannot easily disentangle itself from European institutions due to its own incomplete unification. This is not to say that Italy is not a risk to the stability of the euro area. There are plenty of reasons to worry, starting with the banking system, which Dhaval addresses in his missive. However, the market’s obsession with the referendum is overdone and thus presents an investment opportunity. I hope you enjoy the European Investment Strategy report and I encourage you to take a look at Dhaval’s research closely, if you are not already a subscriber. Kindest regards, Marko Papic Highlights An Italian referendum 'no' is not really revolting. Some people are voting no for no change to the current constitution's vital checks and balances. Lean against any knee-jerk widening of the Italian sovereign yield spread versus France that followed a no vote. Lean against any knee-jerk rally in Italian banks that followed a yes vote. A 50bps spike in the JPM Global Government Bond Yield in just 3 months is normally a bad omen for risk-asset performance. Retain a cautious stance to risk-assets on a 3-month horizon. Feature After shock victories for Brexit and Donald Trump at the polls, a 'no' vote in Italy's December 4 referendum on constitutional reform would be the next worrying sign of a growing grassroots revolt against the establishment. Or would it? An Italian 'No' Is Not Really Revolting The votes for Brexit or Donald Trump were clearly votes for change. At first glance, an Italian no would also look like a revolt, with the potential to trigger political uncertainty and instability in the euro area's third largest economy. Chart of the WeekItalian Banks Are Tracking Japanese 'Zombie' Banks The truth is more nuanced. Clearly, some Italians are voting no to reject Prime Minister Renzi. But others - including former Prime Minister Mario Monti - are voting no for no change. These voters want to leave in place the current constitution's vital checks and balances. If Italians vote yes to constitutional reform, the upper house of parliament - the Senate - would be relegated to an advisory chamber. Meanwhile, an already approved new electoral law for the lower house of parliament - the Chamber of Deputies - hands an automatic 55 percent majority of seats to the largest party. Some people fear that this combination would amount to excessive executive power. So they are voting no to mitigate the danger. Granted, a no vote might also force Renzi to resign, but this would not necessarily trigger new elections. President Sergio Mattarella would likely explore options for a new government - perhaps a technocratic government - which the parties in the current governing coalition have a strong incentive to support until the next elections are due in 2018. Even if there were early elections, it is improbable that they would result in a government led by the populist 5 Star Movement. If 5 Star was the largest party, it would hold a 55 percent majority of seats in the lower house, but only 30 percent in the upper house, in proportion to its popular vote share (Chart I-2). Therefore, it could not form a government. Under the current constitution, the government needs the support of both houses. The irony is that a yes vote - by giving the executive excessive powers - would make it more likely for a populist party like 5 Star to form a government in 2018 or beyond. Still, even this might prove a tall order. Italy's constitutional court is reviewing the electoral law change that gives 55 percent of lower house seats to the largest party. The court will likely demand more proportionality, making it hard for any one party to win an outright majority. This means more coalition governments, which 5 Star rejects. Hence, an Italian no will not be the equivalent of the Brexit vote or U.S. election of Donald Trump. Fears that it will unleash a dangerous phase of populism and political instability in Italy are overblown. Yet in the last three months, the Italian sovereign yield spread has widened sharply versus France (Chart I-3). Note also that the 65-day fractal dimension of the Italy versus France sovereign bond performance is close to its technical limit, indicating excessive pessimistic groupthink. Chart I-2The 5 Star Movement Could Not Form A ##br##Government Under The Current Constitution Chart I-3Italy's Political Risk Premium Has ##br##Increased, But Is It Justified? If December 4 brings a no vote in the Italian referendum combined with the election of a far-right President of Austria - whose role is largely ceremonial - the knee-jerk market response might still be fright. In which case, a further widening in the Italy/France yield spread would be a tactical entry opportunity, given that political risk is overstated. Fixing Italian Banks Needs A 'Deep-V' Or A 'Long-L' The real question in Italy is not about an imminent populist backlash. The real question is what does the cure for Italy's banking malaise look like? The answer is either a 'deep-V', meaning a banking crisis forces a quick workout; or a 'long-L', meaning no banking crisis but a very long struggle back to normal health. As an investor, neither seems particularly appealing. Italy's banking malaise has built up stealthily, generating frequent financial tremors but without an outright crisis. In contrast, the housing-related credit booms in Ireland, Spain, the U.K. and the U.S. did eventually cause housing busts and full-blown financial crises - requiring urgent government-led and central bank-led bailouts. Today, Italian banks' non-performing loans (NPLs) account for 18% of gross lending, and NPLs net of provisions equal 85% of equity capital. A few years ago, Irish banks looked even worse. Irish NPLs peaked at 25% of gross lending in 2013 and net NPLs peaked at 100% of equity capital. Following government bailouts Irish banks have recovered well (Chart I-4 and Chart I-5). Likewise, the Spanish government created a 'bad bank' in 2012 to offload bank NPLs. Subsequently, Spanish banks' NPLs as a share of gross lending has almost halved. Chart I-4Ireland Looked Worse Than Italy##br## For NPLs As A Share Of Loans Chart I-5Ireland Looked Worse Than Italy ##br##For NPLs As A Share Of Capital Compared to Ireland and Spain, Italy's avoidance of outright crisis (thus far) appears a blessing. Unfortunately, it is now a curse. In waiting so long, Italy cannot follow Ireland, Spain, the U.K. and the U.S. in their escapes from their banking woes. The EU Bank Recovery and Resolution Directive (BRRD), which came into full force on January 1 2016, has blocked the bailout escape route. The BRRD does allow state intervention in a banking crisis. But the overarching aim is to protect banks' critical functions and stakeholders, specifically: payment systems, taxpayers and depositors. Therefore, in a banking crisis "other parts may be allowed to fail in the normal way... after shares in full... then evenly on holders of subordinated bonds and then evenly on senior bondholders." For bank investors, this would constitute the 'deep-V' cure: likely intense pain up-front albeit with much better long-term prospects thereafter. Alternatively, without a crisis, the process to recognise and expunge NPLs would be largely up to the private sector and markets. But a long chain of events from the repossession of assets under bankruptcy law, to valuation, to full divestment from the banks' balance sheets could take years. Indeed, the Chart of the Week shows a striking parallel between Italian bank profits and Japan's 'zombie' bank profits, if we lag the Japanese experience by 17 years. Japan perfectly illustrates this alternative 'long-L' cure: no outright crisis, just a long and seemingly never-ending struggle back to normal health. Either way, absent any further information, we would lean against any knee-jerk rally in Italian banks that followed a yes vote on December 4. What Happens When Bond Yields Spike? Turning to the broader financial markets, a bigger concern is the impact that sharply higher bond yields will have on growth and/or on risk-asset valuations. Higher long-term borrowing costs depress credit growth as captured in the credit impulse (Chart I-6 and Chart I-7). A depressed credit impulse then almost always drags down subsequent GDP growth. The recent spike in U.S. 15-year and 30-year mortgage rates has already caused mortgage refinancing applications to plunge by 40% since July (Chart I-8). Chart I-6Higher Bond Yields Depress##br## Credit Growth In Europe... Chart I-7...And In ##br##The U.S. Chart I-8Mortgage Applications##br## Have Plunged Prior to the current incidence, a 50bps rise in the JPM Global Government Bond Yield in just 3 months has occurred only eight times this century (Chart I-9). Table I-1 lists those eight occasions and the subsequent 3-month performance of the equity market. On three out of the eight occasions, the equity market rose modestly, but on the other five it fell. Chart I-9The Bond Yield Has Spiked Table I-1What Happens When Bond Yields Spike? But perhaps the most interesting finding is that on all eight occasions, the equity market's subsequent 3-month performance consistently deteriorated, on average by -7%, compared to the preceding 3-month performance. For reference, today's preceding 3-month performance is just 0.7%. Given this evidence, it is prudent to retain a cautious stance to risk-assets on a 3-month horizon. Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com Fractal Trading Model* The Italy versus France sovereign bond underperformance indicates excessive pessimistic groupthink. However, in this instance we would wait until after Italy's December 4 referendum on constitutional reform before initiating the countertrend trade. For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-10 * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. Fractal Trading Model Fractal Trading Model

Recommended Allocation The Meaning Of Trump Sudden large shocks in markets are rare. But the election of Donald Trump as U.S. President is one such. After a shock of this magnitude, markets tend initially to overreact, then correct, before settling on a new course. Market action since November 9th has caused many asset prices to overshoot short term. It is likely that U.S. bond yields, inflation expectations, the performance of bank and materials stocks, and the U.S. dollar (Chart 1) will correct over the next month or so, perhaps triggered by the Fed's likely rate hike on December 14th or simply by shifting expectations for Trump's economic policies. But what is the likely long-term course, which should set our asset allocation for the next 6 to 12 months? We think investors should take Trump at least partly at his word when he says he will enact tax cuts and increase infrastructure investment. BCA's Geopolitical Strategy service sees few constraints on Trump from Congress in the short term.1 The OECD in its latest Economic Outlook has given its imprimatur, arguing that "a stronger fiscal policy response is needed," and estimating that U.S. fiscal stimulus could add 0.1 percentage point to global growth next year and 0.3 points in 2018.2 If such a policy boosted growth and inflation, it would be negative for bonds. The only question, with 10-year U.S. Treasury bond yields having already risen by almost 100 bps since July, is how much of this is priced in. In the long run, government bond yields are broadly correlated with nominal GDP growth (Chart 2). In H1 2016, U.S. nominal GDP growth was 2.7%, and for 2016 as a whole probably about 3.2%. If it picks up to 4-5% in 2017 (2.5-3% real, plus inflation of 1.5-2%), an additional rise of 50-100 bps in the 10-year yield would not be surprising (though ECB and BoJ asset purchases might somewhat limit the rise in yields). Moreover, growth was already accelerating before Trump's victory. The effects of 2015's commodity shock and industrial and profits recessions have passed, with U.S. Q3 GDP growth revised up to 3.2% and the Fed's NowCasting models suggesting 2.5%-3.6% for Q4. The Citi Economic Surprise Index has surprised on the upside in recent weeks both in the U.S. and Europe - though not in emerging markets (Chart 3). And the Q3 earnings season in the U.S. was well above expectations, with EPS coming in at +3.3% YoY (compared to a consensus forecast pre-results of -2.2%). Analysts' forecasts for 2017 EPS growth are a comparatively modest 11%. Chart 1Some Short-Term Overshoots Chart 2Bond Yields Relate To Nominal Growth Chart 3Growth Was Already Surprising On The Upside But whether this new world will be positive for equities is harder to answer. Trump's unpredictability raises policy uncertainty: how much emphasis, for example, will he put on trade protectionism or confrontational foreign policy? This should raise the risk premium. The Fed's response will also be key. Futures have now priced in the rate hike in December and (almost) the two further rate hikes in the Fed's dots for 2017 (Chart 4). But the market still sees the long-term equilibrium rate (as expressed in five-year five-year forwards) as only just over 2%, compared to the Fed's 2.9%. And, although Janet Yellen has suggested that the Fed will act only after Trump's policies take effect ("We will be watching the decisions that Congress makes and updating our economic outlook as the policy landscape becomes clearer," she said), if core PCE inflation continues to pick up in 2017 beyond the current 1.7% and a strong stimulus package is implemented, the Fed might accelerate its rate hikes. More worryingly, Trump's fundamental views on monetary policy are unknown: does he, as a businessman, like low rates, or will he listen to his "hard money" advisers who believe the Fed has been too lax? Since he can appoint six FOMC governors in his first year in office, he will be able to influence monetary policy. Too fast a rise in Fed rates would be negative for equities. On balance, in this environment we see equities outperforming bonds over the next 12 months. It is unusual for the stock-to-bond ratio to decline outside of a global recession (Chart 5) - and, with the extra boost from fiscal policy (with Trump possibly joined by Japan, the U.K., China and others), a recession is unlikely over our forecast horizon. Chart 4Market Has Priced In 2017 Fed Hikes - ##br##But Not The Long-Term Chart 5Stocks Don't Often ##br##Underperform Outside Recession Accordingly, we are raising our recommendation for global equities to overweight, and lowering bonds to underweight. The problem is timing: we recognize that there may be a better entry point over the next couple of months. Some investors may, therefore, want to implement the change gradually. In addition, some recent market moves are not fundamentally justified: for example, we cannot see how the materials sector would be a significant beneficiary from a Trump fiscal stimulus. We plan to make further detailed adjustments to our equity country and sector recommendations and bond-class recommendations in the next Quarterly Portfolio Update, to be published on December 15th. Currencies: Stronger U.S. growth and tighter monetary policy suggest that the USD will continue to appreciate. The dollar looks somewhat expensive but is still well below the peak of overvaluation at the end of previous bouts of strength in 1985 and 2002. The Bank of Japan's policy of capping the 10-year JGB yield at 0% has worked well (pushing the yen down by 12% against the dollar in the past two months) and, as rates elsewhere rise, this implies further long-run yen weakness. The euro is likely to weaken less, with eurozone growth recently surprising on the upside and the ECB therefore likely to reconsider the amount of asset purchases at some point next year, though probably not at its meeting on December 8th. Emerging market currencies continue to look particularly vulnerable. Equities: In common currency terms, U.S. equities are more attractive than European ones. In local currency terms, however, the call is closer since the strong dollar will depress U.S. earnings relative to those in Europe, and an acceleration of global economic growth should help the more cyclical eurozone stock market. On the other hand, Europe faces structural issues, such as the chronically poor profitability of its banking system, and political risk from a series of upcoming elections (starting with the Italian referendum on December 4th). We continue to like Japan (on a currency hedged basis) and expect that the BoJ's policy will be bolstered by government fiscal and employment policies. We remain underweight on emerging markets. They have always been vulnerable during periods of dollar strength, and political side-effects from their bout of economic weakness in 2011-5 are starting to spread, recently to Turkey, Malaysia, India, Brazil, Korea and South Africa. Fixed Income: The risk of tighter Fed policy and higher yields suggest investors should remain underweight duration. We have liked U.S. TIPS over nominal bonds all year and, with 10-year breakeven inflation still only at 1.8%, they remain attractive in the current environment. We reduced high-yield bonds to neutral on September 30th, on the grounds that investors were no longer being sufficiently compensated for default risk: they have subsequently given -3% return, while equities rallied. We recommend investment grade credits for those investors who need to pick up yield (Chart 6). Commodities: After the OPEC agreement on production cuts, we expect the oil price to move towards $55 in the first few months of 2017 as inventories are drawn down. Over the longer run the risk is to the upside as a dearth of new projects, following cancellations last year, will tighten the supply/demand balance. Metals prices have strengthened since Trump's victory, with the CRB Raw Industrials Index up sharply (Chart 7). This makes little sense. Trump's stimulus will be centered on tax, not infrastructure. China remains a far more important factor: the U.S. represented only 7% of global steel consumption in 2015, for example, compared to 43% for China. And China's recent stimulus is running out of steam. Chart 6Yield On Investment Grade Credits ##br##Still Attractive Chart 7Trump Shouldn't Have ##br##This Much Effect On Metals Prices Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com 1 Please see Geopolitical Strategy Special Report,"U.S. Election: Outcomes and Investment Implications," dated November 9, 2016, available at gps.bcaresearch.com. 2 Please see OECD Global Economic Outlook, November 2016, available at http://www.oecd.org/economy/outlook/economicoutlook.htm. Recommended Asset Allocation

Highlights U.S. bond yields and the U.S. dollar will rise further. Consistently, EM currencies and local bonds will continue selling off. There is meaningful downside in EM exchange rates. We recommend short positions in the following basket of EM currencies versus the U.S. dollar: KOR, MYR, IDR, TRY, ZAR, BRL, COP and CLP. Within domestic bond portfolios, overweight low-beta defensive markets as well as Russia and Mexico. Our underweights are Turkey, South Africa, Malaysia and Indonesia. The latest exponential rise in commodities prices on Chinese exchanges is an unsustainable speculative frenzy. Feature Emerging market (EM) risk assets will likely continue to be driven by both rising U.S. bond yields and a strong U.S. dollar over the next two months or so. Beyond the next couple of months, the focus of the markets will likely switch to China: renewed weakness in growth and possible instability in its financial markets, with negative implications for China plays globally and for commodities prices in particular. The combination of these two negative forces will lead to a considerable drop in EM currencies in the next six months or so. In turn, EM currency depreciation will trigger broad liquidation of EM risk assets. BCA's Emerging Markets Strategy service believes that EM risk assets will continue to sell off in absolute terms, and underperform their DM/U.S. peers. EM Local Bonds The total return (including carry) index of JPM GBI-EM1 local currency bonds in U.S. dollar terms has rolled over at a critical resistance level (Chart I-1). The total return index of EM local bonds has also relapsed relative to the total return of 5-year U.S. Treasurys, failing to break above its long-term moving average (Chart I-1, bottom panel). Consistently, domestic bond yields have troughed at important technical levels in several key countries such as Brazil, Turkey, Colombia, Russia, South Africa and Malaysia (Chart I-2A and Chart I-2B). Chart I-1EM Local Bonds' Total ##br##Return In US$: Failed Breakout Chart I-2AHave EM Domestic ##br##Bond Yields Bottomed? Chart I-2BHave EM Domestic ##br##Bond Yields Bottomed? In short, EM local bonds are exhibiting negative technical dynamics that corroborate our downbeat fundamental analysis. Consequently, we believe the total return JPM GBI-EM index in U.S. dollar terms will drop to new lows for the following reasons: Currency swings are responsible for most of the fluctuations in EM local bond total returns. As we have elaborated numerous times and re-assert in this report, the outlook for EM exchange rates remains gloomy. Foreign holdings of EM local currency bonds are substantial (Table I-1). Even though there have been improvements in a few countries, current account and fiscal deficits generally remain wide in the majority of developing nations (Chart I-3A and Chart I-3B). In other words, a number of EM economies are still at risk from a slowdown in foreign funding. Table I-1Foreign Holdings Of EM Local Bonds Chart I-3ACurrent Accounts And Fiscal Deficits Chart I-3BCurrent Accounts And Fiscal Deficits Chart I-4U.S. And EM Local Yields Notably, the bar for exchange rate depreciation is very low in EM economies with current account deficits. It takes only a reduction in net capital and financial inflows - i.e., net outflows are not necessary - for these countries' currencies to depreciate significantly. As net foreign funding diminishes, exchange rates of countries with current account deficits should weaken and interest rates should rise in order to compress domestic demand, which in turn would equalize the current account deficit to net inflows in capital and financial accounts. Finally, the spread of EM local bonds (the yield for GBI-EM global diversified index) over duration-matched (5-year) U.S. Treasury yields has not risen much (Chart I-4). Heightened risks in EM currencies warrant higher local bond yield spreads over U.S. Treasurys. Bottom Line: Absolute return investors should stay away from EM local currency bonds. U.S. Bond Yields And The Dollar: More Upside We expect U.S./DM bond yields to keep rising as re-pricing in global fixed income markets continues. The decline in DM bond yields in recent years until the latest selloff was enormous, and some sort of mean reversion should not come as a surprise. Our bias is that this selloff will likely continue until sometime in January, when U.S. President-elect Donald Trump takes office. This riot in the bond market could, in retrospect, resemble a typical "sell the rumor, buy the news" pattern. In other words, by the time President-elect Trump takes office, a lot of bad news will already be priced into the U.S. bond markets, creating a buying opportunity. In our July 13 Weekly Report,2 we argued that: "In the U.S., the combination of a healthy labor market and a heavily overbought fixed-income market have created the backdrop for a material rise in U.S. interest rate expectations/bond yields. As U.S. rate expectations climb, the U.S. dollar should gain support. This in turn will create headwinds for EM currencies and other EM risk assets." Then, we reiterated this view in our July 27 Weekly Report: "Nowadays, there is little talk in the investment community about a bond bubble and the potential for much higher bond yields. Indeed, "lower for longer" has begun to dominate the investor lexicon. This is a sign that many G7 bond bears have likely capitulated. Investor consensus on bonds has become quite bullish, and many investors are long duration. When many bears capitulate, the odds of a market selloff inevitably rise. "Importantly, the increase in G7 bond yields might not be gradual as many expect because of the following: with yields at such low levels, bonds' duration is high and price changes become very sensitive to changes in yield... Such (large) price changes (drops) would amount to large losses for bond investors, and forced selling could intensify. As a result, the unwinding of long positions could be abrupt and volatile." For now, odds are that U.S. bond yields will rise further. Given global bond funds have seen massive inflows in recent years, the latest drop in prices of various bonds has been substantial and will likely trigger withdrawals and redemptions from bond funds, prompting forced selling. This is true for all types of bond portfolios, including DM government and corporates, EM credit (U.S. dollar bonds) and EM local currency bonds. U.S. bond yields are still low, even from the perspective of the past several years, and the market-implied terminal fed funds rate is still 80 basis points below the median projection of the Federal Open Market Committee's longer-run rate (Chart I-5). Given that U.S. interest rate expectations are not high at all, they will rise further (Chart I-6) as the uptrend in U.S. wages persists - driven by an already reasonably tight labor market (Chart I-7). Chart I-5U.S. Interest Rate Expectations Are Still Low Chart I-6U.S. Wage Growth Is Accelerating Chart I-7More Upside In U.S. Treasurys Yields Finally, the U.S. dollar will continue to be buoyed by rising U.S. interest rate expectations. Our composite momentum indicator for the broad trade-weighted U.S. dollar has bounced off the zero line (Chart I-8). This constitutes a strong technical confirmation of the durable bullish market trend in the dollar. Bottom Line: Odds are that the rise in U.S. bond yields is not over. As U.S. bond yields rise further, EM currencies and bonds will sell off. Long-Term EM Currency Trends We have several observations on the long-term performance of EM currencies and financial markets: In the long run, there is no guarantee that the majority of EM currencies will appreciate in real terms (adjusted for inflation differentials). In fact, even countries such as Korea and Taiwan - which have been very successful in their economic development and have tremendously grown their income per capita - have seen their real (inflation-adjusted) exchange rates depreciate over the past several decades (Chart I-9). The case for long-term appreciation in real terms is even weaker for exchange rates in countries that exhibit chronically high inflation rates and/or current account deficits. This has been true for many non-Asian EM currencies (Chart I-10). Chart I-8The U.S. Dollar Is ##br##In A Genuine Bull Market Chart I-9Long-Term Currency ##br##Downtrends In Korea And Taiwan Chart I-10EM Currency Trends: ##br##A Long-Term Perspective Importantly, most losses to foreign investors in EM financial markets often occur via currency depreciation. This is even truer in the current bear market downtrend. The JPM ELMI+ currency total return index (including cost of carry) seems to be about to break down (Chart I-11). In EM ex-China, the real effective exchange rate is still elevated (Chart I-12). Given their poor productivity growth outlook, the real effective exchange rates will be inclined to depreciate. Chart I-11EM Currency Return With Cost ##br##Of Carry Versus U.S. Dollar Chart I-12Weak Productivity Means ##br##Further Currency Depreciation To limit the upside in domestic interest rates - both in bond yields and interbank rates - many developing nations' central banks will inject more local currency liquidity into their respective systems.3 This might help cap local interest rates, but is bearish for their currencies. The Turkish central bank has been among the most aggressive in this disguised money printing, and not surprisingly the value of its currency has collapsed (Chart I-13). There is no long-term history for EM currencies, as before 1998 most developing nations' exchange rates were pegged. Yet when one examines EM equities' relative performance against the S&P 500, it emerges that there is no single EM bourse that has outperformed U.S. stocks on a consistent basis in the very long run. Chart I-14A and Chart I-14B demonstrate that among 11 EM equity markets that have a long-term history, none have outperformed the S&P 500 over the past 30-35 years. Chart I-13Turkey's Central Bank Has Been ##br##Pumping Local Currency Into The System Chart I-14AEM Equities Versus The S&P 500: ##br##A Long-Term Perspective Chart I-14BEM Equities Versus The S&P 500: ##br##A Long-Term Perspective This goes to reveal that the starting point of underdevelopment and the mark "emerging" does not guarantee consistent outperformance even in the long run. In fact, EM's relative performance against the U.S. has followed multi-year cycles, and we believe the current bear market and underperformance is not yet over. While EM underperformance is long in duration, economic and financial adjustments remain incomplete. DM QE programs and China's still-growing credit bubble have delayed the adjustment. As a rule, the longer a financial or economic imbalance/excess lingers, the more protracted the adjustment will be. Bottom Line: EM exchange rates will continue depreciating. We recommend short positions in the following basket of EM currencies versus the U.S. dollar: KRW, MYR, IDR, TRY, ZAR, BRL, COP and CLP. For a complete list of our open currency and fixed-income trades please refer to page 18. Country Allocation For EM Local Bond Portfolios Chart I-15 demonstrates the relationship between developing countries' foreign funding requirements and their real (inflation-adjusted) local bond yields. The foreign funding requirement is calculated as the sum of the current account deficit and foreign debt service obligations over the next 12 months. We use inflation-linked (real) bond yields for markets where they are available. In other cases, we subtract the headline inflation rate from nominal bond yields to derive the real one. Chart I-15Real Bond Yields And Foreign Funding Requirements: A Cross Country Comparison The higher the foreign funding requirement, the higher the real yield must be to attract foreign capital, all else equal. On this diagram, the value pockets are Brazil (its real yield of 6.3% offers the best value by far), Indonesia, Russia and India. Domestic real yields in these countries are relatively high compared to their foreign funding requirements, which is a proxy for exchange rate risk. In contrast, Turkey, Chile, Colombia, Hungary and Malaysia have low real yields relative to their large foreign funding requirements. However, there are other factors that are shaping local yields. For example, Brazilian real yields look very attractive on this matrix because the latter does not account for public debt dynamics. The fiscal dynamics in Brazil are dreadful.4 On the contrary, Chilean local bonds appear expensive, but the country's fiscal outlook is very healthy. After considering all factors that affect local bond yields as well as incorporating the currency outlook, we recommend the following allocations: Overweight Korea, Thailand, Poland, Hungary, the Czech Republic, Russia and Mexico (Chart I-16). For investors who can invest in Chinese, Taiwanese and Indian local bonds, we also recommend overweighting these markets within an EM domestic bond portfolio. Underweight Turkish, South African, Malaysian and Indonesian local currency bonds (Chart I-17). We will publish our analysis on Indonesia soon. Stay neutral on domestic bonds' total return in U.S. dollar terms in Brazil (with a negative bias because of the considerable currency risk), Chile and Colombia (Chart I-18). Chart I-16Our Recommended ##br##Overweights In Local Bonds Chart I-17Our Recommended ##br##Underweights In Local Bonds Chart I-18Local Bonds ##br##Warranting A Neutral Allocation A Word On China's Commodities Frenzy Speculative fever is running high in Chinese commodities exchanges. Frenetic commodities trading in China has seen prices skyrocket of late (Chart I-19). Prices often rise a limit during a day. We have the following observations: This stampede into commodities is a reflection of rotating bubbles in China. Mania forces rotated from property to stocks, then to corporate bonds, and then back to housing, again. It seems to be shifting into commodities now. While the mainland's industrial sector and real demand for commodities have registered gradual improvement in recent months, the sharp spike in commodities prices largely reflects speculative activity much more than real demand. In fact, net imports of base metals have been flat for the past six years (zero growth in six years), and all swings have most likely been related to inventory cycles (Chart I-20). Chart I-19The Spike In Commodities ##br##Prices Trading In China Chart I-20China: Net Import Of Base Metals Like any speculative frenzy, this is momentum-driven and will one day crash. Timing the reversal is impossible. A lot depends on policymakers' willingness to confront this speculative bubble and investor psychology. Notably, onshore corporate bond yields and swap rates have recently begun rising. As in DM bonds, the rise in yields from very low levels is causing large price drops. As and if yields rise further, losses in corporate bonds will become considerable and investors (especially ones managing retail investors' money) will head for the exits, triggering liquidation. This, along with the eventual unraveling of commodities speculation poses substantial potential risk to global, or at least EM, financial markets. Bottom Line: The latest exponential rise in commodities prices on Chinese exchanges is an unsustainable speculative frenzy that will end badly. Arthur Budaghyan, Senior Vice President Emerging Markets Strategy & Frontier Markets Strategy arthurb@bcaresearch.com 1 The JPMorgan Government Bond Index-Emerging Markets (GBI-EM) indices are emerging market debt benchmarks that track local currency bonds issued by Emerging Market governments. 2 Please see Emerging Markets Strategy Weekly Report, titled "Risks To Our Negative EM View," dated July 13, 2016. 3 Please see "EM: Is The Liquidity Upturn Genuine And Sustainable?" Parts I & II, dated November 25, 2015 and December 2, 2015, respectively. 4 Please refer to the Emerging Markets Strategy Special Report, "Brazil: The Honeymoon Is Over," dated August 3, 2016. Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights Duration: The odds of further bond bearish catalysts emerging during the next 6-12 months are still quite elevated. Maintain below benchmark duration. Global Bond Strategy: The most likely candidates for another bond bearish catalyst would be an announcement of substantial fiscal stimulus from Japan and/or a hawkish policy shift from the Fed. Investors should remain overweight core Europe, underweight U.S. Treasuries and neutral on JGBs. U.S. High-Yield: Given current spread levels and our default loss expectations, valuation in the U.S. high-yield market sends neither a strong buy nor sell signal. Feature In a U.S. Bond Strategy Special Report1 published in August we observed that, since the financial crisis, material increases in global bond yields have all been associated with a policy catalyst (Chart 1). We identified three such catalysts: the Fed's 2010 announcement of QE2, the Fed signaling its willingness to slow the pace of asset purchases in 2013, and the European Central Bank's (ECB) announcement of its own QE program in 2015. Now we can add the election of Donald Trump as a fourth catalyst that has spurred a tantrum in global bond markets. Chart 1The Four Post-Crisis Bond Tantrums The common factor that links all of these catalysts is that each causes the market to quickly re-assess its expectations about the future pace of monetary tightening. Interestingly, this re-assessment can be caused by either the announcement of a program that is perceived to be extremely stimulative or the announcement that monetary stimulus will be scaled back. Examples of the former include both the Fed's and ECB's QE announcements as well as the recent U.S. election. An example of the latter would be the 2013 taper tantrum. As in August, the goal of this report is to perform a quick survey of the major global economies in order to assess the likelihood that another bond-bearish catalyst emerges during the next 6-12 months. While we find it difficult to see a catalyst of the same scale as those shown in Chart 1, we assign high odds to the possibility that the announcement of fiscal easing in Japan will add to the bearish pressure on global bonds. We also assign high odds to the possibility that upside inflation surprises in the U.S. cause the Fed to adopt a more hawkish forward guidance, further increasing the bearish pressure on global bonds. We assign low odds to the possibility that ECB policy will contribute to the global bond selloff. U.S. Chart 2Fed Wants Breakevens To Head Higher The recent "Trump Tantrum" has sent yields sharply higher, and expectations priced into the U.S. bond market are now not far from the Fed's median rate hike expectations, especially at the short-end of the curve (Chart 2). In the U.S., the next most likely catalyst for sharply higher global bond yields would be the Fed signaling that it will adopt a quicker pace of rate hikes. Specifically, the Fed would need to cease revising its funds rate forecasts lower - which has been the pattern for the last few years - and start revising them higher. While the market was quick to price-in the likelihood of greater fiscal stimulus and rising deficits under the incoming government, the Fed will take a more cautious approach. In fact, with inflation still below target (Chart 2, bottom panel) and market-based measures of inflation compensation still depressed, the Fed will be in no rush to signal a more hawkish policy stance. We expect the Fed will follow through with an expected rate increase in December, but that the median expectation will continue to call for only two more hikes in 2017. The Fed is only likely to shift toward a more hawkish policy stance once inflation expectations are more firmly anchored around levels consistent with the Fed's inflation target. This corresponds to a range of 2.4% to 2.5% on the 5-year, 5-year forward TIPS breakeven inflation rate (Chart 2, second panel). Assuming that U.S. economic growth continues to accelerate into next year, as we expect, then the 5y5y TIPS breakeven rate could reach this target sometime in the middle of 2017. At that point, a more hawkish Fed policy becomes more likely. In the meantime, while the "Trump Tantrum" is likely to take a pause in the near-term (next 1-2 months), it may not have run its course just yet. If U.S. growth is strong in 2017 and the Trump administration appears to be making progress implementing its more stimulative policies, then the Treasury curve will likely resume its bear-steepening trend in the first half of next year.2 Euro Area Chart 3Strong Growth, But Plenty Of Slack According to the OECD and others, including the European Commission and ECB, trend GDP growth in the Eurozone is below 1%. In fact, most estimates center around 0.7%. This means that as long as GDP growth is maintained above these levels we should expect the labor market to continue to tighten. At least for now, the data suggest that growth is likely to remain well above trend. Led by gains in both the services and manufacturing indexes, the euro area's composite PMI jumped from 53.3 to 54.1 in November. The composite PMI has a good track record of leading European GDP growth (Chart 3), and the current reading is consistent with GDP growth of 2%. Despite strong growth, the ECB's policy stance is likely to remain accommodative for quite some time and is unlikely to spur a global bond tantrum within our 6-12 month investment horizon. The fact that core inflation remains below 1% (Chart 3, panel 3) tells us that the output gap in the euro area is still very wide. It will take a prolonged period of strong growth for the output gap to close and for inflationary pressures to mount. In prior cycles inflation has not begun to accelerate until the unemployment rate was below 9% (shaded regions in Chart 3). An announcement from the ECB that it will cease its asset purchase program because the economy has made adequate progress toward its economic and inflation goals would likely spur a large rise in global bond yields. However, this is unlikely to occur until the unemployment rate is below 9% and inflation is in an uptrend. As we argued in a recent Global Fixed Income Strategy report,3 the ECB will be able to alter the rules regarding the quantity of bonds available for purchase as is necessary to keep the program in place. Japan The Bank of Japan (BoJ) recently switched to a policy framework that involves targeting a level of yields as opposed to a quantity of purchases. In our view, this sends a pretty strong signal that monetary policy is close to being exhausted and that fiscal policy must take up the baton of Abenomics. While the timing and amount of any additional fiscal spending is not clear, it is probably necessary if policymakers are serious about reaching their 2% inflation goal. Chart 4Policy Action Required In Japan At present, the Japanese Diet is currently deliberating the third revision to the second supplementary budget and government officials have signaled that there will be more coordination between monetary and fiscal policy in the future. The government is also debating ways to boost household income, including raising government wages, lifting the minimum wage and providing tax incentives for the private sector to be more generous on the wage front. While any fiscal measures would not spur an increase in nominal JGB yields (because the BoJ will retain the cap), they would spur an increase in inflation expectations and a decline in real yields (Chart 4). We also think that the reflationary impulse would be felt by bond markets in the rest of the world, and that large enough fiscal stimulus from Japan would pressure global bond yields higher even though JGBs remain capped. Admittedly, the cap on nominal JGB yields would limit the contagion from Japanese fiscal stimulus to the rest of the global bond market. As would the impact of a depreciating yen relative to the euro and U.S. dollar. However, we also suspect that the shift toward greater fiscal stimulus in both the U.S. and Japan would cause investors to revise their global growth expectations higher, and that this impact would dominate in terms of the impact on global bond yields. Investment Conclusions The odds of further bond bearish catalysts emerging during the next 6-12 months remain quite elevated. The most likely candidates would be an announcement of substantial fiscal stimulus in Japan and/or a hawkish policy shift from the Fed. The ECB is unlikely to contribute to the bearish pressure on global bonds during the next 6-12 months. As such, we continue to recommend a below benchmark duration stance on a 6-12 month horizon. In global bond portfolios, investors should remain overweight core Europe, underweight U.S. Treasuries and neutral JGBs. Valuation & Expected Returns In U.S. High-Yield A commonly used tool for assessing value in the high-yield bond market is a default-adjusted spread. That is, we formulate an expectation for default losses during our investment horizon and compare it to the spread that is currently on offer. If the current spread is elevated compared to our expectation for default losses then the default-adjusted spread is high and we would see good value in high-yield bonds relative to equivalent-duration Treasuries. This week we examine two different formulations of a default-adjusted spread for the U.S. high-yield market and test how well each corresponds to excess junk returns. The first measure we look at is a true ex-ante measure. It relies only on data that are available in real time, and can therefore be used as part of a trading strategy. Specifically, our ex-ante default-adjusted spread is calculated as the average option-adjusted spread from the Bloomberg Barclays U.S. High-Yield index less an expectation of default losses for the subsequent 12 month period. Expected default losses are calculated by taking the Moody's baseline forecast for the U.S. speculative grade default rate during the next 12 months and multiplying it by 1 minus our forecast of the recovery rate for this same period. We forecast the recovery rate based on its historical relationship with the default rate. The second measure we examine is an ex-post default-adjusted spread. In this case we look at the average spread of the index less actual default losses that are realized during the subsequent 12 months. As such, this measure can only be calculated after the fact. Comparing the ex-ante and ex-post measures, we see that both tend to reside within a range of 200 to 300 basis points. However, the ex-post measure periodically shows a negative value while the ex-ante measure is more often above 300 bps (Chart 5). This tells us that when forecasting default losses it is more common to underestimate default losses, rather than overestimate them. Chart 5Distribution of Default-Adjusted Spreads Over Time The next thing we look at is how closely each measure aligns with high-yield excess returns (Charts 6 & 7). Our ex-ante measure explains 34% of the variation in high-yield excess returns since 2002 (when our sample begins). Predictably, the ex-post measure, which removes the error surrounding the default loss forecast, explains a greater proportion of the variation in excess junk returns (53%). Our sample period is also longer for the ex-post measure, beginning in 1995. Chart 612-Month Excess High-Yield Returns Vs.##br## Ex-Ante Default-Adjusted Spread (2002 - Present) Chart 712-Month Excess High-Yield Returns Vs. ##br##Ex-Post Default-Adjusted Spread (1995 - Present) The current average option-adjusted spread for the High-Yield index is 459 bps. If we incorporate the Moody's baseline forecast for the default rate during the next 12 months (4.1%) and our forecast for the recovery rate (39%), then we calculate an ex-ante default-adjusted spread of 210 bps. Using the relationship in Chart 6, this translates into an expected 12-month excess return of -26 bps. If we assume there is no error in our forecast then we can use the relationship in Chart 7. In that case, our expected 12-month excess return would be +55 bps. Of course, that exercise imposes a linear relationship between excess returns and the default-adjusted spread and doesn't consider that there is considerable variation in actual excess returns around this trendline. For that reason, in Charts 8 & 9 we split both our default-adjusted spread measures into intervals of 50 basis points. For each interval we display the average 12-month excess return along with a 90% confidence interval for where those returns are likely to fall. Chart 812-Month High-Yield Excess Returns & 90% Confidence Intervals: ##br##Ex-Ante Default-Adjusted Spread Chart 912-Month High-Yield Excess Returns & 90% Confidence Intervals:##br## Ex-Post Default-Adjusted Spread Specifically, the blue dots in Charts 8 & 9 show the 12-month excess return that is earned on average when the default-adjusted spread falls into a particular interval. The top and bottom edges of the vertical lines correspond to the upper and lower limits of the 90% confidence interval. More statistics related to the 12-month excess returns that have been observed when the default-adjusted spread falls into a specific interval can be found in the Appendix to this report. The main message from these charts is that a default-adjusted spread below 100 bps is a powerful sell signal, while a default-adjusted spread above 350 bps is a powerful buy signal. Between those two thresholds the signal is less clear. Bottom Line: Given current spread levels and our default loss expectations, valuation in the U.S. high-yield market sends neither a strong buy nor sell signal, but is consistent with small positive excess returns. Our inclination is to remain cautious on U.S. high-yield for the time being, but to look for opportunities to upgrade from more attractive valuations. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Special Report, "The Tantrum Theory Of Global Bond Yields", dated August 16, 2016, available at usbs.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, "Toward A Cyclical Sweet Spot?", dated November 22, 2016, available at usbs.bcaresearch.com 3 Please see Global Fixed Income Strategy Weekly Report, "The ECB's Next Move: Extend & Pretend", dated October 25, 2016, available at gfis.bcaresearch.com Appendix Table 112-Month High-Yield Excess Returns & Ex-Ante Default-Adjusted Spread Table 212-Month High-Yield Excess Returns & Ex-Ante Default-Adjusted Spread Fixed Income Sector Performance Recommended Portfolio Specification

Highlights Portfolio Strategy The rise in Treasury yields is approaching a threshold that has often caused equity market indigestion. Stay focused on current monetary conditions rather than fiscal unknowns. The bear market in lodging stocks has played itself out: take profits on an underweight position. The sell-off in home improvement retail shares is overdone, and a contrarian long position should pay off despite the backup in mortgage rates. Recent Changes S&P Hotels Index - Take profits of 3% and raise to neutral. Table 1Sector Performance Returns (%) Feature Momentum may carry the market higher in the short run, but from current valuation levels, stocks, the dollar and bond yields can only climb sustainably in tandem if a non-inflationary economic boom is taking hold. In that sense, equities appear to be taking their cue solely from the anticipated U.S. political shift while ignoring the tightening in monetary conditions and hints of emerging market financial strains. The equity market outlook hinges on a judgement call as to whether the action in the currency and Treasury yields is reflective or restrictive? There are no easy answers, but below we discuss some of the variables that influence this decision. Chart 1 shows that the 10-year Treasury yield has climbed above fair value. Equity bulls may rejoice because yields have sauntered much deeper into undervalued territory before stocks have run into trouble. The big difference this time is that the greenback is also climbing. Parallel powerful rises in both the currency and yields are rare, and typically culminate in steep market pullbacks. Importantly, most of the recent yield rise reflects an increase in inflation expectations. The real component, i.e. economic growth expectations, has been far more muted (Chart 2). Chart 1Stocks, Yields, And The Dollar##br## Can't Climb Together For Long Chart 2Inflation Expectations ##br##Are Driving Up Yields Equities shrugged off the surge in yields during the 2013 taper tantrum. However, yields never rose above fair value then, and the increase was almost entirely due to the real component rather than a rise in inflation expectations, i.e. it was more reflective than restrictive (Chart 2). Meanwhile, equities had just been through a difficult stretch in 2012 on fears the euro was going to break apart, and sovereign yields in the periphery were in the early stages of a long descent (Chart 3). In other words, there was a structural tailwind for equities. In addition, the U.S. dollar was range-bound during that period, overall profit growth was strong, business lending was picking up and corporate bond spreads stayed tight (Chart 3). The outlook today is much different. Euro area periphery yields are up sharply, EM bond spreads are flaring out, profit growth is much weaker and the U.S. is importing deflation through U.S. dollar strength (Chart 3), particularly against China and other developing market currencies. Thus, we are uncomfortable making comparisons between today and 2013 broad market resilience. The speed of upward adjustment in Treasury yields also influences equity prices. At the moment, yields are rising faster than profit growth. The overall market has typically become more volatile and often corrects when the growth in yields outpaces profit growth (Chart 4). Chart 3The 2013 Taper Tantrum##br## Is Not A Good Guide Chart 4Too Far,##br## Too Fast? The most painful equity corrections have occurred when this gauge drops below -10%, as the latter suggests that inflation expectations are increasing rapidly, warning of valuation and monetary tightening ahead. This threshold is in danger of being breached on any further rise in yields. However, if the currency continues climbing, yields are unlikely to rise much further, if at all, underscoring that the next big tactical sub-surface market move may be a recovery in yield-dependent sectors as investors begin to fret about the deflationary and profit-sapping impact of a strong dollar. Against this backdrop, we caution against getting too comfortable extrapolating market momentum, because recent gains could be erased just as quickly as they accrued if monetary conditions keep tightening. On a sub-surface basis, value is being created in interest rate-sensitive sectors and destroyed in cyclical sectors, primarily industrials, as discussed last week. Meanwhile, we maintain a domestic vs. global focus, and recommend buying into the pullback in housing stocks. Buy Home Improvement Retailers Like many other interest rate-sensitive groups, home improvement retailers (HIR) have lagged recently, fueled by the surge in bond yields, and hence, mortgage rates. We doubt this is sustainable. U.S. currency strength will refocus attention on the lack of top-line growth in global-oriented industries, which will reverse recent countertrend intra-sector capital flows, and ensure that bond yields are capped. The housing market slowed this year by most metrics (housing starts, permits, sales growth), which undermined remodeling activity. In response, building supply store sales cooled (Chart 5, bottom panel). Recent earnings reports from housing-geared industries such as appliances and furniture vendors have also disappointed. Analysts have been quick to slash both sales and earnings growth estimates (Chart 5). However, as often happens, an overreaction appears to be occurring. There is little indication of a return to punitively deflationary industry conditions. In fact, the producer price index for appliance and furniture makers has shot up in recent months, heralding stronger HIR pricing power (Chart 6, second panel). Lumber prices are also up sharply, despite U.S. dollar strength, which will boost the top-line and profit margins (Chart 6). At a fixed spread over lumber prices, the higher the latter go, the more profit earned at a constant volume sold. We continue to be encouraged by the long-term outlook. Household formation is accelerating now that the unemployment rate is below 5%. Building permits are below average levels, even excluding the housing bubble period (Chart 7). Chart 5Housing Slowdown Already Reflected Chart 6No Sign Of Deflationary Stress Chart 7Still Early In The Mortgage Cycle Consumers have only recently become comfortable taking on mortgage debt, and first time buyers represent a rising share of total home sales. Banks are ready and willing to extend mortgage credit (Chart 7, bottom panel), unlike most other credit. Ergo, housing activity still has legs. While the backup in Treasury yields will no doubt make housing somewhat less affordable, Chart 8 shows that even a 100 basis point rise would not push affordability back to average levels. Mortgage payments would still be well below the long-term average as a share of income, and effective mortgage rates are still extremely low. Therefore, we would not be surprised to see stable housing metrics in the coming months, despite the yield back up. Existing house prices are flirting with new highs (Chart 7), despite the early stage of mortgage re-leveraging, which bodes well for future house price increases. If homeowners are confident that house prices will stay solid, they will be more inclined to make home improvement investments. These factors are represented in our HIR model. The model is climbing steadily, exhibiting a rare positive divergence from relative share prices (Chart 9). Our inclination is to side with the objective message from the model. The valuation case for the group has improved markedly. The forward P/E is well below the average of the last decade and the dividend yield is now on a par with that of the broad market. Typically, a positive yield differential has been a bullish relative performance signal (Chart 10). Chart 8Higher Yields Are Not A Game Changer Chart 9Our Model Remains Firm Chart 10Discounting A Weak Housing Market Most importantly, the industry continues to generate sky-high return on equity, and free cash flow is booming. The implication is that shareholder-friendly stock buybacks and dividend increases should continue apace, especially compared with the overall corporate sector. At current valuation levels, there is room for a playable recovery in relative performance, especially if Treasury yields level off on the back of relentless U.S. dollar strength. Bottom Line: Home improvement retail (BLBG: S5HOMI - HD, LOW) stock price weakness is a buying opportunity. We recommend an above-benchmark allocation. End Of The Bear Market In Hotel Stocks The S&P hotels index has been in a relative performance bear market since late last year when we reduced it to underweight, but downside risks have diminished even though a number of players have lowered 2017 guidance and revenue per room (REVPAR) expectations. Relative value has been created by the past year of underperformance. A variety of valuation metrics show that the price ratio is plumbing recessionary-type levels (Chart 11). Most notably, the relative price/sales ratio is almost on a par with the lows during the Great Recession, when a steep contraction was anticipated for the foreseeable future. Such a dire forecast is not in the cards, even if economic growth disappoints an increasingly optimistic consensus. The plunge in net earnings revisions has not been confirmed by a downturn in hours worked. Typically, these two series move hand-in-hand (Chart 12). Instead, hours worked continue to trend higher suggesting that reduced profit guidance is bringing analyst expectations to more attainable levels rather than signaling impending doom. After all, persistent hotel construction growth means that demand needs to run hot in order to keep deflationary pressures at bay. This has been a tall order in the past year, as tight business budgets and lackluster discretionary consumer spending have kept REVPAR under wraps (Chart 13). Occupancy rates remain below previous expansionary run rates, leaving revenue per room more exposed than normal to demand soft spots. Chart 11End Of Bear Market Chart 12An Undershoot In Estimates Chart 13Slow, But Steady, Growth REVPAR could be supported by decent consumer spending. Wage growth, and thus aggregate income, are perking up, job security has risen and income expectations are on the upswing. Consumers are behaving as if income gains will be permanent, given the increase in consumer loan demand. Low fuel prices and the surge in vehicle miles driven are consistent with solid lodging outlays. The latter have recently reaccelerated, and are supporting better than market hotel pricing power (Chart 13). Importantly, hotel profit margins are no longer under extreme duress. Decent pricing power gains and an easing in the industry's total wage bill inflation have combined to support an increase in our profit margin proxy (Chart 14). All of this implies that profit conditions are stabilizing, just as valuations have been squeezed, warranting an upgrade to neutral. Why not a full shift to overweight? There are a number of factors to consider. The lodging industry is battling secular crosscurrents. On the positive side, the lodging industry has consistently managed to increase its share of total consumer spending, in real terms (Chart 15), with periodic underperformance phases, typically during recessions. This likely reflects well-timed capacity investments and strong brands. As a result, hotel pricing power has also been in a structural uptrend (Chart 15). This cycle, pricing power has lagged, consistent with subdued REVPAR gains, but hotels have still managed to aggressively grow earnings per share. While buybacks have undoubtedly played a role in this advance, EPS is following a typical pattern. In the last four decades, hotels have suffered four major recession-related earnings contractions. After each contraction, profits ultimately surpassed their previous peak by more than 75%, on average. The duration of the upcycle averaged five years. This cycle the recovery has already lasted more than six years, but hotel profits have only increased 30% from the 2007 peak. That implies substantial profit upside ahead just to reach the average, albeit pricing power will need to kick in as it has in past cycles. On the downside, consumers are still showing a penchant for spending more on essentials compared with non-essentials. The ratio of retail sales at cyclical stores to non-discretionary stores has been highly correlated with relative performance (Chart 16, top panel). Chart 14The Margin Squeeze Is Over Chart 15Structural Tailwinds... Chart 16... And Headwinds That raises some question about the latest burst of strength in lodging outlays, especially in view of the pruning in business travel budgets, as confirmed by anecdotes from recent earnings reports. BCA's capital spending model is not forecasting any improvement (Chart 16, bottom panel). Lingering in the background has been the relentless increase in lodging construction. Capacity growth represents a long-term threat to pricing power (Chart 16), over and above the threat from new entrants such as AirBnB. Expansion explains why real hotel consumer prices have not come close to hitting new highs even though real hotel spending has. Hotel capacity expansion heralds intensifying deflationary pressure. Meanwhile, hotels have sizeable global operations, exposing profitability to risks of incremental U.S. dollar strength. Consequently, we would prefer to await signs of an impending improvement in capital spending, and thus, business travel, and/or a sharp downturn in hotel construction spending, before lifting positions all the way to overweight. Bottom Line: Lift the S&P hotels index (BLBG: S5HOTL - MAR, CCL, RCL, WYN) to neutral, locking in an 3% relative performance profit since our initial underweight call nearly a year ago. A further upgrade is tempting, but awaits relief from pricing power constraints. Current Recommendations Current Trades Size And Style Views Favor small over large caps and growth over value.

Highlights Commodity prices and the dollar can occasionally rise together. The 1999-2001 and the 2005 experiences suggest a supply shock is required. If commodities were to rally alongside a strengthening dollar in 2017, this would be an oil-led move. Metals have very little potential upside as improving DM growth drains liquidity from EM economies. Favor petro currencies (CAD and NOK) relative to the antipodeans (AUD and NZD). Stay short AUD/CAD. USD/JPY is in a major bull market. However, near-term risks are to the downside. Feature It has become axiomatic among investors to assume that a dollar bull market is synonymous with a commodity bear market. While the relationships usually holds, there have been episodes where the narrow trade-weighted dollar and natural resource prices moved in tandem, not in opposite directions: 1982 to 1984, 1999 to 2001, and in 2005. The recent surge in base metals raises that possibility, but as DM economies suck in global liquidity away from EM ones, the prospect for a positive correlation between most commodities and the dollar is still remote. When Do Commodities And The Dollar Walk Together? Commodities and the dollar usually move in opposite direction. Since 1980, there has only been three episodes of consistent commodity strength despite dollar appreciation: 1982 to 1984, 1999 to 2001, and in 2005 (Chart I-1). What defines each of these episodes? In the early 1980s, the rally in commodities was concentrated outside of the energy complex. The U.S. economy was rebounding from the 1980s double-dip recession, and Japan was in the middle of its economic miracle. Their vigorous growth resulted in a large positive demand shock, boosting Japan and the U.S.'s share of global copper consumption from 34% to 37%. This undermined any harmful effect on metal prices from a rising dollar (Chart I-2). Chart I-1Commodities Can Rise ##br##Alongside The Dollar Chart I-2Early 1980s: U.S. Growth Was ##br##Able To Boost Metal Prices From 1999 to 2000, the rally in commodity was not broad based. In fact, it was concentrated in the energy sector (Chart I-3). It reflected three factors: After being decimated in 1997 and 1998, EM stock prices managed to stage a temporary rebound; one that mostly reflected bombed out equity and currency valuations. However, the muted response of non-oil commodities suggests that this rebound had little economic impact. Energy was buoyed by the vigorous growth in DM, with OECD oil consumption growing 1% annually between 1998 and 2001. Finally, as oil prices fell below US$10/bbl in late 1998 global oil production contracted sharply, plummeting by more than 4 million barrels, or 5% of total production. Not only could Saudi Arabia and Russia not withstand the pain of lower oil prices, but the latter was in the midst of a massive economic crisis that disrupted the local oil industry's ability to finance its operations. While most commodities in the 2005 episode experienced subtle upward drift, once again, energy was the true winner (Chart I-4). Supply disruptions in the Gulf of Mexico following the record-breaking 2004 and 2005 hurricane seasons contributed to removing slightly more than one million barrels from the market. Additionally, oil had captured investors' imagination, with the peak-oil theory being all the rage. This combination explains why oil was the primary beneficiary of Chinese and EM economic strength while base metals could not overcome the dollar's hurdle. Chart I-31999-2001: Commodity##br## Rally Was An Oil Rally Chart I-42005: Commodity##br## Rally Was An Oil Rally Bringing it all together, the dollar and commodities where able to rise as one in the 1980s because they responded to the same positive U.S. growth shock. However, during the 1999-2001 and 2005 commodity rallies in the face of strong dollar, the supply/demand imbalance in oil was paramount. Bottom Line: The dollar and commodity prices can occasionally move together. This happens when a supply shock affects a natural resource as important as oil, lifting its price despite the greenback hurdle. Outside of energy, in general prices still displayed little upside through these episodes. Giant Sucking Sound Our bullishness on the dollar is built on our positive outlook for U.S. growth and rates, a view only reinforced by Trump's electoral victory.1 This does not mean we expect the same boost to metal consumption that we saw in the early 1980s. Today, combined Japanese and U.S. copper consumption only accounts for 11% of global consumption. For iron ore, the U.S. represents only 4% of global consumption. Even if the U.S. were to spend $1trillion over five years on infrastructure (an extremely optimistic assumption), it will not constitute the same relative boost to global demand as the U.S. expansion during the 1980s did (Chart I-5). Additionally, metals will remain slightly oversupplied. In fact, inventories have been rising and more supply of iron ore is coming upstream in 2017, as additional Pilbara iron ore deposits are being unleashed on the markets. In the case of copper, our commodity specialists expect supply to continue to grow in the years ahead. But still, could EM lift the demand for metals enough to play the same role as the U.S. did in the early 1980s? We doubt it. When it comes to China, the current growth improvement is likely as good as it gets. The Keqiang index - a measure of industrial activity in the Middle Kingdom - is approaching post-2011 highs, but the demand for loans remains very depressed (Chart I-6). Moreover, the Chinese fiscal impulse - which has buoyed the country's economy for much of 2016 - has rolled over and is now in negative territory, suggesting that the Keqiang index will weaken in 2017. This will weigh on Chinese imports of machinery and raw materials, representing a deflationary shock for other EM. Chart I-5Metals Are About China, Not The U.S. Chart I-6China: The Best Is Behind Us At the current juncture, additional deflationary forces on EM would be an unwelcomed development. The structural headwinds plaguing EM economies are still in place. EM remain burdened by too much capacity, too much debt, and too little productivity (Chart I-7). More worryingly, strong DM growth will do very little to lift EM economies and assets out of their structural funk. Instead, DM strength is likely to hurt EM. As Chart I-8 shows, since 2009 improvements in DM leading economic indicators (LEIs) have led to falling EM LEIs. Chart I-7EM Structural Headwinds Chart I-8DM Hurting EM EM nations are not very dependent on DM as a source of growth. Intra EM trade has been responsible for most of the growth in EM exports as shipments to the DM economies and the U.S. now account for only 28% and 15% of EM total exports, respectively. While this explains why DM growth cannot lift EM growth, it still does not explain why DM growth leads to deteriorating EM activity. The glue binding this paradox is global liquidity. In a nutshell, when DM growth improves, DM economies suck in global liquidity, which results in a tightening of EM monetary and financial conditions. This combined constriction acts as a large brake on EM growth. Underpinning the relationship between liquidity and growth are a few relationships: First, DM real rates are a relatively clean measure of growth expectations. As Chart I-9 shows, U.S. real yields and the growth expectations embedded in U.S. stocks prices correlate closely with each other. Second, when DM real yields rise, EM reserve accumulation - a measure of high-powered liquidity - moves into reverse (Chart I-10). This suggests that rising DM real yields prompt investors to abandon EM markets, attracted by improving risk-adjusted returns in DM. Chart I-9Real Interest Rates: ##br##A Read On Expected Growth Chart I-10The Liquidity ##br##Channel Third, rising DM rates puts downward pressure on EM FX (Chart 10, bottom panel). Being associated with a reversal of carry trades this is another indication that capital is leaving EM economies. Additionally, falling EM exchange rates tighten EM financial conditions by hampering the financial viability of EM borrowers with foreign currency debt. Fourth, given that the exogenously-driven fall in liquidity already hurts EM growth, rising EM borrowing costs in response to increasing DM real rates amplify the economic drag. By causing the return on EM bonds to fall (Chart I-11), this generates further outflows from EM, and also tightens EM financial conditions. Finally, rising DM yields have been associated with underperforming EM equities relative to DM equities (Chart I-12), giving investors another reason to pull money out of EM. These dynamics have implications for commodity currencies. BCA's view is that DM real yields have upside from here, and therefore EM liquidity and financial conditions are set to tighten. Not only will this hurt EM assets, but a flattening BRICs yield curve should also lead to falling commodity currencies (Chart I-13). Chart I-11The Financial ##br##Channel Chart I-12EM/DM Stocks: A Function ##br##Of DM Real Rates Chart I-13Tightening EM Liquidity Conditions##br## Hurt Commodity Currencies However, differentiation is needed. Tightening EM liquidity and financial conditions are likely to hurt the metal market where there is no broad-based supply deficit. However, like in the late 1990s, oil could actually do well under a strong dollar scenario. For one, the OECD and the U.S. represent much larger shares of oil demand than they do for industrial metals (Chart I-14). In the context of robust U.S. economic growth and consumer spending, we could see continued upward momentum in global oil demand. This is crucial as the oil market is already in a deficit following the collapse in oil capex in 2015 and 2016 (Chart I-15). Additionally, our Commodity and Energy Strategy team argues that OPEC and Russia are very likely to cut production next week. Economic strains and the desire for asset sales in Saudi Arabia and Russia are creating the needed incentives.2 In this environment, oil currencies (CAD and NOK) should outperform antipodeans (AUD and NZD). The outlook for the AUD is the poorest. It is the currency most exposed to metals, the segment of the commodity market most aligned with EM growth. NZD could be at risk too. While it is not exposed to metals like the AUD, the kiwi is very exposed to EM spreads, a variable that is likely to suffer if DM yields continue to rise.3 Buying a basket of CAD and NOK relative to AUD and NZD makes sense here. In terms of our trades, we shorted AUD/CAD too early. However, the economic backdrop described above suggests that the economic rationale for this trade is growing ever more potent. In fact, from late December 1998 to January 2000, CAD rallied against the USD, while the AUD was flat. Additionally, technicals and positioning point to a favorable entry point at the current juncture (Chart I-16). Chart I-14Oil Is Still About The U.S. Chart I-15Favorable Supply/Demand Backdrop For Oil Chart I-16A Good Entry Point For Shorting AUD/CAD Bottom Line: In 2017, the relationship between commodity prices and the dollar is likely to resemble the 1999-2001 outcome. While tightening EM liquidity conditions could weigh on metals, supply concerns and a strong U.S. economy could lift oil prices. This environment would favor the CAD and the NOK relative to the AUD and the NZD. A Countertrend Bounce In The Yen? As we discussed last week, the move in USD/JPY makes sense based on the BoJ policy dynamics we analyzed in our September 23 report titled "How Do You Say "Whatever It Takes" In Japanese?". However, despite our bearish disposition toward the yen, we worry that a countertrend correction in USD/JPY is in the offing. USD/JPY is approaching a formidable resistance. The tell-tale sign of a USD/JPY bull market has been when the pair moves above its 100-week moving average (Chart I-17). We do expect such a move to ultimately materialize. However, with the 100-week MA currently at 114.8, this key indicator is a stone throw away from the present exchange rate of 113.39 and might prove to be a temporary resistance. Additionally, a congestion zone exists between 113 and 114.5, reinforcing this risk. Increasing the danger at the 114 level is the recent high degree of groupthink behavior displayed by this pair. As was the case for the U.S. bonds, the fractal dimension measure for USD/JPY is now below 1.25, highlighting the risk of a countertrend move (Chart I-18). Chart I-17USD/JPY: Key Resistance In Sight Chart I-18A Countertrend Move In USD/JPY Moreover, we agree with our U.S. Bond Strategy service and expect a pause in the U.S. bond sell-off.4 With the tight relationship between USD/JPY and 10-year Treasury yields fully alive, any rebound in bond prices would imply a rebound in the yen. Finally, our intermediate-term timing indicator shows that USD/JPY is 5% overvalued on a tactical time frame, a level where the likelihood of a temporary reversal is heightened. Based on the above observations, today we are opening a tactical short USD/JPY position at 113.39, with a target of 107 and a stop at 115.2. We are also closing our long NOK/JPY trade at a profit of 5.3%. Bottom Line: While the cyclical outlook for USD/JPY continues to point upward, tactically, USD/JPY is facing some downside risk. We are implementing a tactical short USD/JPY trade with a target at 107 and closing our long NOK/JPY trade. Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com 1 Please see Foreign Exchange Strategy Weekly Report, "Dollar: The Great Redistributor", dated October 7, 2016, and Foreign Exchange Strategy Weekly Report, "Reaganomics 2.0?", dated November 11, 2016, available at fes.bcaresearch.com 2 Please see Commodity & Energy Strategy Weekly Report, "The OPEC Debate", dated November 24, 2016, available atces.bcaresearch.com 3 Please see Foreign Exchange Strategy Special Report, "Global Perspective On Currencies: A PCA Approach For The FX Market", dated September 16, 2016, available at fes.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Toward A Cyclical Sweet Spot?", dated November 22, 2016, available at usbs.bcaresearch.com Currencies U.S. Dollar Chart II-1USD Technicals 1 Chart II-2USD Technicals 2 The dollar has crossed a crucial resistance level, and the DXY is now trading close to 102. Positive data this month have contributed to this rally. Durable goods orders came in at 4.8% for October, up from 0.4% in September. This has lifted manufacturing PMI for November to 53.9, showing strength in the supply side of the U.S. economy. Minutes from the November 1-2 FOMC meeting indicate a clear hawkish consensus for December's meeting. A probability of a hike is now fully priced in and is reflected in the almost 14-year high reached by the DXY following the release of the minutes. We should see some stability in the DXY coming up to the December meeting. Otherwise, the U.S. economy seems strong. Upcoming data should ultimately buoy the strength in the dollar, but short-term movements will be limited. Report Links: One Trade To Rule Them All - November 18, 2016 Reaganomics 2.0? - November 11, 2016 When You Come To A Fork In The Road, Take It - November 4, 2016 The Euro Chart II-3EUR Technicals 1 Chart II-4EUR Technicals 2 Draghi remains resolute in his commitment to reach the inflation target. Easy monetary policy has helped support recent growth in the euro area. Low policy rates have increased credit supply, leading to higher lending volumes to households, NFCs and SMEs. Key indicators, such as this month's composite PMI which went up to 53.7, from 53.3, highlight continued decent growth in Europe. Nevertheless, core inflation remains weak at 0.75%, which entails a high likelihood for easy policy going forward. Persistently low rates and structural weaknesses will continue to weigh on bank profitability. Banks may eventually respond by limiting credit growth in the future and hampering overall activity. The short-run outlook for the Euro still remains solid against crosses. EUR/USD has hit a support level, but momentum indicates strong downward pressure against the dollar, so attention to this resistance level is warranted. Report Links: When You Come To A Fork In The Road, Take It - November 4, 2016 Relative Pressures And Monetary Divergences - October 21, 2016 The Pound Falls To The Conquering Dollar - October 14, 2016 The Yen Chart II-5JPY Technicals 1 Chart II-6JPY Technicals 2 USD/JPY has appreciated by more than 7% since the day Donald Trump was elected president. From 1990 up until the day Trump got elected, the yen depreciated at such a high rate in such a short time frame in only 4 occasions. We are taking a tactical short position in USD/JPY, because although we continue to be yen bears on a cyclical basis, the current sell-off seems overdone. USD/JPY has reached highly overbought technical levels and it is near its 100-week moving average of 114.8, which should act as a temporary resistance. More importantly, the sell-off in U.S. bond yields, a major driver of the recent plunge in the yen is likely to pause for the time being. USD/JPY will once again become an attractive buy at around 107. Report Links: One Trade To Rule Them All - November 18, 2016 When You Come To A Fork In The Road, Take It - November 4, 2016 USD, JPY, AUD: Where Do We Stand - October 28, 2016 British Pound Chart II-7GBP Technicals 1 Chart II-8GBP Technicals 2 On Wednesday the Treasury released its Autumn Statement, outlining fiscal policy for the coming year. Philip Hammond, Chancellor of the Exchequer, offered no surprises as he vouched to continue to rebalance the budget, albeit at a slower pace. The fiscal impulse looks to increase slightly, yet stay negative for the next 4 years. Such a hawkish fiscal stance should be a drag on growth in an economy that cannot afford any setbacks as it prepares to exit the European Union. However, despite this grim outlook we are still monitoring the pound as an attractive buy, given that it is very cheap. In fact GBP/USD had very little movement after the announcement, which suggests that much of the risks for the U.K's economic outlook are already priced into the cable. Report Links: The Pound Falls To The Conquering Dollar - October 14, 2016 The Dollar: The Great Redistributor - October 7, 2016 Long-Term FX Valuation Models: Updates And New Coverages - September 30, 2016 Australian Dollar Chart II-9AUD Technicals 1 Chart II-10AUD Technicals 2 The Australian economy continues to encounter structural weaknesses from a deteriorating mining sector, for which the outlook remains pessimistic. An interesting observation is that the mining investment-cut is considerably mature, as RBA Assistant Governor Christopher Kent states "about 80% of the adjustment" is done. However, weak Asian EM fundamentals and a questionable outlook for China imply impending demand-side problems, which will weigh, not only on Australian terms of trade, but also the Australian economy, as emerging Asia represents 66% of Australia's total exports. An additional hurdle for the terms of trade is a rising USD, which could drag down commodity prices and the AUD. In the short run, the MACD line for AUD/USD also points to downside in the near future, as the currency approaches a possible resistance level at 0.72. Report Links: One Trade To Rule Them All - November 18, 2016 When You Come To A Fork In The Road, Take It - November 4, 2016 USD, JPY, AUD: Where Do We Stand - October 28, 2016 New Zealand Dollar Chart II-11NZD Technicals 1 Chart II-12NZD Technicals 2 We continue to hold a bearish stance towards NZD/USD, as the dollar bull market and weakness in Asian currencies will ultimately weigh on the kiwi. However, the outlook for the NZD against other commodity producers is not as clear. Prices for dairy products, which constitute over 30% of New Zealand exports, have skyrocketed and are now growing at 46% YoY. This trend is set to continue in the short term, as Chinese dairy imports continue to rebound, recording a 9.7% growth rate compared to last year. Furthermore, real GDP is growing at a 3.5% pace, the highest in the G10. That being said, we are reticent to be too bullish on this currency, as inflation remains very low and increasing migration is putting a lid on wages. However if inflation picks up, the NZD could become attractive relative to its commodity peers. Report Links: Long-Term FX Valuation Models: Updates And New Coverages - September 30, 2016 Global Perspective On Currencies: A PCA Approach For The FX Market - September 16, 2016 The Fed is Trapped Under Ice - September 9, 2016 Canadian Dollar Chart II-13CAD Technicals 1 Chart II-14CAD Technicals 2 Recent data has come out below expectations: Core CPI came in at 1.7%. Wholesale sales are contracting at -1.2%. Retail sales excluding autos are at 0%. These figures support the view that there is an underlying weakness in the Canadian economy which will keep the BoC from reaching its inflation target. However, as the U.S. continues to be the largest consumer of oil in the world, with around 20% of global consumption, stronger U.S. growth will support oil demand, which in conjunction with tighter supply, will support oil prices. This will support the CAD against other commodity producing currencies. Structural weaknesses and an upward trend in USD/CAD since May suggest that the CAD could experience more downside momentum against USD. Nevertheless, it is important to monitor next week's OPEC meeting, the outcome of which will dictate the CAD. Report Links: When You Come To A Fork In The Road, Take It - November 4, 2016 Relative Pressures And Monetary Divergences - October 21, 2016 The Pound Falls To The Conquering Dollar - October 14, 2016 Swiss Franc Chart II-15CHF Technicals 1 Chart II-16CHF Technicals 2 The decline in EUR/CHF appears to have subsided for the time being. Last week we mentioned that the SNB would not tolerate much more downside on this cross, and would not be shy to intervene if necessary. This view has shown to be valid, as EUR/CHF has found support around 1.07. This floor imposed by the SNB means that the performance of the franc against the dollar should mirror EUR/USD for the time being. This implies that USD/CHF should have limited upside in the short term, as EUR/USD has hit a major support level around 1.05 that has been in place for the last 2 years. On a cyclical basis, monetary divergences should continue to weigh against the euro, which makes us bullish on USD/CHF on this time frame. Report Links: Long-Term FX Valuation Models: Updates And New Coverages - September 30, 2016 Global Perspective On Currencies: A PCA Approach For The FX Market - September 16, 2016 Clashing Forces - July 29, 2016 Norwegian Krone Chart II-17NOK Technicals 1 Chart II-18NOK Technicals 2 The U.S. continues to be world's largest consumer of crude oil, with 20% of total consumption, while China leads in both the copper and nickel markets, accounting for nearly half of global consumption and consuming over 5 times as much as the U.S. in both markets. This divergence implies that if U.S. outperforms the rest of the world, and if the rising dollar continues to weigh on EM economies, oil should outperform base metals in the commodity space and consequently petro currencies like the NOK should outperform other commodity currencies. Additionally the NOK is supported by a current account surplus of 6%, and high inflation is prompting Norges Bank to back off from its dovish stance. While we like the NOK on its crosses, we are more bearish on the NOK versus the USD, as USD/NOK remains very sensitive to the dollar. Report Links: The Pound Falls To The Conquering Dollar - October 14, 2016 The Dollar: The Great Redistributor - October 7, 2016 Long-Term FX Valuation Models: Updates And New Coverages - September 30, 2016 Swedish Krona Chart II-19SEK Technicals 1 Chart II-20SEK Technicals 2 The Swedish economy continues to show signs of strength. Recent data supports this view: Consumer confidence for November is at 105.8, compared to 104.8 for October. Producer Price Index came in at 2.2% annually for October. A strong consumer sector has lifted inflation expectations in Sweden. Strong PPI numbers validate this, as they foretell a potential rise in CPI as producers pass on their costs to consumers. Despite this strength, SEK may see limited upside. As mentioned last week, most of the movement in the SEK can be attributed to the USD. Rate hike expectations have now been fully priced in for the Fed, so it is likely that movements in the USD will be muted, and hence the SEK could find some support, at least for now. Report Links: One Trade To Rule Them All - November 18, 2016 The Pound Falls To The Conquering Dollar - October 14, 2016 Long-Term FX Valuation Models: Updates And New Coverages - September 30, 2016 Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Closed Trades

The Tactical Asset Allocation model can provide investment recommendations which diverge from those outlined in our regular weekly publications. The model has a much shorter investment horizon - namely, one month - and thus attempts to capture very tactical opportunities. Meanwhile, our regular recommendations have a longer expected life, anywhere from 3-months to a year (or longer). This difference explains why the recommendations between the two publications can deviate from each other from time to time. Highlights In November, the model underperformed global equities and the S&P in USD and in local-currency terms. For December, the model reduced its allocation to cash and stocks and boosted its weighting in bonds (Chart 1). Within the equity portfolio, most of the decrease in allocation came at the expense of EM, Sweden, Netherlands, U.S., and New Zealand. The model increased its weighting in Swedish, French, U.K., and Canadian bonds. The risk index for stocks deteriorated in November, while the bond risk index improved significantly. Chart 1Model Weights Feature Performance In November, the recommended balanced portfolio lost 1.5% in local-currency terms and was down 3.4% in U.S. dollar terms (Chart 2). This compares with a gain of 1.3% for the global equity benchmark, and a 3.7% gain for the S&P 500 index. Given that the underlying model is structured in local-currency terms, we generally recommend that investors hedge their positions, though we do provide recommendations from time to time. The sharp bond selloff and weakness in EM equity markets both took a toll on the model's performance in November. Weights The model cut its allocation to stocks from 66% to 53%, and increased its bond weighting from 26% to 47%. The allocation to cash was brought down to zero from 8%, while commodities remain excluded from the portfolio (Table 1). The model trimmed its allocation to Latin American equities by 4 points, Sweden by 3 points, and the Netherlands by 3 points. Also, weightings were reduced in U.S., New Zealand, Spanish, and Emerging Asian stocks. In the fixed-income space, the allocation to Swedish paper was boosted by 12 points, France by 7 points, Canada by 5 points, the U.K. by 3 points, and Italy by 1 point. Allocation to New Zealand bonds was decreased by 6 points and U.S. Treasurys by 1 point. Chart 2Portfolio Total Returns Table 1Model Weights (As Of November 24, 2016) Currency Allocation Local currency-based indicators drive the construction of our model. As such, the performance of the model's portfolio should be compared with the local-currency global equity benchmark. The decision to hedge currency exposure should be made at the client's discretion, though from time to time, we do provide our recommendations. The dollar appreciated significantly in November following the U.S. presidential election. Our Dollar Capitulation Index spiked and is currently at levels that suggest the rally in the broad trade-weighted dollar could pause (Chart 3). Chart 3U.S. Trade-Weighted Dollar* And Capitulation Capital Market Indicators The momentum indicator for commodities has moved further into overbought territory, pushing up the overall risk index. This asset class remains excluded from the portfolio (Chart 4). The deterioration in the liquidity and momentum indicators has lifted the risk index for global equities to the highest level in over 2 years. Our model cut its weighting in equities for the fourth month in a row (Chart 5). Chart 4Commodity Index And Risk Chart 5Global Stock Market And Risk The risk index for U.S. stocks increased sharply in November. With stocks reaching new highs, the model trimmed its allocation to this bourse. The markets took note of the growth-positive aspects of Trump's policies, but seem complacent about the stronger dollar, higher interest rates, and the potential for trade protectionist policies (Chart 6). The risk index for euro area equities has ticked up slightly in November. However, unlike its U.S. peers, it remains in the low-risk zone. Above-trend growth could provide support for euro area equities. (Chart 7). Chart 6U.S. Stock Market And Risk Chart 7Euro Area Stock Market And Risk The risk index for Dutch equities ticked up slightly and the model has downgraded this asset. That said, the weighting in Dutch equities remains the highest among its euro area counterparts (Chart 8). Improvements in the value and momentum measures for Latin American stocks have been largely offset by a deteriorating liquidity reading. As a result, the risk index did not decline much after the selloff. The model decreased its allocation to this asset (Chart 9). Chart 8Dutch Stock Market And Risk Chart 9Latin American Stock Market And Risk Over the course of only a few months, the risk index for bonds has swung from an extremely high risk level to the low-risk zone. Momentum has been the primary driving force behind this move and currently suggests that yields could pull back in the near term (Chart 10). The risk index for U.S. Treasurys declined significantly in November. While the model used the latest selloff to boost its allocation to bonds, it preferred to add allocation to bond markets outside of Treasurys. (Chart 11). Chart 10Global Bond Yields And Risk Chart 11U.S. Bond Yields And Risk After the rise in yields, Canadian bonds are massively oversold based on our momentum measure. The extremely low-risk reading has prompted the model to allocate to this asset (Chart 12). German bonds are oversold, but the reading on the cyclical measure has become considerably more bund-unfriendly. The model opted not to include bunds in the overall boost to its bond allocation. (Chart 13). Chart 12Canadian Bond Yields And Risk Chart 13German Bond Yields And Risk The risk reading in French bonds is more favorable than for bunds. Apart from oversold momentum, the value reading has also improved. The model increased its allocation to French bonds (Chart 14). The cyclical component of the risk index for Swedish bonds keeps moving in a bond-bearish direction. But that is completely overshadowed by extremely oversold conditions. In fact, the overall risk index for Swedish bonds is the lowest within our bond universe. Much of the increase in overall bond allocation ended up in Swedish paper (Chart 15). Chart 14French Bond Yields And Risk Chart 15Swedish Bond Yields And Risk Following sharp gains, the 13-week momentum measure for the U.S. dollar has reached levels at which some consolidation may take place. But the recovery in the 40-week rate of change measure indicates that it would probably be a pause in the dollar bull market rather than a trend change. With the December rate hike baked in, the Fed's communication about the policy next year holds the key to the path of the dollar - in addition to the fiscal policy of the next administration (Chart 16). The Japanese yen has been a major victim of the dollar rally. The 13-week momentum measure is approaching levels that halted the yen weakening trend in 2013 and 2015. However, this time around, it is not coupled with the same signal from the 40-week rate of change measure. The BoJ is sticking to its easy monetary policy, and some additional support on the fiscal front could drag the yen lower, notwithstanding a possible hiatus in the short term. Short term the yen could benefit from an EM pullback (Chart 17). After the latest bout of depreciation, the euro seems poised for another attempt to break below 1.05. The 13-week and 40-week momentum measures do not preclude this from happening. However, it would probably take the ECB to reaffirm its dovish message to push EUR/USD technical indicators into more oversold territory (Chart 18). Chart 16U.S. Trade-Weighted Dollar* Chart 17Yen Chart 18Euro Miroslav Aradski, Senior Analyst miroslava@bcaresearch.com

Feature Happy Thanksgiving to all our U.S. clients. We wish you the best the holiday has to offer, as you share blessings with friends and family. In this holiday-shortened week, we are publishing a joint report with our colleagues at BCA's Energy Sector Strategy (NRG) service. We succinctly examine the pros and cons of the debate over whether OPEC will or will not agree to and uphold a *real* production cut, as it has promised, at its much-anticipated meeting on November 30. Disagreement on the likely outcome of the meeting runs high. In late September, OPEC announced an agreement in principle to cut oil production at the formal November meeting to a level of 32.5-33.0 MMb/d. This would represent a 500,000-750,000 b/d reduction from August production levels, and an 830,000-1,330,000 b/d reduction from the IEA's latest OPEC production estimate for October of 33.83 MMb/d. In addition, non-OPEC behemoth Russia has signaled a potential willingness to contribute its own production freeze or cut to the agreement in an effort to support higher oil prices. Chart 1With A 1 MMb/d Cut, ##br##Draws Would Be Greater There are compelling arguments to be made both supporting the likelihood of a production cut as well as for being skeptical that such an agreement will be reached and adhered to. Even within BCA, there is disagreement. This service, the Commodity & Energy Strategy (CES), which sets the BCA house view on oil prices, pegs the odds at greater than 50% that there will be a meaningful cut of 1 MMb/d+, anchored by large cut pledges from OPEC's leader, the Kingdom of Saudi Arabia (KSA), and Russia. The NRG team, dissents; they think it is more likely that no deal is reached, and if a deal is announced, it will not be adhered to. Regardless of whether there is an announced agreement to cut production or not, both CES and NRG expect KSA's production to decline by 400,000-500,000 b/d between August and December according to KSA's normal seasonal management of production levels; we would not include this expected seasonal reduction in the calculation of a *real* cut. In our analysis on Chart 1, we include a *real* cut of 1MMB/d below the normal seasonality of KSA's production, which lasts for six months. In H2 2017, we assume the cut is dissolved and the market also receives an extra 200,000 b/d of price-incentivized production from the U.S. shales. How To Bet On A Cut, The Out-Of-Consensus Call Chart 2Without A Cut,##br## Inventories Still Will Be Drawn In 2017 CES's view for a cut (established November 3) was significantly out-of-consensus until recent chatter from OPEC increased the perception that an agreement could be reached. Still, there remains significant doubt a freeze or cut can be accomplished. Without a cut, NRG and CES share a constructive outlook for oil markets heading towards steepening deficits during 2017 (Chart 2). Note: BCA's estimates show a tighter oil market than the EIA's estimates: Our Q3 2016 production estimates are lower than the EIA's by ~300,000 b/d due to differences in our assessments in Brazilian, Russian and Chinese production; our Q3 2016 consumption estimate is higher than the EIA due to our higher assessment of U.S. summer-time demand (the EIA has consistently underestimated U.S. demand over the past few years). A production cut coupled with a natural tightening in the market brought about by the price-induced supply destruction over the past 18 months would make 2017 inventory draws even greater, lifting oil prices higher, and providing even greater upward support to our favorite investment recommendations (Chart 1). Below we outline the investment recommendations that would benefit from an OPEC cut, spanning individual equities, ETFs, and commodity calls: Direct Commodity Investment: CES recommends two pair trades on oil contracts and call options. Long February 2017 $50/bbl Brent Calls vs. short February 2017 $55/bbl Brent Calls to play the spike in oil prices that would come from a successful OPEC cut, which was recommended November 3 and was up 50.41% as of Tuesday's close. Long August 2017 WTI contract vs. short November 2017 WTI contract to play an expected flattening of the forward curve, which also was recommended November 3 and it up 48.61% as of Tuesday's close. Oil Producers: NRG recommends overweight-rated Permian oil producers EOG, PXD, FANG and PE, which will be leaders in expanding production into an improving oil price market. Service Companies: NRG recommends overweight-rated completion-oriented services companies HAL, SLB and SLCA, which will benefit most from increased U.S. shale spending. Equity-Backed ETFs: NRG recommends overweight-rated ETFs XLE, FRAK, and OIH as vehicles that provide more diversified investment exposure to higher oil prices and oilfield service activity than individual equities. Oil-Backed ETF. Tactically buying the U.S. Oil Fund ETF (USO) would provide good direct exposure to a quick oil price surge. However, USO should not be held as a longer-term investment because the inherent cost of continually rolling contracts consistently erodes USO's value versus the equity-backed ETFs XLE and OIH. This longer-term underperformance informs NRG's underweight rating on USO. Risks To Our Views: Oil and natural gas prices that differ materially from our forecasts, possibly due to slower-than-expected global economic growth and/or greater than expected supply growth. Poor operational execution and/or changes to regulatory restrictions could negatively impact the financial and stock performance of our recommendations. A week ahead of the OPEC meeting, in the wake of recently recovering production in Libya and Nigeria, and amid campaigning by Iran and Iraq to be excluded from participation in the cuts, it is impossible to know for certain how the complicated politics of OPEC and Russia will play out. Below we outline the competing objectives and risks that will be in play. Case Against A Cut Undeniably, a cut in production, particularly a coordinated cut where several countries share the burden of restricting production, would raise oil prices and enhance 2017 oil export revenues for all OPEC producers. However, that near-term benefit for pricing and revenue has been obvious for the past two years, and yet neither KSA nor Russia has been willing to cut production, feeling the potential to lose longer-term market share outweighed the immediate revenue benefits of a cut. The hazard of a price-increasing production cut, is that the higher oil price would essentially subsidize non-OPEC competitors with higher cash flows, and would simultaneously bolster the confidence of capital markets that OPEC will support prices at a floor of $50, reducing the risk of future investments. These two effects would jointly encourage increased capital investment into establishing new production, especially by the fast-acting U.S. shale producers, whose rampant investment and production growth from 2010-2015 was, by far, the leading contributor to the 2015-2016 oversupply of oil. Encouraging a resurgence of drilling and production would certainly lead to faster production growth from the U.S. shales in 2017-2018, allowing those producers to grow market share under the umbrella of OPEC's production sacrifices that created the higher prices. OPEC has just endured a lot of economic pain through the oil price decline. The economic purpose of this pain was to starve global producers of operational cash flow and dissuade the inflow of new capital, thus choking off the reinvestment required to continue to grow oil production. By and large, this goal has been achieved, with U.S. shale producers slashing capital expenditures by 65% from 2014 to 2016, and the International Oil Companies (IOCs) cutting capital expenditures by 40% over the same period. As a result, after the substantial surge in global oil production in 2014-2015 that created the current over-supply, the capital starvation caused by low oil prices will result in essentially no global production growth in either 2017 or 2018, allowing for demand growth to erode the oversupply of production during 2016, and to eat into the overstocked inventories of crude during 2017-2018. KSA has created fear and uncertainty throughout global producers and capital markets by steadfastly refusing to use its production-management powers to support a floor under oil prices. We are skeptical that KSA will ultimately agree to reverse this strategy, by now establishing a price floor. Such a reversal would undermine the profound market-share message KSA has delivered to competitors (at the cost of great financial pain), and weaken its perceived resolve to allow oil prices to be set by the market. As such, the NRG team believes KSA will not agree to cut production beyond the already-expected seasonal reduction in production, and that this position will scuttle September's tacit agreement to cut production at the official meeting next week. Such a scenario would be fairly similar to how KSA undermined the production-freeze discussions in Doha in April, by insisting other OPEC members - Iran, in particular - share in the production limitations in order to engender KSA's support; a condition that other members were unwilling to accept. The Case For A Cut The case to expect a cut agreement acknowledges that such a cut would subsidize competitors and diminish the impression of KSA's resolve and/or ability to out-last competitors through an oil price down-cycle. The case for a cut concludes that the benefits of higher 2017 oil prices simply outweigh these market share and reputational costs. The benefits that OPEC and Russia would receive are: Critical Need For Higher Revenue. If KSA and Russia each cut 2017 production by 500,000 below current expectations, and oil prices jumped $10/bbl as a result, KSA's 2017 oil export revenues would increase by close to $17.5 billion, and Russia's would increase by almost $8.25 billion. If the financial pain endured by these countries is substantially greater than NRG has estimated, this near-term revenue lift could be more critical than we appreciate, overwhelming the reputational and longer-term market-share losses resulting from the reversal of policy. Borrowing capacity for each country also would increase, as a result of higher revenues. With both states seeking to tap international debt and equity markets, this increased revenue would increase their borrowing capacity. Higher Value For Asset Sales. KSA is preparing to IPO Saudi Aramco. Bolstering the spirits of capital markets with higher oil prices would be expected to increase the proceeds received from this equity sale, increase the market value of the company, reduce debt-service costs, and improve access to debt markets, which KSA and Saudi Aramco are both likely to tap more frequently in the future as the country tries to diversify the economy away from oil. Similarly, two weeks ago, Russia signed a decree to sell a 19.5% stake in Rosneft by the end of 2016. An immediate oil price strengthening and messaging that KSA and Russia would support a pricing floor would inflate the value of this sale, given the high correlation between Brent crude oil prices and Rosneft's equity price. Production Stability Not As Strong As It Seems. Russia's production levels in 2016 have been surprisingly strong, exceeding our expectations. The collapse of the Russian Ruble has allowed for continued internal investment despite the substantial reduction to dollar-denominated oil revenues. Still, it is likely that Russian producers are pulling very hard on their fields, over-producing the optimal level in an effort to scratch out higher revenues. Such over-production is not sustainable ad infinitum, and Russia may know that its fields need a rest in 2017 anyhow, so a 4-5% production cut is ultimately not much of a sacrifice. Make Room For Libya & Nigeria. Both Libya and Nigeria are trying to overcome substantial civil obstacles to allow production to increase back towards oilfield capabilities. If these problems were solved, we estimate Libya could increase production by 400,000-600,000 b/d while Nigeria could add 200,000-300,000 b/d. If KSA, OPEC, and Russia believe these countries will be able to re-establish shut-in production, they may conclude a production cut is necessary to make room for the growth, and to keep prices from collapsing. Entrenching U.S. Shale As The Marginal Barrel: If KSA and Russia can agree to a 1 MMb/d cut, U.S. shale-oil producers would be the first to take advantage of expected higher prices, given the fast-response nature of this production. This actually would work to the advantage of KSA and Russia and other low-cost producers in and outside OPEC, by firmly entrenching U.S. shale oil as the marginal barrel for the world market. On the global cost curve, shale sits in the middle some $30 to $40/bbl above KSA and Russia, which means that, as long as the global market is pricing to shale economics at the margin, these mega-producers earn economic rents on their production. In order to retain those rents, KSA and Russia will have to find a way to keep shale on the margin - i.e., regulate their production so that prices do not rise too quickly and encourage more expensive output to come on line. For KSA and Russia, it is better to climb the shale cost curve than to encourage the next tranche of production - such as Canadian oil sands - to come on to the market too quickly, or to further incentivize electric vehicles and conservation with run-away price increases, with too-sharp a production cut. Allowing prices to trade through a $65 - $75/bbl range or higher would no doubt produce a short-term revenue jump for cash-strapped producers - particularly those OPEC members outside the GCC. But it also would make most of the U.S. shales economic to develop, and incentivize other "lumpy," expensive production that does not turn off quickly once it is developed (e.g., oil sands and deepwater). This ultimately would crash prices over the longer term, making it difficult for the industry to attract capital. This is not an ideal outcome for KSA's planned IPO of Aramco, or Russia's sale of 19.5% of Rosneft, or their investors. Global Reinvestment Needs To Be Re-Stimulated. Stimulating non-OPEC reinvestment with higher oil prices and increased price-floor confidence may actually be needed in the not-too-distant future. IOCs have barely started to show the negative production ramifications of their 40% cuts to capex; cuts which will grow deeper in 2018. We expect these production declines to show up increasingly over the next four years, and there is not much the IOCs can do to stop it, since their mega-project investments generally require 3-5 years from the time that spending decisions are made until first oil is produced. With such huge cuts to future expenditures, and enormous amounts of debt incurred by the IOCs to pay for the completion of legacy mega-projects that will need to be repaid ($130B in debt added in the past two years), OPEC could see a looming shortage of oil developing later this decade if IOC-sponsored offshore production falls into steep declines, as we think is likely. To orchestrate a softer landing, to prevent oil prices from spiking too high due to a shortage of production, to head-off an acceleration in the pursuit of alternative fuels and/or the recessionary impact of an oil price spike, KSA may actually want to accelerate the re-start of global investment. Bottom Line: There are strongly credible and well-reasoned arguments that support the expectations for a successful establishment of a production cut from OPEC and Russia, as well as to doubt that such an agreement will be achieved (and adhered to) amid the political and economic competition between OPEC members and against non-OPEC producers. A successful agreement to cut production in excess of 1 MMb/d, as CES believes is likely, would be the more out-of-consensus call, with substantially bullish implications for oil prices and for our oil-levered investment strategy and stock recommendations. Even without a production cut, the NRG service remains strongly constructive on the investment strengths of high-quality Permian oil producers and the completion-oriented service companies that will benefit from increased U.S. shale spending. If a production cut is achieved, our investment cases become even stronger, as the U.S. shale producers and service companies would be the greatest beneficiaries of an upward step-change in oil prices. Matt Conlan, Vice President Energy Sector Strategy mattconlan@bcaresearchny.com Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com SOFTS Dairy: Moderate Upside In 2017H1 Dairy prices may have another 5%-10% upside over next three to six months, based on tightening supply in the global dairy market. China will become more important in the global dairy market. The country's dairy imports will continue heading north. Downside risks include elevated global dairy product inventory, a supply boost from major exporters, and a continuing strengthening dollar. We have been cautiously bullish on global dairy market since last October.1 Since then, the Global Dairy Trade (GDT) All-Products Price Index, which is widely used as a benchmark price for the market, has rallied over 50% in the past seven months off its November - March lows (Chart 3, panel 1). Chart 3Dairy: Tactically Bullish Now the question is: will the rally continue? A review of what had happened in 2015 and so far this year may be a good start of our analysis. A Terrible 2015 The GDT index tumbled to the lowest level on record in early August 2015. A sharply drop in Chinese dairy imports; the Russian import ban on dairy products; robust supply growth across major dairy producing countries; and the EU's decision to scrap its production quotas created a perfect storm for the global dairy market last year - resulting in an extremely oversupplied market, stock builds and depressed dairy prices (Chart 3, panels 2, 3 and 4). An Improving 2016 Fundamentals have improved since April, as major dairy exporting countries responded to low dairy prices, while Chinese dairy imports revived. Fonterra, the world's biggest dairy exporter, and Murray Goulburn, Australian's biggest dairy company, both announced retrospective price cuts in April to dairy farmers in New Zealand and Australia, which hit both countries' dairy industries hard. Many farmers exited the dairy business, given their production costs were well above farm-gate milk prices. As a result, dairy farmers In New Zealand have cut the national dairy cow herd size by 3.3% yoy in 2015 and then a further 1.5% in 2016, based on USDA data. In Australia, dairy farmers have sent more cows into slaughterhouse as well. According to Dairy Australia, in the past 12 months to August 2016, 109,102 head were sold, an increase of 33% on the previous year. New Zealand and Australia are the world's largest and the fourth largest dairy exporters, respectively. In June, one month before the start of the new season (July 2016 to June 2017), farm-gate milk prices set by major dairy processors in Australia were still much lower than most dairy farmers' production costs, further damaging the country's dairy production outlook for the 2016/17 season. In July, August and September, Australian milk production fell sharply for three consecutive months, with a yoy contraction of 10.3%, 9.3%, and 10.2%, respectively. In July, the European Commission funded a €150 million program to pay farmers to cut their milk production. At the same time, the region also intervened with a stock purchase program and a private-storage aid scheme to help remove excess supply from the market. The EU region is the world's second biggest exporter. Its production increase due to the removal of its quota system was one main reason for last year's price drop. The recent supportive policy has worked well - the region's milk volumes decreased in September for the third consecutive month. In the meantime, Chinese dairy imports have rebounded 9.7% yoy for the first nine months of this year, a significant improvement from last year's 44.4% contraction over the same period. China is the world biggest dairy importer, accounting for 51% of global fluid milk imports, and 40% of dry, whole-milk powder imports (Chart 4, panel 1). Chart 4China Needs More Dairy Imports In comparison, the number of Chinese cow herds only accounts for 6% of global total cows for milk production, which is clearly far from meeting its domestic demand (Chart 4, panel 2). Early this year the country loosened up the "one-child" policy, and now allows "two-kids" in a family, starting this year. This will increase the country's baby formula's demand. The country's dairy product intake per capita is still far below Asian peers like Japan and Korea. Growing family wealth and increasing demand for healthy dairy food will continue boosting the dairy consumption in China. Due to the limited pasture land in the country for raising cows, we expect China's dairy imports will continue heading north. What about the price outlook in the remainder of 2016 and 2017? Most of the positive factors aforementioned are still in place. In the near term, we do not see significant supply increase. Despite the 61% price rally in the GDT price index over the past seven months, most of the price increase still has not passed to farm-gate milk prices in major producing countries (except New Zealand). Hence, for the remainder of 2016 and 2017H1, we expect prices will be prone to the upside. Pullbacks are always possible. But overall we still expect another 5% to 10% upside over next three to six months for the GDT price index. Beyond 2017H1, the price outlook is less clear. If prices either go sideways or up, milk production in major producing countries should eventually recover. For now, we hold a neutral view for dairy prices in 2017H2. Downside Risks Chart 5Downside Risks First, global dairy stockpiles are much higher than previous years (Chart 5, panel 1). According to the European Commission, at the end of September, around 428 thousand metric tons (kt) of skimmed-milk powder (SMP) was in public intervention stocks, while another 73 kt SMP was in private storage. In addition, there also is about 90 kt butter and 19 kt cheese stored privately. As the EU still is aiming to cut milk production to boost dairy prices, we believe the odds of an unexpected release from storage in a fast and massive manner is low. The release will likely be gradual. Second, much of New Zealand's milk production is dependent on weather conditions, which have improved from mid-August. Moreover, Fonterra increased its farm-gate milk price to $6 per kgMS (kilogram milk solid) from $5.25 per kgMS last week, which was the third increase over the past four months. Since August, farm-gate milk price in New Zealand has already been up 41% and well above the country's production cost. A combination of both factors may boost the country's milk production more than the market expected. In this case, prices could decline in 2017H1. Third, if the U.S. dollar continues strengthening versus the RMB and other major exporters' currencies, this will tend to discourage purchases from China and encourage sales from New Zealand, the EU and Australia, which will be negative to dairy prices (Chart 5, panel 2). We will monitor these risks closely. Ellen JingYuan He, Editor/Strategist ellenj@bcaresearch.com 1 please see Commodity & Energy Strategy Weekly Report for softs section "Oil Markets Pricing In $20/Bbl Downside," dated October 1, 2015, available at ces.bcaresearch.com Investment Views And Themes Recommendations Tactical Trades Commodity Prices And Plays Reference Table Closed Trades