Commodities & Energy Sector

Highlights Just to be clear: The balance of price risks in oil markets remains to the upside - particularly if we see a supply shock resulting from the loss of as much as 2mm b/d of exports from Iran and Venezuela. Neither the supply side nor the demand side in base metals evidence outsized risks, which keeps us neutral ... for now. Still, downside risks for commodities - mostly via threats to trade - loom. In line with our House view, we believe markets are too complacent re the effects of a global trade war.1 However, focusing only on the trade war obscures growing risks to EM imports and exports arising from the Fed's rates-normalization policy, which is pushing the USD higher. A strong USD retards EM trade growth, which is particularly bearish for metals and oil (Chart of the Week). Chart of the WeekStronger USD, Slower EM Import Growth##BR##Bearish For Base Metals And Oil Stronger USD, Slower EM Import Growth Bearish For Base Metals And Oil Stronger USD, Slower EM Import Growth Bearish For Base Metals And Oil An oil-supply shock taking prices above $120/bbl, as one of our scenarios does, would generate a short-term inflationary impulse, and would depress aggregate demand, particularly in EM. Ultimately, it would become a deflationary impulse, as higher energy prices consume a larger share of discretionary incomes, and slow growth. A slowdown in EM trade on the back of a strong USD also would generate a deflationary impulse, as EM income growth slows and aggregate demand falls. Either way, the Fed's rates-normalization policy will be put on hold as current inflation risks morph to deflation risks, if the downside becomes dominant. Highlights Energy: Overweight. The U.S. Strategic Petroleum Reserve (SPR) will release 11mm of oil from its reserves in the October - November period, to allay concerns over the likely loss of 1mm b/d of Iranian exports to U.S. sanctions. We've been expecting this ahead of U.S. mid-term elections, but don't think it will fill the gap in lost exports. Base Metals: Neutral. Union and management leaders at BHP's Escondida mine in Chile averted a strike, after agreeing a contract at the end of last week. Precious Metals: Neutral. Gold rallied more than $35/oz off its lows of last week, as markets took notice of record speculative short positioning, which many view as a bullish contrary indicator. Gold was trading to $1195/oz as we went to press. Ags/Softs: Underweight. The USDA is expected to roll out a $12 billion relief package for farmers on Friday, which includes direct purchases of commodities that were not exported due to tariffs, according to agriculture.com's Successful Farming publication. Feature Overall, the balance of price risks in the industrial commodities are neutral (in base metals) and to the upside (in oil). In the base metals, we think fear of a Sino - U.S. trade war has market participants jittery, and may be getting to the point where it is starting to affect expectations for capex and investment on the production side, and growth on the demand side. Given our expectation EM trade will hold up this year (Chart 2), we continue to expect base metals demand to remain fairly stable, and perhaps pick up as China rolls out modest stimulus measures later this year.2 Chart 2USD Strength Slows EM Trade Growth USD Strength Slows EM Trade Growth USD Strength Slows EM Trade Growth We remain bullish oil demand - expecting growth of ~ 1.6mm b/d on average in 2018 - 19, and continue to expect a supply deficit next year, which will push Brent prices from $70/bbl on average in 2H18 to $80/bbl next year.3 However, if we see continued strength in the USD beginning to degrade actual EM demand, we will be forced to revise our assessment. Downside Risks To Metals And Oil Loom As mentioned above, we are aligned with our House view, and believe markets are all but ignoring the risk of an all-out trade war, spreading from the well-covered Sino - U.S. standoff to the broader global economy. The global economy already appears to be registering the first signs of a trade slowdown, according to the World Bank's July 2018 global outlook, where it observes "softening demand for imports in advanced economies - with the exception of the United States - and weaker exports from Asia."4 We also are picking it up in our modeling (Chart 2). The Bank also notes the slowdown in trade "is accompanied by rising barriers to trade, moderating growth in China, higher energy prices, and elevated policy uncertainty." A prolonged trade war that spreads globally would be especially devastating to EM economies, as two-thirds of them are commodity exporters of one sort or another.5 Fed Policy Is An EM Growth Risk As important as a trade war is for global growth, focusing too heavily on it obscures growing risks to EM imports and exports arising from the Fed's rates-normalization policy, which is pushing the USD higher. Table 1USD Vs. Fed Policy Variables Trade, Dollars, Oil & Metals ... Assessing Downside Risk Trade, Dollars, Oil & Metals ... Assessing Downside Risk Per the Richmond Fed's Summary, the Fed is charged by Congress to "promote effectively the goals of maximum employment, stable prices, and moderate long term interest rates."6 One of the models we use to forecast the broad trade-weighted USD is a Fed policy-variables model, which uses lagged U.S. nonfarm payrolls, core PCEPI (the Fed's preferred measure), U.S. 10-year real rates, and U.S. short-term real-rate differentials vs. DM rates as proxies for these policy goals. We throw lagged copper futures prices in to pick up current industrial activity, as well (Table 1). This model highlights the long-term equilibrium between the USD TWIB and the Fed's policy variables going back to 2000.7 We average the output of the policy-variables model with four other models using close-to-real-time variables, and some other proxies for the Fed's policy variables to generate our forecast (Chart 3). Chart 3BCA USD TWIB Forecast BCA USD TWIB Forecast BCA USD TWIB Forecast The USD TWIB and EM trade volumes form a cointegrated system, as shown in Chart 2. Based on our modeling, we expect EM trade to hold up reasonably well over the next year, with y/y growth remaining positive most of the time. But, as close inspection of the chart reveals, the rate of p.a. growth is slowing as a result of the Fed's rates-normalization policy. This means the rate of growth in EM demand for base metals and oil will slow, although the level of demand will remain high following 20 years of solid growth.8 As a House, we expect the USD TWIB to rise another 5% over the next year, which, given the elasticities in our model, would translate into more than 10% declines in copper and Brent prices, all else equal. The Oil Wildcard As regular readers of this service know, we do not believe "all else equal" applies to commodity markets, particularly oil. We have been highlighting the risks of a confluence of negative supply shocks for months - i.e., the loss of up to 2mm b/d of oil exports from Iran and Venezuela - and the implications of this for prices (Chart 4). This is apparent in our ensemble forecasts, which reflect the physical deficit we expect to the end of 2019 (Chart 5). Chart 4U.S. SPR Release Doesn't Cover Lost Iranian Exports Trade, Dollars, Oil & Metals ... Assessing Downside Risk Trade, Dollars, Oil & Metals ... Assessing Downside Risk The U.S. government has taken notice of these risks. However, we believe this week's announcement by the Trump administration to release 11mm barrels of crude oil from the U.S. SPR over the October - November period might hold gasoline prices down ahead of the U.S. midterms, but will do next to nothing to make up for the lost export volumes we are expecting in 2019 (Chart 4). Chart 5BCA Continues To Expect Physical Deficits BCA Continues To Expect Physical Deficits BCA Continues To Expect Physical Deficits An oil-supply shock taking prices above $120/bbl - the projection from one of our scenarios in Chart 4 - would generate a short-term inflationary impulse in U.S. data the Fed follows. This would depress aggregate demand, particularly in EM, as oil is priced in USD. The Fed likely looks through this spike, but, should it misread the inflation impulse and tighten more aggressively, it would be delivering a double-whammy to EM economies: Higher oil prices and a stronger USD. Many EM governments have relaxed or removed subsidies on fuel prices following the 2015 collapse in oil prices engineered by OPEC. While some governments may re-introduce subsidies, not all will cover all of the price increase in such a shock.9 So, even if some subsidies are re-introduced, a price spike likely would hit EM consumers harder than previous high-price epochs. There is a non-trivial likelihood such an oil-price spike would trigger a recession in the U.S. - and likely in DM and EM economies - per Hamilton's (2011) analysis.10 This would force the Fed to change course and resume its accommodative policies. Ultimately, this would become a global deflationary impulse, as higher energy prices erode discretionary incomes, and slow growth. Bottom Line: An oil-supply shock and slower EM trade growth on the back of a strong USD ultimately produce deflationary impulses. Either way, Fed rates-normalization policy will be put on hold if these downside risks become the dominant theme in industrial commodity markets, and the current inflation risks morph to deflation risks. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com Hugo Bélanger, Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com 1 Please see BCA Research's Global Investment Strategy Weekly Report "How To Trade A Trade War," published July 13, 2018. It is available at gis.bcaresearch.com. 2 BCA Research's Geopolitical Strategy is expecting policymakers to deploy modest fiscal stimulus and reflationary policies to counter growing threats from the country's trade war with the U.S. This will be supportive, at the margin, for bulks and base metals. Please see "China: How Stimulating Is The Stimulus?" published by our Geopolitical Strategy August 8, 2018. It is available at gps.bcaresearch.com. 3 Please see BCA Research's Commodity & Energy Strategy Weekly Report "OPEC 2.0 Sailing Close To The Wind," which contains our most recent supply-demand balances and forecasts. It was published August 16, 2018, and is available at ces.bcaresearch.com. 4 Please see The World Bank's Global Monthly, July 2018, p. 2. 5 Please see remarks by World Bank Senior Director for Development Economics, Shantayanan Devarajan, who notes, "two-thirds of developing countries ... depend on commodity exports for revenues." His remarks are in "Global Economy to Expand by 3.1 percent in 2018, Slower Growth Seen Ahead," World Bank press release on June 5, 2018. 6 Please see Steelman, Aaron (2011), "The Federal Reserve's "Dual Mandate": The Evolution Of An Idea," published on the Federal Reserve Bank of Richmond's website. 7 We use a cointegration model to estimate these policy-driven regressions. The output is stout (R2 is greater than 0.95), and it has good out-of-sample results. We use a weighted-average of the five forecasts based on root-mean-square-errors to come up with our USD_TWIB forecast. 8 The World Bank estimates the seven largest EM economies - Brazil, China, India, Indonesia, Mexico, the Russian Federation, and Turkey - accounted for ~ 100% of the increase in metals consumption and close to 70% of the increase in energy demand over the past 20 years. Please see "The Role of Major Emerging Markets In Global Commodity Demand," in the Bank's June 2018 Global Economics Prospects, beginning on p. 61. 9 Please see BCA's Commodity & Energy Strategy Weekly Report, "OPEC 2.0 Scrambles To Reassure Markets," published June 28, 2018. It is available at ces.bcaresearch.com. 10 For an excellent discussion of the correlation between oil-price shocks and recessions, please see Hamilton, James D. (2011), "Historical Oil Shocks," Prepared for the Handbook of Major Events in Economic History. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trade, Dollars, Oil & Metals ... Assessing Downside Risk Trade, Dollars, Oil & Metals ... Assessing Downside Risk Trades Closed in 2018 Summary of Trades Closed in 2017 Trade, Dollars, Oil & Metals ... Assessing Downside Risk Trade, Dollars, Oil & Metals ... Assessing Downside Risk

Highlights Portfolio Strategy Looming inflation, the synchronized global capex upcycle and rising real Treasury yields all argue for preferring oil-related to gold-exposed equities. Recent Changes Initiate a long S&P oil & gas exploration & production / short global gold miners pair trade today. Table 1 Deflation - Reflation - Inflation Deflation - Reflation - Inflation Feature Chart 1No Contagion Yet No Contagion Yet No Contagion Yet Stocks recovered smartly from the Turkey induced pullback last week, and continue to flirt with all-time highs. While the risk of contagion remains acute, three key high-frequency financial market metrics suggest that the SPX will likely escape unscathed. The second panel of Chart 1 shows that both the Japanese yen and the Swiss franc, the two ultimate safe havens, have barely budged vis-a-vis the U.S. dollar and also the junk bond market remains extremely calm (third panel, Chart 1). We will continue to closely monitor these indicators to gauge the risk of contagion in U.S. equities. The greatest risk, however, is China's economic footing, particularly its foreign exchange policy (bottom panel, Chart 1). Any further steep devaluation in the renminbi will prove destabilizing and bring back memories of August 2015 when Chinese policy easing caused the dollar to spike and short-circuited SPX EPS growth. Relatedly, there is also a risk that China moves forward more aggressively on capital account liberalization, likely leading to a renminbi devaluation at least initially. Re-reading this Bank For International Settlements paper (starting on page 35 penned by Mitsuhiro Fukao, an ex-Director of Economic Research at the Bank of Japan) and taking a cue from Japan's experience was insightful.1 But, it remains difficult to predict what China's ultimate reaction function to Trump's trade rhetoric will be (Mathieu Savary, BCA's foreign exchange strategist, will be addressing this in one of his upcoming reports). While a tactical 5-10% pullback cannot be ruled out as the seasonally weak month of September is nearing, from a cyclical perspective our strategy would be to "buy the dip" if one were to materialize. Importantly, this bulletproof equity market that refuses to go down has two stealthy allies on its side: pension plans that are forced into equities and corporate treasurers that execute buybacks. Granted, EPS have delivered and suggest that upbeat fundamentals remain the key market support pillars. As a result, the S&P 500 is on track to register a tenth consecutive positive total return year, which is unprecedented in previous expansions. The only other time that the (reconstructed) SPX rose every year for 10 years in a row was in the late 1940s, however, two recessions occurred during that equity market run (Chart 2). While we are undoubtedly in the later stages of the bull market and the business cycle, there is a big difference between "late-cycle" and "end-of-cycle". Keep in mind that the current backdrop is unusual. A large fiscal package has hit late in the game likely extending the cycle. Thus, gauging where we are in the cycle is important. Chart 3 shows a stylized liquidity cycle and our sense is that we are in the early innings of the inflation stage. The handoff from reflation to inflation has happened and during this stage excesses take root eventually morphing, more often than not, into a mania. Chart 2Impressive Streak Continues Impressive Streak Continues Impressive Streak Continues Chart 3Liquidity Cycle Deflation - Reflation - Inflation Deflation - Reflation - Inflation From a macro perspective inflation is slated to rear its ugly head. Nominal GDP is far exceeding the 10-year Treasury yield, and this yield curve type steepening is bullish for SPX top line growth (Chart 4). As a reminder, in Q2 the GDP deflator jumped to 3.35% pushing nominal GDP growth to 7.41%. Money velocity2 is also enjoying a slingshot recovery. Nominal GDP growth is outpacing M2 money supply growth by roughly 150bps. The U.S. money multiplier (M2 over the monetary base, not shown) is also at a 5-year high. This is an inflationary backdrop (bottom panel, Chart 5) and should also boost SPX revenues and thus continue to underpin the broad equity market. Similarly, the NY Fed's Underlying Inflation Gauge (UIG) is firing on all cylinders and is a harbinger of a further pickup in core inflation in the coming months. As a result, SPX sales growth remains on a solid foundation (Chart 6). Chart 4SPX Sales Rest On Solid Foundations SPX Sales Rest On Solid Foundations SPX Sales Rest On Solid Foundations Chart 5A Little Bit Of Inflation... A Little Bit Of Inflation... A Little Bit Of Inflation... Chart 6...Is A Boon For The SPX ...Is A Boon For The SPX ...Is A Boon For The SPX This week we are initiating a market and asset class neutral pair trade to benefit from the inflationary backdrop. Initiate A Long Oil & Gas E&P / Short Gold Miners Pair Trade One way to benefit from this onset of the inflation stage/mania phase is to go long oil & gas exploration & production/short global gold miners. On the underlying commodity front, the handoff from reflation to inflation has historically been a boon to the oil/gold ratio (OGR). Importantly, the prices paid subcomponent of the ISM manufacturing survey has gone parabolic compared with the new order sub index, roughly doubling since the 2016 nadir. This depicts an inflationary backdrop and is signaling that the OGR will play catch up in the coming months (Chart 7). Chart 7CHART 7 Reflation To Inflation Handoff CHART 7 Reflation To Inflation Handoff CHART 7 Reflation To Inflation Handoff Similarly, another surging inflation indicator also suggests that the OGR has ample room to run. The GDP deflator has recently eclipsed the 3% mark and since exiting deflation following the end of the recent global manufacturing recession it is up over 370bps. Chart 8 shows that if this multi-decade positive correlation were to hold then the OGR could double from current levels. Chart 8GDP Deflator On The Rise GDP Deflator On The Rise GDP Deflator On The Rise Finally, the NY Fed's UIG is also closely correlated with OGR momentum, corroborates the other firming inflation signals and hints that more gains are in store for the OGR (bottom panel, Chart 9). Global macro tailwinds are also clearly in favor of oil at the expense of gold. BCA's global industrial production gauge of 40 DM and EM countries continues to expand at a healthy clip. Oil is a global growth barometer, whereas gold represents one of the few true safe havens in times of duress. Taken together, the implication is that a catch up phase looms for the OGR (middle panel, Chart 9). The relative commodity backdrop is the most important determinant of relative share prices as it dictates the direction of relative profitability (middle panel, Chart 10). Therefore, as the OGR goes so do relative share prices. Chart 9Enticing Global Macro Backdrop Enticing Global Macro Backdrop Enticing Global Macro Backdrop Chart 10Buy Oil & Gas E&P... Buy Oil & Gas E&P... Buy Oil & Gas E&P... Beyond this enticing relative commodity complex outlook, the synchronized global capex upcycle, one of BCA's key themes for the year, is underpinning the relative share price ratio. U.S. capex in particular is outpacing GDP growth and oil & gas investment is the key driver. The V-shaped recovery in the Baker Hughes oil & gas rig count data (bottom panel, Chart 10) confirms this upbeat energy capital outlay backdrop. Moreover, capex intentions from the Dallas Fed survey point to more upside in relative share prices (bottom panel, Chart 11). Meanwhile, keep in mind that the U.S. has been at full employment for 18 months now (in other words the unemployment gap closed in February of 2017) and the economy is firing on all cylinders. Real rates have also shot the lights out recently. In fact the 5-year real Treasury yield is perched near 1%, a multi-year high. Given that gold does not yield any income, it suffers when real yields rise and vice versa (for additional details on the relationship between gold and interest rates, please refer to the early-May piece penned by our sister publication U.S. Bond Strategy titled "A Signal From Gold?").3 Similarly, relative share prices thrive when real yields advance and retreat when the TIPS yield sinks (top panel, Chart 12). Chart 11...At The Expense Of Gold Miners ...At The Expense Of Gold Miners ...At The Expense Of Gold Miners Chart 12Bullion TIPS Over Bullion TIPS Over Bullion TIPS Over Unsurprisingly, the Fed has been tightening monetary policy since December 2015. Nevertheless, the "Fed Spread" (2-year Treasury yield compared with the fed funds rate) is steepening and continues to point to additional gains in the share price ratio (bottom panel, Chart 12). Given that both the ECB and the BoJ have remained ultra-accommodative, a hawkish Fed has boosted the U.S. dollar. However, most commodities are priced in greenbacks, thus the currency effect is a washout and is neither closely correlated to the OGR nor to the share price ratio. Two risks to this high octane, high momentum pair trade are: an EM accident induced risk off phase and a global recession likely due to a flare up in the global trade war (policy uncertainty shown inverted, top panel, Chart 9). In either of these scenarios, investors will likely seek the refuge of bullion's perceived safety as the bond market will almost immediately start pricing in easier monetary policy with investors flocking into the ultimate safe haven asset, U.S. Treasurys. Netting it all out, an enticing macro backdrop with the onset of the inflation stage, the synchronized global capex upcycle and rising real Treasury yields all argue for preferring oil-related to gold-exposed equities. Bottom Line: Initiate a market- and currency-neutral long S&P oil & gas exploration & production/short global gold miners pair trade today. The ETF ticker symbols the S&P oil & gas exploration & production and the global gold mining index are: XOP and GDX, respectively. Anastasios Avgeriou, Vice President U.S. Equity Strategy anastasios@bcaresearch.com 1 BIS Papers No 15 "China's capital account liberalisation: international perspectives", Monetary and Economic Department, April 2003. 2 "The velocity of money is the frequency at which one unit of currency is used to purchase domestically- produced goods and services within a given time period. In other words, it is the number of times one dollar is spent to buy goods and services per unit of time. If the velocity of money is increasing, then more transactions are occurring between individuals in an economy". Source: Federal Reserve Bank of St. Louis. 3 Please see BCA U.S. Bond Strategy Weekly Report, "A Signal From Gold?" dated May 1, 2018, available at usbs.bcaresearch.com. Current Recommendations Current Trades Size And Style Views Favor value over growth Favor large over small caps

Dear client, Our publishing schedule will be shifting over the next two weeks. Next Friday, we will publish a Special Report aggregating various pieces from our colleague Matt Gertken of BCA's Geopolitical Strategy detailing the reforms taking place in China and their past and future evolution, and the economic and investment implications for China and the rest of the world. Matt argues that Chinese reforms are in place and here to stay, which should deepen the malaise in EM and support the dollar. We will not publish any report on August 31st. We will resume our regular publishing schedule on September 7. I hope you enjoy the rest of your summer. Best regards, Mathieu Savary Highlights The 1997 Asian Crisis was a deflationary event, causing commodity prices, commodity currencies and the yen to fall against the dollar, but it had a limited impact on the euro. When Russia collapsed in 1998, the LTCM crisis hit the U.S. banking system, with fears of solvency dragging Treasury yields lower, hurting the dollar against the yen and the euro. Today is not 1997, but the tightness of the U.S. economy suggests the Federal Reserve will need a large shock before abandoning its current pace of a hike per quarter; additionally, global liquidity conditions are tightening and China is slowing. The EM crisis is therefore not over, and vulnerable Brazil, Chile, Mexico, Colombia and South Africa could still experience significant pain. Unlike in 1998, the hot potato is not hiding in the U.S. but in Europe. A contagion event is therefore more likely to hurt the euro than 20 years ago; meanwhile, the yen stands to benefit. DXY could hit 100, and commodity currencies still have ample downside, the AUD in particular. Continue to monitor our China Play Index to gauge if Chinese stimulus could delay the day of reckoning for EM; this index can also be employed as a hedge for investors long the dollar or short EM plays. Feature "Misfortune tests the sincerity of friends." - Aesop This summer is oddly reminiscent of that of 1997. The Federal Reserve is tightening policy because the U.S. economy is not only at full employment but is also growing strongly and generating increasing domestic inflationary pressures. But the most familiar echoes come from outside the U.S. Specifically, emerging market trepidations are once again front page news as the Turkish lira, which had already fallen by 24% between January 2018 and July 31st, dropped by an additional 28% at its worst in a mere two weeks. Consequently, investors are now fretting about the risks of contagion across EM markets, one that could reverberate among G10 economies as well. We too worry that the echoes of 1997 are becoming increasingly louder. EM economies have built up large stocks of debt, and have financed themselves heavily by tapping foreign investors. However, these investors can be rather fickle friends, and we are set to test their sincerity. In this piece, we review how the events of 1997-'98 unfolded, what it meant for G10 currencies, and whether the same lessons can be applied today. We find that in 2018, an EM crisis could ultimately be more supportive for the dollar versus the euro, as unlike in 1998, where the hot potatoes were held by U.S. hedge funds, this time the mess sits squarely in Europe. Tom Yum Goong Goes Viral Initiated in the second half of the 1980s, the peg of the Thai baht seemed like a very successful experiment. The stability created by this institutional setup not only contributed to keeping Thai inflation at manageable levels, but by incentivizing capital inflows in the country it also helped Thailand build up its capital stock. At the time, this yielded a large growth dividend, with real GDP growth averaging 9% from 1985 to 1996. However, the economic boost generated by this cheap financing had a dark side. The Thai current account balance ballooned to a deficit of 8% of GDP in 1995-'96. As Herb Stein famously expressed, if something cannot go on forever, it will stop. Like in Aesop's fable where one of two travelers climbed up a tree to avoid a bear, leaving his friend to fend off the bear on his own, foreign investors abandoned Thailand, which was left on its own to finance its large current account deficit. While the Bank of Thailand was able to fend off the attacks for a few weeks, on July 2nd, 1997, it abandoned its efforts. The THB was left to float freely and dropped 56% against the USD over the subsequent six months. Other EM countries including Malaysia, Brazil and Korea, to name a few, had implemented similar U.S. dollar pegs. They too enjoyed stable inflation, growing money inflows and improved growth, but also experienced growing current account deficits and foreign currency debt loads. It did not take long for investors to extrapolate Thailand's woes to other countries. The Malaysian ringgit and the Indonesian rupiah began falling soon after the THB, while the Korean won began its own steep descent four months later (Chart 1). The economic pain was felt globally. The collapse in EM Asian exchange rates and the deep recessions experienced in these countries caused their export prices to collapse, which created a global deflationary shock (Chart 2). This shock was compounded by a fall in commodity prices that materialized as market participants realized that demand for commodities from the crisis-stricken countries was set to evaporate (Chart 2, bottom panel). Chart 1How The Thai Crisis Morphed Into An Asian Crisis How The Thai Crisis Morphed Into An Asian Crisis How The Thai Crisis Morphed Into An Asian Crisis Chart 2The Asian Crisis Was A Deflationary Shock The Asian Crisis Was A Deflationary Shock The Asian Crisis Was A Deflationary Shock Not only did this deflationary shock lift the USD against EM currencies and commodity currencies, it also caused inflation breakevens in the U.S. to fall significantly (Chart 3). However, because the U.S. economy remained robust through the second half of 1997 and in the early days of 1998, real rates did not respond much (Chart 3, bottom panel). Markets where not very concerned that this shock would force the Fed to cut rates, as it did not seem to affect the outlook for U.S. growth and employment. However, this combination of stable real rates in the face of weaker growth in EM, as well as the collapse in commodity prices ended up having large second-round effects. Russia defaulted in August 1998, prompting a collapse in the ruble. To patch up its finances, Russia began pumping ever more oil out of the ground, causing oil prices to fall below US$10/bbl in December 1998, deepening the malaise in commodity prices. This caused the Brazilian real to collapse in 1999, and the Argentinian peso to follow in 2002 (Chart 4). Chart 31997: Falling Breakevens, Stable Real Yields 1997: Falling Breakevens, Stable Real Yields 1997: Falling Breakevens, Stable Real Yields Chart 4Asian Crisis Goes Global Asian Crisis Goes Global Asian Crisis Goes Global Among these contagions, the Russian default was the event with the greatest systemic impact. This was because it was a direct hit to the U.S. banking system. Long Term Capital Management, a large Connecticut-based hedge fund, had accumulated massive bets on Russia. The country's default plunged the fund into the abyss. However, LTCM had liabilities to banks to the tune of US$125 billion. The exposure was perceived as an existential threat to the banking sector, and the market began to anticipate a repeat of the 1907 panic.1 Junk bond spreads jumped, the S&P 500 fell by 18%, and U.S. government bond yields collapsed by 120 basis points (Chart 5). The Fed was forced to respond, coming out of hibernation and cutting rates by 75 basis points between September and November of 1998. As the Fed forcefully responded to this shock and 10-year Treasury yields fell, the dollar, which had managed to stay somewhat stable against the synthetic euro from July 1997 to August 1998, fell 11%. Within the same one-year window starting in July 1997, the yen dropped 23%, dragged lower by the competitive pressures created by weaker Asian currencies. However, as soon as U.S. bond yields collapsed, the yen began to surge, rising by 36% from August 1998 to January 1999 (Chart 6). Only once the Fed started increasing rates anew did the euro and the yen level off. Chart 5The Russian Default Was The Real Shock For The U.S. The Russian Default Was The Real Shock For The U.S. The Russian Default Was The Real Shock For The U.S. Chart 6The Dollar Buckled After LTCM The Dollar Buckled After LTCM The Dollar Buckled After LTCM In aggregate, the dollar's performance through the 1997-1998 period was very mixed. The trade-weighted dollar managed to rise from July 1997 to August 1998. Nevertheless, this was a complex picture. During this timeframe the dollar rose against EM currencies - against the CAD, the AUD, the NZD and the JPY - but was flat against the euro. The USD then fell against everything from August 1998 to the first half of 1999. Only once the Fed started hiking again in the summer 1999, was the greenback able to resuming its broad ascent, one that lasted all the way until late 2001. Bottom Line: In 1997, the first domino to fall was Thailand. Since many East Asian economies suffered the same ills - current account deficits, foreign currency debt loads and falling foreign exchange reserves - Asian currencies followed, dragging the yen lower in the process. This generated a deflationary shock that hurt commodity prices and commodity currencies, leading to the infamous Russian default of 1998. The associated LTCM bankruptcy threatened the survival of the U.S. banking system, forcing bond yields much lower as the Fed cut rates three times. The dollar suffered because of this policy move, especially against the yen. However, once the Fed resumed its hiking campaign, the dollar recovered across the board, making new highs all the way to late 2001 and early 2002. Is 2018: 1997, 1998, Or 2018? In one key regard, today is not the late 1990s: Dollar pegs are few and far between. However, in many respects, similarities abound. First and most obviously, EM foreign currency debt loads, as measured against exports, GDP or reserves, are at similar levels to those prevailing in the late 1990s (Chart 7). This means that EM economies suffer when the dollar rises, as it represents an increase in their cost of capital, and thus a tightening in financial conditions. Second, the Fed has been increasing interest rates. Most importantly, the Fed is growingly concerned that domestic inflationary pressures in the U.S. are intensifying, courtesy of strong growth - at least relative to potential; a high degree of capacity utilization, especially in the labor market (Chart 8); and, unique to today, the U.S. has received a large degree of unneeded fiscal stimulus. Chart 7EM Dollar Debt Is High EM Have More ##br##Foreign-Currency Debt Than In The 1990s EM Dollar Debt Is High EM Have More Foreign-Currency Debt Than In The 1990s EM Dollar Debt Is High EM Have More Foreign-Currency Debt Than In The 1990s Chart 8The Foreign Pain Threshold For The Fed Is Much Higher ##br##Now Than In 2015 or 2016 The Foreign Pain Threshold For The Fed Is Much Higher Now Than In 2015 or 2016 The Foreign Pain Threshold For The Fed Is Much Higher Now Than In 2015 or 2016 This means it will take a lot of pain to derail the Fed from its desire to hike rates once a quarter. This also makes the current environment very different from 2015, the most recent episode of EM tumult. In 2015-2016, the Fed easily abandoned its hiking campaign. When it hiked rates in December 2015, the Fed anticipated increasing rates four times over the following 12 months. It delivered only one hike in December 2016. The reason was straightforward: Unlike today, the U.S. economy was still replete with slack (Chart 8) and was not on the receiving end of a large fiscal stimulus program, suggesting the Fed could not tolerate the deflationary impact of tightening financial conditions. Third, global liquidity is tightening, which is hurting the global growth outlook. Today, global excess money, as defined by the growth of broad money supply above that of loan growth in the U.S., the euro area and Japan, is contracting. Today, as in 1997, this indicator forebodes important weaknesses in global industrial production (Chart 9). U.S. liquidity is particularly important. Not only is dollar-based liquidity crucial to financing the large stock of dollar-denominated foreign debt, but the U.S. is also driving the fall in global excess money. The pick-up in U.S. economic activity is sucking liquidity from both the rest world and from the financial system to finance U.S. loan growth (Chart 10). This phenomenon was also at play in 1997. Chart 9Excess Money Is Contracting Global Excess ##br##Money Contracting, Just Like In Early 1997 Excess Money Is Contracting Global Excess Money Contracting, Just Like In Early 1997 Excess Money Is Contracting Global Excess Money Contracting, Just Like In Early 1997 Chart 10The U.S. Economy Is ##br##Sucking In Liquidity The U.S. Economy Is Sucking In Liquidity The U.S. Economy Is Sucking In Liquidity Why does this matter? Simply put, U.S. financial liquidity; built as a composite of 3-month T-bills, total bank deposits minus bank loans, bank investments, and M2 money supply; is a wonderful leading indicator. The current collapse in financial liquidity suggests that the global economy is about to hit a rough patch. As Chart 11 illustrates, the weakness of this indicator points to declines in our Global Leading Economic Indicators and in global commodity prices. This suggests the indicator is foretelling that a deflationary scare could materialize, an event normally also associated with a stronger dollar and downside in EM export prices (Chart 12). In a logically consistent fashion, the liquidity indicator is also warning that the AUD, CAD and NZD have substantial downside, while EM equity prices could also suffer more (Chart 13). Finally, it also highlights that even the U.S. stock market may not be immune to upcoming troubles (Chart 14). Chart 11U.S. Financial Liquidity Points To Weaker Growth... U.S. Financial Liquidity Points To Weaker Growth... U.S. Financial Liquidity Points To Weaker Growth... Chart 12...And A Stronger Dollar But Weaker EM Export Prices... ...And A Stronger Dollar But Weaker EM Export Prices... ...And A Stronger Dollar But Weaker EM Export Prices... Chart 13...Falling EM Stocks And Commodity Currencies... ...Falling EM Stocks And Commodity Currencies... ...Falling EM Stocks And Commodity Currencies... Chart 14...And Maybe Even A Correction In U.S. Stock Prices ...And Maybe Even A Correction In U.S. Stock Prices ...And Maybe Even A Correction In U.S. Stock Prices Fourth, gold is sending a similar signal as in the late 1990. As we have argued in the past, gold is a very good gauge of global liquidity conditions. During the Asian Crisis and the Russia/LTCM fiasco, industrial commodity prices only experienced a serious decline after the Thai baht had dragged down Asia into a tailspin. However, gold had been falling since 1996, a move predating the fall in Asian currencies (Chart 15). The precious metal was confirming that global liquidity was tightening and being sucked back into the booming U.S. economy. Today, gold prices are sending an ominous signal. After forming a large tapering wedge from 2011 to 2018, gold prices have broken down below the major upward-sloping trend line that had defined the bull market that began in 2001 (Chart 16). This indicates that gold may be starting another leg of a major bear market. Moreover, as the bottom panel of Chart 16 illustrates, it is true that net speculative positions in the yellow metal have plunged, but they remain far above the large net short positions that prevailed in the late 1990s. If gold is indeed entering another major down leg, this would confirm that tightening liquidity will further hurt EM asset prices, commodity prices and non-U.S. economic activity. Chart 15As Early As 1996, Gold Warned Of Upcoming Problems In Asia As Early As 1996, Gold Warned Of Upcoming Problems In Asia As Early As 1996, Gold Warned Of Upcoming Problems In Asia Chart 16Is A Secular Bear Market In Gold Beginning? Is A Secular Bear Market In Gold Beginning? Is A Secular Bear Market In Gold Beginning? Finally, adding insult to injury is China. The current communist party leadership is hell-bent on reforming the Chinese economy, moving it away from its dependence on capex and leverage. Consequently, China is in the midst of a major deleveraging campaign concentrated in the shadow banking sector, which has already caused money growth and total social financing to plumb to new lows (Chart 17). This is deflationary for the global economy as weaker Chinese credit weighs on capex, which in turns weighs on Chinese imports, as 69% of China's intake from the rest of the world are commodities and intermediate as well as industrial goods. Chart 17Chinese Monetary And Credit Conditions Remain ##br##Tight China Deleveraging Is Biting Chinese Monetary And Credit Conditions Remain Tight China Deleveraging Is Biting Chinese Monetary And Credit Conditions Remain Tight China Deleveraging Is Biting Chart 18No Capitulation ##br##Yet No Capitulation Yet No Capitulation Yet Moreover, the recent wave of renminbi weakness is exacerbating these deflationary pressures. The 9% fall in the yuan versus the dollar since April 11th represents a competitive devaluation that will hurt many EM countries. It also implies downside in China's import volumes, as it increases the prices paid by Chinese economic agents for foreign-sourced industrial goods and commodities.2 All these forces suggest that the pain that started in Argentina and Turkey could continue to spread across other vulnerable EM economies. It is doubtful that economies with large debt loads, large upcoming debt rollovers and other underlying economic problems will find it easy to receive financing in an environment of declining global liquidity, a strong dollar, budding deflationary pressures and a slowing China. Making this worry even more real, EM investors have not capitulated, as bottom-fishing has prompted massive inflows into Turkey in recent days (Chart 18). 2018 may not be 1997 or 1998, but it is likely to be a year to remember. Bottom Line: EM currency pegs to the dollar may not be as prevalent as they were back in the 1990s, but enough risks are present that contagion from Argentina and Turkey to other EM economies is a very real risk. Specifically, the domestic economic situation in the U.S. warrants higher interest rates, which suggests the Fed is unlikely to be fazed by EM market routs unless they become deep enough to present a threat to U.S. growth itself. Moreover, global liquidity conditions are tightening as the U.S.'s economic strength is sucking in capital from around the world. This combination means that EM countries with large dollar debt loads are likely to find debt refinancing a very onerous exercise. Finally, China is slowing and letting the RMB fall, which is exerting a deflationary impact on the world. Investment implications An environment of slower global economic activity, tightening global liquidity conditions and a potential deflationary scare is positive for the dollar. But 1998 shows that if the hot potato hides in the U.S. and the Fed is forced to ease aggressively, the dollar could nonetheless suffer. In order to get a sense as to whether the dollar can continue to strengthen or not, it is important to get a sense of where the exposure to an EM accident may lie. To begin this exercise, we need to first assess which EM countries are most vulnerable to catching the "Turkish Flu." To do so, we collaborated with our colleague Peter Berezin and his team at BCA's Global Investment Strategy to build a heat map of vulnerable EM economies. This heat map is based on the following factors: current account balance, net international investment position, external debt, external debt service obligation, external funding requirements, private sector savings/investment balance, private sector debt, government budget balance, government debt, foreign ownership of local currency bonds, and inflation. This method shows that after Turkey and Argentina, the next six most vulnerable countries are Colombia, Brazil, Mexico, Chile, South Africa, and Indonesia in this order (Chart 19). Chart 19Vulnerability Heat Map For Key EM Markets The Bear And The Two Travelers The Bear And The Two Travelers While our long-term valuation models show that the Colombian peso is already trading at a significant discount to its fair value, the BRL, the CLP, the ZAR, and the MXN are not (Chart 20). This highlights that these markets could provide serious fireworks in the coming months. Moreover, they all have their own idiosyncrasies that accentuate these risks. Brazil will soon undergo elections that will likely not result in a market-friendly outcome.3 Chile has an extremely large dollar-debt load, copper prices are tanking and the CLP is very pricey. Finally, South Africa is contemplating the kind of land expropriations reminiscent of those that plunged Zimbabwe into chaos - not a good optic for a still-expensive currency. So, who is most exposed to this potential mess? The answer is the euro area, most specifically, Spain. As Chart 21 shows, the exposure of Spanish banks to the most vulnerable EM markets totals nearly 170% of the banking system's capital and reserves. This means that 30% of the capital and reserves of the banking systems in the euro area's five largest economies is exposed to these markets. Making the risk even more acute, French banks have large exposure to Spain, and German banks to France. This combined exposure dwarfs the exposure of the U.K., Japan or the U.S. to the most vulnerable EM economies. To be fair to Spain, Spanish banks often have set up their foreign affiliates as separate legal entities. This means that the impact on the balance sheets of the Spanish banking system of defaults in vulnerable EM countries may be more limited than seems at face value. Yet, this is far from certain. Chart 20BRL, CLP, ZAR, And MXN Are Too Expensive##br## In Light Of Their Vulnerabilities BRL, CLP, ZAR, And MXN Are Too Expensive In Light Of Their Vulnerabilities BRL, CLP, ZAR, And MXN Are Too Expensive In Light Of Their Vulnerabilities Chart 21Who Has More Exposure To EM? The Bear And The Two Travelers The Bear And The Two Travelers As a result, we would not be surprised if the European Central Bank is forced by an EM accident to back away from its desire to abandon its extraordinary accommodative stance. The ECB would first use forward guidance to message that a hike will be delayed ever further in the future. The ECB may even be forced to resume government and corporate bonds purchases past 2018. This is a potential nightmare scenario for the euro. In fact, as Chart 22 illustrates, a euro at parity may not be a far stretch. Historically, the euro bottoms when it trades 10% below our fair value model, based on real short rate differentials, relative yield curve slopes and the ratio of copper to lumber prices. Such a discount would correspond to EUR/USD at parity. Because under such circumstances the Fed could be forced to pause its own hiking cycle for a quarter or two, a move to EUR/USD between 1.10 and 1.05 seems more likely than a collapse to parity right now. This also means that in conjunction with BCA's Geopolitical Strategy team, we recommend our clients close overweight positions in Spanish assets. Chart 22The Euro Still Has Downside If EM Go Bust The Euro Still Has Downside If EM Go Bust The Euro Still Has Downside If EM Go Bust What about the yen? In the late 1990s, the yen fell against the U.S. dollar as Asian currencies were collapsing, but surged once the Fed backtracked and bond yields tanked in 1998. This time could follow a different road map. Japan does not compete against Brazil, Colombia, Mexico, Chile and South Africa in the same way as it was competing against industrial companies in countries like Taiwan, Singapore or South Korea. This means that Japan is unlikely to need to competitively devalue to remain afloat if the BRL, COP, MXN, CLP and ZAR collapse further. However, since an EM shock is likely to prove to be a deflationary event, this means that bond yields could experience downside, especially as positioning in the U.S. bond market is massively crowded to the short side (Chart 23). A countertrend bull market in bonds would greatly flatter the yen. As a result, we are maintaining our short EUR/JPY bias over the coming months. The G10 commodity currency complex is also at risk. Not only does tightening dollar liquidity imply further weakness in this group of currencies, so does slowing EM activity and a deflationary scare. Additionally, the CAD and the NZD are not trading at much of a discount to their fair value, and the AUD trades at a premium (Chart 24). This means we would anticipate these currencies to suffer more in the coming quarters, led by the AUD, which is not only the most expensive of the group, but also the most geared to EM economic activity. Being short AUD/CAD still makes sense. Chart 23A Bond Rally Would ##br##Support The Yen A Bond Rally Would Support The Yen A Bond Rally Would Support The Yen Chart 24TDollar-Bloc Currencies Offer Limited Cushion##br## In The Event of An EM Selloff TDollar-Bloc Currencies Offer Limited Cushion In The Event of An EM Selloff TDollar-Bloc Currencies Offer Limited Cushion In The Event of An EM Selloff Finally, the pound is its own animal. GBP/USD is now quite cheap, but the U.K.'s large current account deficit of 3.9% of GDP, which is not funded through FDIs anymore, means that Great Britain remains vulnerable to tightening global liquidity conditions. Moreover, Brexit negotiations will heat up in the fall, as the March 2019 deadline for reaching a deal with the EU looms large. This means that political tumult in the U.K. will remain a large source of risk for the pound. We will explore the outlook for the pound in an upcoming report this September. Currently, our long DXY trade is posting an 8.5% profit, with a target at 98. The above picture suggests that the dollar could move well past 98, especially as the momentum factor that is so important to the greenback still plays in favor of the USD.4 As a result, we are upgrading our target on the dollar to 100. However, we are also tightening our stop loss to 94.88. We will update our stop loss to 97 if the DXY hits 98 in the coming weeks, in order to protect gains while still being exposed to the dollar's potential upside. Bottom Line: Beyond Turkey and Argentina, the EMs most vulnerable to tightening global liquidity conditions are Brazil, Colombia, Mexico, Chile and South Africa. Spanish banks have outsized exposure to these markets, which means the euro area is at risk if the "Turkish Flu" becomes contagious. As such, the ECB could be forced to remain easier than it wants to. The euro is still at risk. The yen could strengthen if global bond yields suffer. Hence, it still makes sense to be short EUR/JPY. While the CAD, AUD and NZD are also all vulnerable to a deflationary scare, the Aussie is the worst positioned of the three. Shorting AUD/CAD still makes sense. The DXY is likely to experience significant upside from here, with a move to 100 becoming an increasingly probable scenario. Risks To Our View Chart 25A Gauge And A Hedge Against Chinese Stimulus A Gauge And A Hedge Against Chinese Stimulus A Gauge And A Hedge Against Chinese Stimulus The biggest risk to our view is China. In 2016, a vicious EM selloff was staunched by a large wave of stimulus that put a floor under Chinese economic activity, and caused China to re-lever. The impact was felt around the world, lifting commodity prices and EM assets while plunging the dollar into a vicious selloff in 2017. It is conceivable that such an outcome materializes anew, especially as China is, in fact, injecting stimulus into its economy. However, as we wrote two weeks ago, the current stimulus still pales in comparison to what took place in 2015. Moreover, reforms and deleveraging have much greater primacy now than they did back then.5 BCA believes that the current wave of stimulus is not designed to cause growth to surge again, as was the case in 2015, but is instead aimed at limiting the negative impact of the ongoing trade war with the U.S. Yet, we cannot be dogmatic. Not only is it hard to gauge the actual degree of stimulus currently applied to the Chinese economy, there is a heightened risk that the flow of policy announcements causes a shift in the dominant narrative among market participants. Such a shift in attitudes could easily cause a mass buying of EM assets and commodities, delaying the day of reckoning for vulnerable EM. As a result, we continue to promulgate that investors track the behavior of our China Play Index, introduced two weeks ago (Chart 25).6 Not only does this index provide a live read on how traders are pricing in Chinese developments, but it also provides a great hedge for investors long the dollar, short EM, or short the commodity complex. Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com 1 In the panic of 1907, the Knickerbocker Trust Company went bankrupt, threatening the health of the U.S. banking system. The stock market crashed, money markets went into paralysis, and a consortium of bankers led by J.P. Morgan himself ended up acting as a lender of last resort, staunching the crisis. As a consequence of this panic, the Federal Reserve System was born in 1913. 2 For a more detailed discussion of the deflationary risk created by the RMB, please see Foreign Exchange Strategy Weekly Report, "What Is Good For China Doesn't Always Help The World", dated June 29, 2018, available at fes.bcaresearch.com 3 Please see Emerging Markets Strategy Special Report, "Brazil: Faceoff Time", dated July 27, 2018, available at ems.bcaresearch.com 4 Please see Foreign Exchange Strategy Special Report, "Riding The Wave: Momentum Strategies In Foreign Exchange Markets", dated December 8, 2017, available at fes.bcaresearch.com 5 Please see Foreign Exchange Strategy Weekly Report, "The Dollar And Risk Assets Are Beholden To China's Stimulus", dated August 3, 2018, available at fes.bcaresearch.com 6 Ibid. Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Closed Trades

Highlights Our antennae are twitching wildly, as the Kingdom of Saudi Arabia (KSA) walks back a widely telegraphed commitment to surge production. This occurs against the backdrop of a possible loss of as much as 2mm b/d in exports from Iran and Venezuela next year, with demand expected to remain fairly strong. U.S. President Donald Trump remains silent. We believe the proximate cause of KSA's reversal boils down to one or all of the following: President Trump told KSA to expect an SPR release ahead of November mid-terms; KSA found it difficult to maintain higher production; or Short-term demand for KSA's output is falling, so they reduced production. We have questioned the ability of KSA to sustain production above 10.5mm b/d for an extended period in the past. However, we believe July's 200k b/d cut was produced by a combination of No. 1 and No. 3. We expect KSA to build storage ahead of Iran sanctions. On the back of our updated balances modeling we are maintaining our 2H18 Brent ensemble forecast of $70/bbl, and raising our 2019 forecast to $80/bbl from $75/bbl (Chart of the Week): The front-loaded production increase we expected from OPEC 2.0 could be less than expected. Highlights Energy: Overweight. The U.S. EIA reported U.S. crude and product inventories rose 17.4mm barrels for the week ended August 10, 2018. Markets traded sharply lower as a result, falling more than 3% in WTI and 2% in Brent. As we went to press, October Brent was trading just above $70/bbl. We are maintaining our $70/bbl Brent forecast for 2H18. Base Metals: Neutral. Union leaders at BHP's Escondida mine in Chile, the largest in the world, will take proposed contract terms to members this week.1 We were stopped out of our tactical Dec18 copper call spread with a 10.2% loss. Precious Metals: Neutral. Gold remains under pressure as the broad trade-weighted USD rises. We remain long as a portfolio hedge. Ags/Softs: Underweight. USDA export data show year-to-date wheat and soybean exports are down 20% and 10% y/y in the Oct17 - Jun18 period. Feature Forward guidance from OPEC 2.0's leadership and its predecessor, the regular old OPEC, has not been helpful of late.2 This complicates our balances assessment this month (Chart of the Week), and raises the odds volatility will increase sooner than we expected. Chart of the Week2H18 Brent Forecast Stays At $70/bbl, 2019 Moved Up To $80/bbl 2H18 Brent Forecast Stays At $70/bbl, 2019 Moved Up To $80/bbl 2H18 Brent Forecast Stays At $70/bbl, 2019 Moved Up To $80/bbl KSA's reversal in July of its earlier, widely telegraphed decision to sharply raise production in response to aggressive tweeting from U.S. President Donald Trump beginning in May - to as much as 11mm b/d from just over 10mm b/d in the first five months of this year - was followed by an abrupt output cut of ~ 200k b/d last month. Last month, we expected KSA's crude production to average 10.60mm b/d in 2H18, and 10.50mm b/d next year. In our current balances estimate (Table 1), we now expect the Kingdom's output to average 10.28mm b/d in 2H18 and 10.35mm b/d in 2019, down 300k b/d and 150k b/d, respectively. Table 1BCA Global Oil Supply - Demand Balances (MMb/d) (Base Case Balances) OPEC 2.0 Sailing Close To The Wind OPEC 2.0 Sailing Close To The Wind Russia, OPEC 2.0's other putative leader, also is complicating assessments of liquids production by the producer coalition. Given the signaling it and KSA were providing over the past couple of months, we expected Russia to raise production 80k b/d in 2H18 to 11.27mm b/d, and by 160k b/d in 2019 to 11.35mm b/d. We still expect Russia to raise its production and revised our baseline estimates to 11.32mm b/d and to 11.43mm b/d for this year and next, respectively. However, it is difficult to reconcile our expectation with the 11.13mm b/d 2H18 liquids production expected by OPEC for Russia in its August Monthly Oil Market Report (MOMR), as we are highly confident Russia signed off on that estimate before it was published. Chart 2Physical Deficit Worsens Physical Deficit Worsens Physical Deficit Worsens Our global liquids supply estimate for 2H18 now stands at 101.08mm b/d, down 680k b/d from last month's estimate. For 2019, we lowered our supply estimate by 800k b/d to 101.01mm b/d. But this could end up overstating supply, given what we're seeing from OPEC 2.0 presently. On the demand side, we've lowered our 2018 and 2019 expectations slightly - to 1.67mm b/d and 1.62mm b/d, respectively, or ~ 50k b/d on average versus our previous estimates. This is still relatively stout demand growth - supported by still-strong global trade, particularly in the EM economies - which means storage will be forced to draw harder next year than we expected even a month ago (Chart 2). Physical Deficit Worsens In 2019 We expected OPEC 2.0's supply increase would persist at a higher level during 2H18, which would allow refiners to build precautionary inventories going into next year. This no longer is a tenable assumption, given what is being reported for OPEC 2.0's largest producers - KSA and Russia. In addition, we have amended our base case supply model, to reflect the loss of 1mm b/d of Iranian exports to U.S. sanctions for most of next year; we have this occurring in 250k b/d increments in the Nov18 - Feb 19 period, leaving production from March 2019 on at 2.8mm b/d. This replaces our earlier assumption of a 500k b/d by the end of 1H19. We took this action on the back of the increasingly strident rhetoric from the U.S. administration, and press reports indicating widespread compliance with the sanctions is expected - particularly reports suggesting China and India will not be looking to increase purchases of Iranian crude. Offsetting the higher Iranian export losses we foresee, our base case includes a re-start of Neutral Zone production in 2Q19.3 We expect KSA and Kuwait to each bring 175k b/d back on line, for a total of 350k b/d. It is not clear this is counted in both countries' spare capacity, but if it is, then spare capacity will become tighter within OPEC 2.0 next year. In our scenario analysis, we continue to give a relatively high weight to the loss of Venezuela's exports - anywhere from 800k to 1mm b/d - as that country's oil industry continues to degrade. Our ensemble analysis indicates OECD storage will draw more than previously estimated (Chart 3), on the back of these higher assumed Iranian export losses, and a reduction in OPEC 2.0's front-loaded production increases, particularly in 2019. As storage draws, days-forward-cover (DFC) also will contract (Chart 4). In addition to steepening the backwardation in crude forward curves, we expect implied option volatility to increase in 2019 (Chart 5). Chart 3Storage Will Draw##BR##Harder Next Year Storage Will Draw Harder Next Year Storage Will Draw Harder Next Year Chart 4Days-Forward-Cover##BR##Will Fall In 2019 Days-Forward-Cover Will Fall In 2019 Days-Forward-Cover Will Fall In 2019 Chart 5Implied Volatilities Will Rise,##BR##As OECD Storage Falls OPEC 2.0 Sailing Close To The Wind OPEC 2.0 Sailing Close To The Wind Ensemble Forecast Update In addition to moving the 1mm b/d loss of Iranian exports from a scenario and into our base case - offset somewhat by higher Neutral Zone production - we expect transportation bottlenecks in the Permian Basin to slow production growth in the U.S. shales even more. We have lowered our expected U.S. production growth to 1.21mm b/d this year and 1.22mm b/d in 2019, versus earlier estimates of 1.30mm b/d and 1.34mm b/d, as a result (Chart 6 shows the trajectory we expect from this scenario).4 Coupled with the lower-than-expected production increase from OPEC 2.0 and still-strong demand growth globally, this will lead to tighter markets in 2019. Chart 6Higher Volatility = Wider Expected Price Range Higher Volatility = Wider Expected Price Range Higher Volatility = Wider Expected Price Range We also are including a scenario showing a slowdown in demand growth, which takes y/y growth to 1.43mm b/d in 2018 and 2019, versus our current estimates of average growth of 1.64mm b/d over the two-year interval. Bottom Line: Numerous conflicting data have entered the oil pricing picture over the past month, which greatly complicates our analysis and forecasting. The fact that OPEC 2.0's leadership - KSA and Russia - is providing little in the way of forward guidance does not make this any easier. We admit to being puzzled by KSA's apparent decision to walk back its production increase going into 2019, when the likelihood of losing close to 2mm b/d of exports from Iran and Venezuela becomes markedly higher. Based on our current modeling we expect higher prices next year ($80/bbl vs. our earlier estimate of $75/bbl for Brent), and a steepening of the Brent and WTI backwardations next year. We continue to expect WTI to trade $6/bbl below Brent in 2H18 and 2019. The steepening backwardation will lift implied volatility, particularly next year. We remain long call option spreads along the Brent forward curve in 2019, in expectation prices and volatility will move higher. We continue to believe the balance of price risk is to the upside. However, as the lower-demand scenario in our ensemble forecast shows, an unexpected slowdown in growth can have profound effects on prices. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com Hugo Bélanger, Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com 1 Please see "Chile's Escondida union to take new labor proposal to members," published by reuters.com August 15, 2018. 2 OPEC 2.0 is the name we coined for the producer coalition led by the Kingdom of Saudi Arabia (KSA) and Russia. At the end of June, the coalition's member states agreed to increase production to bring it into line with the originally agreed deal to remove 1.8mm b/d of output from the market. 3 Please see "Kuwait, Saudi to resume output from Neutral Zone in 2019 - Toyo Engineering," published by reuters.com July 2, 2018. 4 We place our scenarios within the context of a market-generated confidence interval, which we calculate using implied volatilities derived from Brent and WTI options markets. Please see Ryan, Bob and Tancred Lidderdale (2009), "Energy Price Volatility and Forecast Uncertainty," particularly Appendix 1 beginning on p. 18, for a derivation of the confidence intervals. The article was published by the U.S. EIA. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table OPEC 2.0 Sailing Close To The Wind OPEC 2.0 Sailing Close To The Wind Trades Closed in 2018 Summary of Trades Closed in 2017 OPEC 2.0 Sailing Close To The Wind OPEC 2.0 Sailing Close To The Wind

Highlights Seasonal capacity restrictions in China during the winter heating months - when pollution from steel mills is particularly high - and continued efforts to limit particulate emissions in major cities will drive steel prices higher. The steel rebar market in China is backwardated, indicating physical markets are tight; inventories have been falling since mid-March. We expect prices to remain elevated going into the winter months, when capacity restrictions kick in. Ongoing capacity reductions in steelmaking will favor higher-grade iron ores, which will widen price differentials versus lower-grade ores. We are recommending a long China rebar futures on the SHFE in 1Q19 vs short 62% Fe iron ore futures on the Dalian DCE in 1Q19 at tonight's close, based on our research. Energy: Overweight. Loadings of Iranian crude are expected to be curtailed beginning this month, as the November 4 deadline for the imposition of U.S. secondary sanctions kick in. Our base case calls for the loss of 500k b/d of exports from Iran; our ensemble forecast includes an estimate of 1mm b/d. Base Metals: Neutral. BHP asked the Chilean government to intervene in the strike called by unions at its Escondida mine. Union officials delayed strike action while talks are being held. Negotiators have until August 14 to reach an agreement. Reuters reported Chile's copper production was up 12.3% y/y in 1H18 to 2.83mm MT.1 Precious Metals: Neutral. U.S. sanctions on trading gold and precious metals with Iran went into effect earlier this week. Ags/Softs: Underweight. Chinese imports of U.S. soybeans could fall 10mm MT over the next year, if pig and chicken farmers switch to lower-protein feed and substitutes like sunflower seeds, and boost local production of the legume, state-run news service Xinhua reported.2 The USDA expects U.S. exports of 55.52mm MT of soybeans in the 2018 - 19 crop year, down 1.22mm MT from last year. Feature Steel prices have performed exceptionally since the beginning of 2Q18, seemingly oblivious to Sino - U.S. trade tensions, a stronger USD, and risks to China's economy roiling other metal markets (Chart of the Week). The MySteel Composite Index we use to track steel prices is up 7% since the beginning of April. With demand growth leveling off, steel's price dynamics highlight the continued relevance of the market's supply-side developments. Most notably, Beijing's battle for blue skies: Winter capacity curbs, and, to a lesser extent, ongoing efforts to retire older, highly polluting capacity will keep prices elevated over the next 9 months. Winter Curbs: China's New Normal As we highlighted in our April 12 weekly, despite the much-ballyhooed reductions in China's steel capacity over the 2017 - 18 winter months, markets in China and globally remained relatively well supplied over the winter.3 However, several key changes this year suggest the impact of these measures will intensify this time around, keeping producers constrained in their ability to ramp up production of the metal. For one, the data suggest strong production levels amid the anti-pollution curbs last winter were a result of an increase in output from regions unaffected by the capacity restrictions (Chart 2). This went a long way in muting the impact of the restrictions in the heavily industrialized Beijing-Tianjin-Hebei region of northern China. Chart of the WeekSteel Oblivious To Pessimism Steel Oblivious To Pessimism Steel Oblivious To Pessimism Chart 22017/18 Winter Cuts: A Net Non-Event Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy This year's curbs will broaden the regions targeted by anti-pollution restrictions. The campaign will encompass 83 cities, up from last year's 28, thereby reducing the potential production ramp up from regions not covered by these measures (Chart 3). This coming winter's closures will cover regions where producers traditionally account for 68% of China's steel output (Chart 4). Chart 3Second Annual Winter Capacity ##br##Restrictions Will Broaden Coverage... Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy Chart 4...And##br## Impact Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy The anti-pollution campaign is one of the three battles prioritized in Xi Jinping's plan for the coming years. These curbs will be implemented during the October 1, 2018 to March 31, 2019 heating season, extending the duration from last year's mid-November to Mid-March period. Because the minimal effect observed per last year's closures was due to specifying too narrow a range of plants and regions, not to non-compliance, we expect the measures announced for this coming winter to be fully implemented. These measures come amid already-tight market conditions. The steel rebar market in China is in backwardation - meaning a physical shortage is pushing up prompt prices relative to those further out the curve. Inventories have been falling since mid-March, reflecting supply-demand dynamics in other steel product markets. Thus, we expect prices to remain elevated going into the winter months. Capacity Impacts Are Difficult To Gauge Opaqueness and discretionary authority in the new rules clouds the outlook on how anti-pollution reforms will impact the steel market. This makes it difficult to estimate their impact with precision. This time around, China's State Council announced that curbs will be implemented in a more scientific and targeted approach, ensuring maximum efficiency to attain the targets. This means the constraints this year will depend on emissions in each region, which will be set at the discretion of local authorities.4 For example, steel mills in six key cities including Tianjin, Shijiazhuang, Tangshan, Handan, Xingtai and Anyang will be asked to keep capacity below 50% this winter, while producers in the rest of the Beijing-Tianjin-Hebei region will keep production running at less than 70% of capacity. Furthermore, a draft plan by the city of Changzhou - which planned to implement the curbs beginning August 3 - suggests production curbs may vary by company, depending on operational situations and emission levels.5 These restrictions are applied to capacity, rather than production. Without up-to-date and accurate information on crude steel-making capacity across the different regions, it is extremely difficult to accurately quantify the impact. Specifics of the plans are up to the discretion of local authorities. Thus, these restrictions can be applied to different stages in the steel-making process (Diagram 1), impacting furnaces, pig iron or sintering plants. In some cases, the output curbs are not only restricted to the winter heating months. Several regions have been implementing curbs throughout the year on an as-needed basis. The cities of Tangshan and Changzhou are two such examples, implementing restrictions during the summer months as well. Furthermore, all industrial plants in the city of Xuzhou remain shut. High profit margins at steel mills may incentivize the shuttered illegal furnaces to restart. The industry ministry acknowledges this threat, and claims it will carry out checks on these producers to ensure they do not come back online. Diagram 1Steelmaking Production Process: Restrictions Can Be Applied To Different Stages Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy Without full knowledge of these details, quantifying the impact of these restrictions is a challenge. Morgan Stanley estimates the impact of these curbs on steel output to be 78mm MT during the winter period by assuming capacity utilization is restricted to 50% in the key cities, while the rest of the areas cut capacity by 30%. The estimated production loss from these restrictions accounts for 9% of China's 2017 crude steel output.6 China's Ongoing Capacity-Reduction Reforms Most of the planned permanent capacity shutdowns have already taken place. Of the targeted 150mm MT of cuts between 2016 and 2020, 115mm MT have already taken place over the past two years. Furthermore, 1H17 witnessed the closure of all illegal induction furnaces producing sub-par quality steel, estimated to account for 140mm MT of crude steel capacity (Table 1).7 Table 1De-Capacity Reforms Still Ongoing Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy We expect the magnitude of cutbacks to slow considerably. Even though the industry ministry issued a statement in February that it plans to meet steel capacity reduction targets this year - two years ahead of schedule. Furthermore, mills face restrictions on new steel capacity. China's State Council announced it intends to prevent new steel capacity additions in the Beijing-Tianjin-Hebei, Guangdong province, and Yangtze River Delta regions, and a cap set at 200mm MT in Hebei by 2020. The capacity replacement plan, which allows a maximum of 0.8 MT of new capacity for each MT of eliminated capacity, will ensure capacity does not grow going forward. In fact, not all mills are eligible to take advantage of the replacement policy. Among others, now-shuttered induction furnace capacity, as well as producers that previously benefited from cash and policy support will not meet the requirements for this program. Steel And Iron Ore Prices Will Not Reconverge As a result of China's reform policies in the steel industry, iron ore prices have diverged from steel. Reduced steel production lowers demand for raw materials, including iron ore. This is reflected in falling Chinese iron ore imports amid contracting production (Chart 5). Chart 5Weak Demand For Iron Ore Weak Demand For Iron Ore Weak Demand For Iron Ore Chart 6EAF Penetration In China: Still Some Catching Up To Do Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy China's reform and anti-pollution campaigns have had serious consequences on iron ore markets. For starters, China is encouraging the adoption of electric arc furnaces (EAF), rather than additional new blast furnaces.8 While the latter primarily uses iron ore, the former uses scrap steel. EAF penetration in China's steel industry significantly lags the rest of the world (Chart 6). This means that even if the capacity-replacement program allows eliminated furnaces to be replaced with newer, more up-to-date capacity, this will not spur demand for iron ore. Instead, we expect to see higher scrap steel prices (Chart 7). Furthermore, as we first highlighted in our January report, China's anti-pollution campaign coupled with high steel profit margins has incentivized the use of higher grade iron ore and iron ore pellets, widening the price spread between high- and low- grade ores (Chart 8).9 Chart 7EAFs Support Scrap Steel Demand EAFs Support Scrap Steel Demand EAFs Support Scrap Steel Demand Chart 8IO Grade Premiums Will Remain Elevated IO Grade Premiums Will Remain Elevated IO Grade Premiums Will Remain Elevated While high-grade ores are more expensive, they emit less pollution in the steelmaking process. Similarly, unlike fines, pellets which are direct charge feedstock, are not required to undergo the highly polluting sintering stage and can be fed directly into the furnace. China's Steel Dynamics Overshadow Global Markets The ongoing supply-side reforms in China are overshadowing events in other markets. Globally, steel is expected to remain in physical deficit this year (Chart 9). This is largely on the back of an increase in world ex-China demand, and the decline in Chinese supply, despite expectations of weaker Chinese demand, and increased supply from the rest of the world (Table 2). Chart 9Physical Steel Deficit Will Persist... Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy Table 2...Despite Weaker Chinese Demand And Stronger RoW Supply Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy These figures do not consider the impact of the ongoing Sino - U.S. trade dispute, which could evolve into a full-blown trade war, weighing on EM incomes and demand. In such a scenario, global demand for steel would take a hit, potentially shifting global markets into surplus. In theory, trade barriers on U.S. steel imports could lead to weaker domestic supply for American users and at the same time, leave more of the metal for use by the rest of the world. The net effect of that would be a higher price for American steel relative to the rest of the world. However, since May, 20,000 requests for steel tariff exemptions have been filed in the U.S., of which the Commerce Department has denied 639. To the extent that American steel users are able to obtain tariff exemptions, the impact of the barriers on global steel markets will be muted. Bottom Line: We expect China's steel market to tighten as we go into the winter season, during which capacity cuts will be broadened to 82 cities, from last year's 28. This will keep steel prices elevated. At the same time, we expect prices of 62% Fe material and lower iron ore grades to weaken, as appetite for the steelmaking raw material contracts during these months. Mills still running in the mid-November to mid-March period will have a preference for higher-grade ores and pellets, keeping premiums on these grades elevated. Barring a significant demand-side shock, expect more upside to steel prices and downside to iron ore prices over the coming 9 months. Based on our research, we are recommending a long China rebar futures on the SHFE in 1Q19 vs. short 62% Fe iron ore futures on the Dalian DCE in 1Q19 at tonight's close. Roukaya Ibrahim, Editor/Strategist Commodity & Energy Strategy RoukayaI@bcaresearch.com 1 Please see "BHP asks for government mediation in talks at Chile's Escondida," published August 6, 2018, by uk.reuters.com. 2 Please see "Economic Watch: China can cut soybean imports in 2018 by over 10 mln tonnes," published August 5, 2018, by xinhuanet.com. 3 Please see Commodity & Energy Strategy Weekly Report titled "Chinese Steel, Aluminum Markets Well Supplied Despite Winter Capacity Cuts," dated April 12, 2018, available at ces.bcaresearch.com. 4 Please see "Chinese steel output cuts to vary from mill to mill next winter," dated July 21, 2018, available at reuters.com. 5 The restrictions will not only apply to the city's steel mills, but also to copper smelters, chemical makers as well as cement producers. Please see "China's Changzhou plans to enforce output curbs in steel, chemical plants," dated July 30, 2018, available at reuters.com. 6 Please see "Shanghai steel resumes rise, coke rallies as China eyes winter curbs," dated August 2, 2018, available at reuters.com. 7 Low-quality steel produced by induction furnaces, also referred to as ditiaogang, is made by melting scrap steel using induction heat, preventing sufficient control over the quality of the steel. Platts estimates ditiaogang production in 2016 to be 30-50mm MT. As we explain in our September 7, 2017 Weekly Report titled "Slow-Down In China's Reflation Will Temper Steel, Iron Ore In 2018," given that ditiaogang is illegal, these closures are not reflected in official steel production figures. Thus the closures of these mills have no impact on actual steel production, but instead raise the capacity utilization rates for Chinese steel producers. 8 China launched a carbon trading system in January 2018, which penalizes blast furnace operators with higher environmental taxes relative to EAF processes. 9 Please see Commodity & Energy Strategy Weekly Report titled "China's Environmental Reforms Drive Steel & Iron Ore," dated January 11, 2018, available at ces.bcaresearch.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy Trades Closed in 2018 Summary of Trades Closed in 2017 Blue Skies Drive China's Steel Policy Blue Skies Drive China's Steel Policy

Highlights Paradox 1: U.S. growth will slow, and this will force the Fed to raise rates MORE quickly. Paradox 2: China will try to stimulate its economy, and this will HURT commodities and other risk assets. Paradox 3: Global rebalancing will require the euro area and Japan to have LARGER current account surpluses. Feature Faulty Assumptions Investors assume that slower U.S. growth will cause the Fed to turn more dovish; efforts by China to stimulate its economy will boost market sentiment towards risk assets; and global rebalancing requires the euro area and Japan to reduce their bloated current account surpluses. In this week's report, we consider the possibility that all three assumptions are wrong. Let's start with the U.S. growth picture. U.S. Growth About To Slow? The U.S. economy grew by 4.1% in the second quarter, the fastest pace since 2014. The composition of growth was reasonably solid. Net exports boosted real GDP by 1.1 percentage points, but this was largely offset by a 1.0 point drag from a slower pace of inventory accumulation. As a result, domestic final demand increased at a robust rate of 3.9%, led by personal consumption (up 4.0%) and business fixed investment (up 7.3%). Unfortunately, the second quarter is probably as good as it gets for growth. We say this not because we expect aggregate demand growth to falter to any great degree. Quite the contrary. Consumer confidence is high and the labor market is strong, with initial unemployment claims near 49-year lows. The Bureau of Economic Analysis' latest revisions revealed a much higher personal savings rate than had been previously estimated (Chart 1). The savings rate is now well above levels that one would expect based on the ratio of household net worth-to-disposable income (Chart 2). This raises the odds that consumer spending will accelerate. Chart 1Households Are Saving More ##br##Than Previously Thought Households Are Saving More Than Previously Thought Households Are Saving More Than Previously Thought Chart 2Consumption Could Accelerate ##br##As The Savings Rate Drops Three Macro Paradoxes Are About To Come True Three Macro Paradoxes Are About To Come True Rising consumer demand will prompt businesses to expand capacity (Chart 3). Core capital goods orders surprised on the upside in June, with positive revisions made to past months. Capex intention surveys remain at elevated levels. So far, fears of a trade war have not had a major impact on business investment. Fiscal spending is also set to rise. Federal government expenditures increased by only 3.5% in Q2, far short of the 10%-plus growth rate that some forecasters were projecting. The effect of the tax cuts have also yet to make their way fully through the economy. Supply Matters Considering all these positive drivers of demand, why do we worry that growth could slow meaningfully later this year or in early 2019? The answer is that for the first time in over a decade, demand is no longer the binding constraint to growth - supply is. Today, there are fewer unemployed workers than job vacancies (Chart 4). The number of people outside the labor force who want a job is near all-time lows. Businesses are reporting increasing difficulty in finding qualified labor. Chart 3U.S. Companies Plan To Boost Capex U.S. Companies Plan To Boost Capex U.S. Companies Plan To Boost Capex Chart 4Companies Are Struggling To Fill Job Openings Companies Are Struggling To Fill Job Openings Companies Are Struggling To Fill Job Openings New business investment will add to the economy's productive capacity over time, but in the near term, the boost to aggregate demand from new investment spending will easily exceed the contribution to aggregate supply.1 The Congressional Budget Office estimates that potential real GDP growth is running at around 2%. What happens when the output gap is fully eliminated, and aggregate demand growth begins to eclipse supply growth? The answer is that inflation will rise. Instead of more output, we will see higher prices (Chart 5). Chart 5Inflationary Pressures Tend To Increase ##br##When Spare Capacity Is Absorbed Three Macro Paradoxes Are About To Come True Three Macro Paradoxes Are About To Come True Rising inflation will force the Fed to engineer an increase in real interest rates, even in the face of slower GDP growth. Such a stagflationary outcome is not good for equities, which is one reason why we downgraded our cyclical recommendation on risk assets from overweight to neutral in June. Higher-than-expected real interest rates will put upward pressure on the U.S. dollar. A stronger dollar will hurt U.S. companies with significant foreign exposure more than it hurts their domestically-oriented peers. If history is any guide, a resurgent greenback will also cause credit spreads to widen (Chart 6). Chinese Stimulus: Be Careful What You Wish For Chinese stimulus helped reignite global growth after the Global Financial Crisis and again during the 2015-2016 manufacturing downturn. With global growth slowing anew, will China once again come to the rescue? Not quite. China does not want to let its economy falter, but high debt levels, and an overvalued property market plagued by excess capacity, limit what the authorities can do (Chart 7). Chart 6A Stronger Dollar Usually Corresponds ##br##To Wider Corporate Borrowing Spreads A Stronger Dollar Usually Corresponds To Wider Corporate Borrowing Spreads A Stronger Dollar Usually Corresponds To Wider Corporate Borrowing Spreads Chart 7China: High Debt Levels Make ##br##Credit-Fueled Stimulus A Risky Proposition Three Macro Paradoxes Are About To Come True Three Macro Paradoxes Are About To Come True Granted, the government has loosened monetary policy at the margin and plans to increase fiscal spending. However, our China strategists feel these actions are more consistent with easing off the brake than pressing down on the accelerator.2 They note that the authorities continue to squeeze the shadow banking system, as evidenced by the continued deceleration in money and credit growth, as well as rising onshore spreads for the riskiest corporate bonds (Chart 8). The Specter Of Currency Wars If Chinese growth continues to decelerate, what options do the authorities have? One possibility is to double down on what they are already doing: letting the RMB slide. Chart 9 shows that the Chinese currency has weakened substantially more over the past six weeks than its prior relationship with the dollar would have suggested. Chart 8Chinese Credit Growth Has Been Slowing Chinese Credit Growth Has Been Slowing Chinese Credit Growth Has Been Slowing Chart 9The Yuan Has Weakened More Than Expected ##br##Based On the Broad Dollar Trend The Yuan Has Weakened More Than Expected Based On the Broad Dollar Trend The Yuan Has Weakened More Than Expected Based On the Broad Dollar Trend Letting the currency weaken is a risky strategy. Global financial markets went into a tizzy the last time China devalued the yuan in August 2015. The devaluation triggered significant capital outflows, arguably only compounding China's problems. This has led some commentators to conclude that the authorities would not make the same mistake again. But what if the real mistake was not that China devalued its currency, but that it did not devalue it by enough? Standard economic theory says that a country should always devalue its currency by enough to flush out expectations of a further decline. Perhaps China was simply too timid? Capital controls are tighter in China today than they were in 2015. This gives the authorities more room for maneuver. China is also waging a trade war with the United States. The U.S. exported only $188 billion of goods and services to China in 2017, a small fraction of the $524 billion in goods and services that China exported to the United States. China simply cannot win a tit-for-tat trade war with the United States. In contrast, China is better positioned to wage a currency war with the United States. The Chinese simply need to step up their purchases of U.S. Treasurys, which would drive up the value of the dollar. Efforts by China to devalue its currency would invite retaliation from the United States. However, since the Trump Administration seems keen on pursuing a protectionist trade agenda no matter what happens, the Chinese may see their decision to weaken the yuan as the least bad of all possible outcomes. Unlike traditional stimulus in the form of additional infrastructure spending and faster credit growth, a currency devaluation would roil financial markets, causing risk asset prices to plunge. Metal prices would take it on the chin, since a weaker RMB would make it more expensive for Chinese businesses to import commodities. China now consumes close to half of the world's supply of copper, zinc, nickel, aluminum, and iron ore (Chart 10). Investors should remain underweight emerging market equities relative to developed markets and shun the currencies of commodity-exporting economies. We are currently short AUD/CAD on the grounds that a China shock would hurt metal prices more than energy prices. The Canadian dollar is highly levered to the latter, while the Aussie dollar is more levered to the former. Global Rebalancing: It's Not About Getting To Zero We have argued before that China's high savings rate explains why the country has maintained a structural current account surplus, despite the economy's rapid GDP growth rate.3 Both the euro area and Japan also have an excessive savings problem, minus the mitigating effect of rapid trend growth. The euro area's excessive savings problem was masked during the nine years following the introduction of the euro by a massive credit boom across much of the region (Chart 11). Germany did not partake in that boom, but it was still able to export its excess savings to the rest of the euro area via a rising current account balance. Chart 10China Is A More Dominant Consumer ##br##Of Metals Than Oil China Is A More Dominant Consumer Of Metals Than Oil China Is A More Dominant Consumer Of Metals Than Oil Chart 11Germany Did Not Take Part ##br##In The Credit Boom Germany Did Not Take Part In The Credit Boom Germany Did Not Take Part In The Credit Boom Germany Needs A Spender Of Last Resort Chart 12 shows that Germany's current account surplus with other euro area members mirrored the country's increasing competitiveness vis-à-vis the rest of the region. In essence, the spending boom in southern Europe sucked in German exports, with German savings financing the periphery's swelling current account deficits. This is the main reason why German banks were hit so hard during the Global Financial Crisis: They were the ones who underwrote the periphery's spendthrift ways. That party ended in 2008. With the periphery no longer the spender of last resort in Europe, Germany had to find a way to export its savings to the rest of the world. But that required a cheaper currency, which Mario Draghi ultimately delivered in 2014 when he set in motion the ECB's own quantitative easing program. So where do we go from here? Germany's excess savings problem is not about to go away anytime soon. The working-age population is set to decline over the next few decades, which means that most domestically oriented businesses will have little incentive to expand capacity (Chart 13). The peripheral countries remain in belt-tightening mode. This will limit demand for German imports. Meanwhile, countries such as Spain have made significant progress in reducing unit labor costs in an effort to improve competitiveness and shift their current account balances back into surplus. Chart 12Competitiveness Gains In The 2000s Allowed ##br##Germany To Increase Its Current Account Surplus Competitiveness Gains In The 2000s Allowed Germany To Increase Its Current Account Surplus Competitiveness Gains In The 2000s Allowed Germany To Increase Its Current Account Surplus Chart 13Germans Need To Have More Children Three Macro Paradoxes Are About To Come True Three Macro Paradoxes Are About To Come True The ECB And The BOJ Can't Afford To Raise Rates The private sector financial balance in the euro area - effectively, the difference between what the private sector earns and spends - now stands near a record high (Chart 14). Fiscal policy also remains fairly tight. The IMF estimates that the euro area's cyclically-adjusted primary budget balance will be in a surplus of 0.9% of GDP in 2018-19, compared to a deficit of 3.8% of GDP in the United States (Chart 15). Chart 14Euro Area: Private Sector ##br##Balance Remains Elevated Euro Area: Private Sector Balance Remains Elevated Euro Area: Private Sector Balance Remains Elevated Chart 15The Euro Area's Fiscal Policy Is Tight The Euro Area's Fiscal Policy Is Tight The Euro Area's Fiscal Policy Is Tight If the public sector is unwilling to absorb the private sector's excess savings by running large fiscal deficits, those savings need to be exported abroad in the form of a current account surplus. Failure to do so will result in higher unemployment, and ultimately, further political upheaval. This means that the ECB has no choice other than to keep rates near rock-bottom levels in order to ensure that the euro remains cheap. Japan has been more willing than Europe to maintain large budget deficits, but the problem is that this has resulted in a huge debt-to-GDP ratio. The Japanese would like to tighten fiscal policy, starting with the consumption tax hike scheduled for October 2019. However, this may require the economy to have an even larger current account surplus, which can only be achieved if the yen weakens further. This, in turn, suggests that the Bank of Japan will not abandon its yield curve control policy anytime soon. We were not in the least bit surprised this week when Governor Kuroda poured cold water on the idea that the BoJ was contemplating raising either its short or long-term interest rate targets. The bottom line is that thinking about global imbalances solely in terms of current account positions is not enough. One should also think about the distribution of aggregate demand across the world. Countries with demand to spare such as the United States can afford to run current account deficits, while economies with insufficient demand such as the euro area and Japan should run current account surpluses. The key market implication is that interest rates will remain structurally higher in the United States, which will keep the dollar well bid. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 1 This is partly because it can take a while for additional capital spending to raise aggregate supply. For example, it may take a few years to build an office tower or a new factory. Corporate R&D investment may not generate tangible benefits for a long time, especially in cases where the research is focused on something complicated (i.e., the design of new computer chips or pharmaceuticals). And even if investment spending could be transformed into additional productive capacity instantaneously, aggregate demand would still rise more than aggregate supply, at least temporarily. Here is the reason: The nonresidential private-sector capital stock is about 120% of GDP in the United States. As such, a one percent increase in investment spending would raise the capital stock by four-fifths of a percentage point. Assuming a capital share of income of 40% of national income, a one percent increase in the capital stock would lift output by 0.4%. Thus, a one-dollar increase in business investment would boost aggregate demand by one dollar in the year it is undertaken, while increasing supply by only 4/5*0.4 = roughly 32 cents. 2 Please see China Investment Strategy Weekly Report, "China Is Easing Up On The Brake, Not Pressing The Accelerator," dated July 26, 2018. 3 Please see Global Investment Strategy Weekly Report, "U.S.-China Trade Spat: Is R-Star To Blame?" dated April 6, 2018. Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

  Overweight The divergence between rising crude oil prices and the performance of exploration & production (E&P) stocks has grown remarkably wide (top panel). We continue to credit the absence of market belief in the longevity of the increase in oil prices. However, we think soaring industry capex means the view on the ground is much more positive. Planned capex has reached a level not seen since the recession, despite oil prices remaining well below the 2010-2014 highs, implying E&P companies will be a solid capex upcycle play, which remains a key BCA theme for the year (middle panel). Considering the diffusion indexes for unfilled orders and current employment in Texas, both of which set decade highs in July (bottom panel), we see little that stands in the way of the recovery. We reiterate our overweight recommendation on the S&P oil & gas E&P index and our high-conviction overweight recommendation on the broader S&P energy index. The ticker symbols for the stocks in this index are: BLBG: S5OILP - COP, EOG, APC, PXD, DVN, CXO, MRO, APA, HES, NBL, EQT, COG, XEC and NFX.   Going Big In Texas Going Big In Texas  

Highlights The eye of the storm is passing over the oil market. OPEC 2.0's recent production increase will temporarily halt the sharp decline in OECD commercial oil inventories, allowing stocks of crude oil and refined products in member states to level off ahead of the sharp drawdowns we expect next year (Chart of the Week).1 This will keep the front of Brent's forward curve in a modest contango going into 4Q18, and suppress short-term price volatility. Thereafter, reduced OPEC 2.0 output post-U.S. midterm elections, and lower Iranian and Venezuelan exports will force OECD inventories to resume drawing sharply, backwardating Brent's forward curve and raising oil price volatility (Chart 2).2 Chart of the WeekOECD Inventories Rebuild Slightly,##BR##Then Resume Falling Next Year OECD Inventories Rebuild Slightly, Then Resume Falling Next Year OECD Inventories Rebuild Slightly, Then Resume Falling Next Year Chart 2Brent, WTI Implied Volatility Vs. Curve Shape:##BR##Implied Vol Is Higher At Storage Extremes Calm Before The Storm In Oil Markets Calm Before The Storm In Oil Markets Chart 3Physical Oil Deficit Returns##BR##To Oil Market Next Year Physical Oil Deficit Returns To Oil Market Next Year Physical Oil Deficit Returns To Oil Market Next Year Highlights Energy: Overweight. The U.S. EIA revised its estimate of OPEC spare capacity down slightly for this year - to 1.7mm b/d from 1.8mm b/d. Spare capacity for next year was raised to 1.3mm b/d from just over 1mm b/d previously. At ~1.5% of global consumption this year and next, spare capacity is chronically low. Base Metals: Neutral. Chinese policymakers could sanction new infrastructure spending and easier credit to counter slower growth related to trade tensions, Reuters reported.3 Precious Metals: Neutral. We were stopped out of our tactical long silver position with a 10% loss. Ags/Softs: Underweight. There is more evidence that U.S. ags are finding new markets. EU imports of U.S. soybeans almost quadrupled in recent weeks. This comes amid the June plunge in prices and a thawing in trade tensions, following talks between EU Commission President Juncker and President Trump late last week.4 Feature The oil market sits in the eye of a pricing storm we expect to hit later this year. Following highly vocal - and twitter-textual - jawboning by U.S. President Donald Trump, OPEC's Gulf Arab producers lifted production in June and again in July.5 Reuters survey data indicate the OPEC Cartel (including new member Congo) lifted production by 70k b/d in July, bringing output to its highest level this year (32.64mm b/d).6 KSA boosted its output to 10.6mm b/d in June, up from less than 10mm b/d in the January - May period. This likely was a combination of higher production and inventory draws. OPEC's compliance level fell to 111% of the 1.2mm b/d of cuts agreed in November 2016, versus compliance levels exceeding 150% earlier this year. This is attributed to sharp declines in Venezuela's output, sporadic losses from Libya and Nigeria, and ongoing declines in non-Gulf OPEC states. We expect Russia, the putative co-head of the OPEC 2.0 coalition, will increase production by 200k b/d in 2H18 (Table 1). Table 1BCA Global Oil Supply - Demand Balances (MMb/d, Base Case Balances) Calm Before The Storm In Oil Markets Calm Before The Storm In Oil Markets Global Oil Market Will Tighten Again Post-U.S. mid-term elections in November - just when the U.S. sanctions are re-imposed against Iranian crude exports - we expect OPEC 2.0 to dial back production increases made at the behest of President Trump. Continued declines in non-Gulf OPEC output, led by ongoing and deep losses in Venezuelan output, and random unplanned production outages also will contribute to a tightening on the supply side going into 2019. Rising geopolitical tensions in the Gulf will keep markets on edge, with a predisposition to push higher. This supply-side tightness will once again come up against strong global oil demand, which we estimate will grow at a 1.7mm b/d rate this year and next. We are not expecting a repeat of the evolution of prices observed following OPEC 2.0's January 2017 agreement, which cut production to reverse the massive accumulation of inventories brought about by the original cartel's market-share war launched in November 2014. This evolution is depicted in the price-decomposition model for Brent shown in Chart 4. We segmented the fundamental price drivers - i.e. demand, supply and inventories - into distinct factors, and estimated an econometric model that allows us to track whether the evolution of prices is consistent with our expectations for these factors. Chart 4Factor Decomposition For Brent Prices Calm Before The Storm In Oil Markets Calm Before The Storm In Oil Markets Our modeling indicates the 2014 - 15 decline in oil prices was driven by a not-often-seen combination of every single factor, with our OPEC Supply-and-Inventory factor accounting for the largest negative contribution to the evolution of prices during this period. Since 2017, our factor model shows Brent prices have been supported by two factors acting simultaneously together: (1) the strong compliance of OPEC 2.0 members to the coalition's production-cutting agreement, which reduced the OPEC Supply-and-Inventory factor's role, and (2) the pickup in global oil demand, particularly in EM economies, which pushed our Global Demand factor up. These effects were partly counterbalanced by the rise in our Non-OPEC Supply factor, which became the largest negative contributor to price movements, driven by strong U.S. shale production growth. Return Of Backwardation Will Spur Volatility Our ensemble forecasts for Brent in 2H18 and 2019 are $70 and $75/bbl, with WTI expected to trade $6/bbl below these levels (Chart 5). The supply-side tightening we expect, coupled with continued demand growth, will once again lead to sharp draws in OECD inventories beginning in 4Q18 and continuing into 2019, as seen in the Chart of the Week. This will steepen the backwardations in the Brent and WTI forward curves (Chart 6). Chart 5BCA Brent And##BR##WTI Forecasts BCA Brent And WTI Forecasts BCA Brent And WTI Forecasts Chart 6Backwardation Will Return##BR##To Brent's Forward Curve Backwardation Will Return To Brent's Forward Curve Backwardation Will Return To Brent's Forward Curve Our research shows that as the slope of the Brent and WTI forward curves steepen - i.e., backwardations become more positive in percentage terms (or contangoes become more negative) - the implied volatility of options written on these crude oil futures increases, as can be seen in Chart 2.7 All else equal, higher volatility makes options written on these crude futures more valuable. Higher Vol ... Higher Prices ... In the different scenarios we use to produce our ensemble forecast, we view the balance of risks to be on the upside. This can be seen in the different paths our scenarios cover over the next year and a half, which include physical and geopolitical variables affecting price expectations (Chart 7).8 Chart 7Higher Volatility = Wider Expected Price Range Higher Volatility = Wider Expected Price Range Higher Volatility = Wider Expected Price Range Our base case assumes the supply and demand estimates shown in Table 1, which include the loss of 500k b/d due to the re-imposition of U.S. sanctions against Iran. However, we also model the loss of 1mm b/d of Iranian exports. Furthermore, we account for the loss of ~ 800k b/d of Venezuelan exports in the event that country collapses and nothing but the 250k b/d of output required to produce refined products for the local market remains online. Lastly, we account for the Permian transportation bottlenecks preventing all of the crude produced in the Basin from getting to refiners or to export markets. In this week's publication, we also include an estimate of the 95% confidence interval derived from Brent and WTI options' implied volatilities, so that our scenarios can be placed in the context of market-derived assessments of the range in which prices will trade. ... Lower Prices ... ? In modeling these risks, we also must account for downside price risks. Most prominent among these is a resolution of the long-simmering U.S. - Iran conflict, which, from time to time, results in physical confrontation. This is an outcome markets were forced to consider earlier this week when President Trump offered to meet Iranian President Rouhani without any preconditions. Among other things, Trump suggested he would have interest in working on a nuclear-arms deal to replace the one negotiated under President Obama's watch, which he scuppered in May. Secretary of State Mike Pompeo walked this remark back later. We believe the odds of such a meeting are extremely low. The odds such meeting would lead to a resolution of animosities - or at least a working understanding between the two sides - are even lower. Even so, investors need to account for this tail risk, which, if realized could take $5 to $10/bbl out of the current oil price structure. That is, until KSA and Russia muster the OPEC 2.0 member states to again reduce production to keep prices at levels that work best for their economies. Bottom Line: Our modeling and the forecasts point to higher prices and a steepening of the backwardation in Brent and WTI forward curves. This will lead to an increase in implied volatilities for options written on these crude oil futures. For this reason, we suggest investors remain long call spreads further out the Brent forward curve in 2019, which can be found in the Strategic Recommendations table on page 10 of this publication. That said, downside risks have emerged, even if, at present, the likelihood of a diplomatic breakthrough that triggers them is remote. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com Hugo Bélanger, Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com 1 OPEC 2.0 is the name we coined for the producer coalition led by the Kingdom of Saudi Arabia (KSA) and Russia. At the end of June, the coalition's member states agreed to increase production, which we estimate will raise its output ~ 275k b/d in 2H18 (vs. 1H18). We expect a physical deficit of ~ 430k b/d in 1H19 (vs 1H18, Chart 3). 2 "Contango" and "backwardation" are terms of art in commodity markets. In oil trading, when prompt-delivery crude is priced below deferred-delivery material markets are in contango; vice versa for backwardation. 3 Please see "Exclusive: China eyes infrastructure boost to cushion growth as trade war escalates - sources," published by uk.reuters.com July 27, 2018. 4 We discussed this possibility under Option 1 in our July 26, 2018, Commodity & Energy Strategy lead article entitled "Policy Uncertainty Could Trump Ag Fundamentals." It is published by BCA Research, and is available at ces.bcaresearch.com. 5 Please see our Special Report entitled "U.S., OPEC Talk Oil Prices Down; Gulf Tensions Could Become Kinetic," published jointly July 19, 2018, by BCA Research's Commodity & Energy Strategy and Geopolitical Strategy. It is available at ces.bcaresearch.com. 6 Please see "OPEC July oil output hits 2018 peak, but outages weigh: Reuters survey," published July 30, 2018, by uk.reuters.com. 7 Chart 2 shows the V-shaped mapping of implied volatility as a function of the slope of the forward curve - , i.e., the difference between the 1st- and 12th-nearby futures divided by the 1st -nearby future (to get the number in %) - against the at-the-money Implied Volatilities of 3rd-nearby Brent and WTI options (also in %). Our findings extend results published in Kogan et al (2009), who show realized volatilities calculated using historical settlements of crude oil futures have a similar V-shaped mapping with the slope of crude oil futures conditioned on 6th- vs. 3rd-nearby futures returns (in %). Please see Kogan, L., Livdan, D., & Yaron, A. (2009). "Oil Futures Prices in a Production Economy With Investment Constraints." The Journal of Finance, 64 (3), 1345-1375. Strictly speaking, volatility is the standard deviation of percent returns, usually measured on a per annum basis. Realized volatility uses futures prices to calculate returns and standard deviations; options' implied volatility is a parameter of an option-pricing model that is solved for once an option's premium, or price, is known (i.e., clears the market). This makes implied volatility a forward-looking market-cleared parameter, provided market participants agree the model used to calculate its value. Research shows implied volatilities do a better job of forecasting actual volatility than historical volatilities constructed using futures prices. See Ryan, Bob and Tancred Lidderdale (2009). "Energy Price Volatility and Forecast Uncertainty." U.S. Energy Information Administration. 8 We do not try to model a closure of the Strait of Hormuz or its prices implications. We do, however, consider this in our Special Report published July 19, 2018, "U.S., OPEC Talk Oil Prices Down; Gulf Tensions Could Become Kinetic," referenced above. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Calm Before The Storm In Oil Markets Calm Before The Storm In Oil Markets Trades Closed in 2018 Summary of Trades Closed in 2017 Calm Before The Storm In Oil Markets Calm Before The Storm In Oil Markets

Feature Downside Risks Haven't Gone Away We downgraded risk assets to neutral in last month's Quarterly Portfolio Outlook,1 citing an increasing number of risks to the equity bull market. Specifically, we warned about the slowdown and desynchronization of global growth, rising U.S. inflation, further deterioration in the trade war, and the ongoing slowdown in China. Markets - particularly in the U.S. - have stabilized somewhat over the past few weeks on the expectation that these risks are not particularly grave, that global growth remains robust, and that central banks will be slow to tighten. We accept that there remain upside risks (which is why we are neutral, not underweight, equities) but think many investors remain too sanguine about the downside risks. On desynchronized growth, it is true that the slowdown in the euro zone seems to have bottomed. The Citi Economic Surprise Indexes (Chart 1) suggest that downward surprises to euro zone and Japanese growth have ended, and that the U.S. is no longer surprising significantly to the upside. However, the likely path of inflation in the two regions looks very different, with U.S. core PCE inflation likely headed towards 2.5% over the next few quarters, while euro zone core inflation is stuck around 1% (Chart 2). Table 1Recommended Allocation Monthly Portfolio Update Monthly Portfolio Update Chart 1A Resynchronization Of Growth? A Resynchronization Of Growth? A Resynchronization Of Growth? Chart 2Core Inflation: Higher In The U.S. Than In The Euro Zone Core Inflation: Higher In The U.S. Than In The Euro Zone Core Inflation: Higher In The U.S. Than In The Euro Zone In particular, we think it is only a matter of time before U.S. wages start to accelerate. Unemployment has not been this low since the late 1960s. As happened then, there is typically a lag between the labor market becoming tight and inflation emerging (Chart 3). With the employment/population ratio for the key working-age demographic now back close to its 2007 level (Chart 4), and 601,000 new entrants to the labor force last month alone, that point is probably not far away. Note, too, that people switching jobs are now seeing large wage rises; those staying are not (Chart 5). With strong corporate profit growth, companies will soon start to raise wages to keep staff and fill vacancies. Chart 3Just A Matter Of Time Before Inflation Accelerates Just A Matter Of Time Before Inflation Accelerates Just A Matter Of Time Before Inflation Accelerates Chart 4Little Slack Left In The Labor Market Little Slack Left In The Labor Market Little Slack Left In The Labor Market Chart 5Switchers Getting Wage Rises; Stayers Not Switchers Getting Wage Rises; Stayers Not Switchers Getting Wage Rises; Stayers Not This all suggests that markets are too nonchalant about the risk of further Fed tightening. The futures market is pricing in only four rate hikes from the Fed over the next 24 months (Chart 6). We think it likely that the Fed will continue to hike by 25 basis points a quarter until something gives. By contrast, the ECB has clearly signaled that it will wait until at least September next year before raising rates; when it does so, it may hike by only 10 basis points. The futures market is close to pricing this correctly (Chart 6, panel 2). We remain concerned about further exacerbation of the retaliatory tariff war. In late July, the European Union and President Trump seemed to agree a truce, especially with regard to auto tariffs. But, even if this proves more than transitory, it is unlikely to be repeated between the U.S. and China. Both sides have raised the stakes so much that it will be politically difficult for either to back down. Further aggressive moves are likely, including a 10% tariff on all USD500 billion of Chinese imports into the U.S, and the Chinese authorities engineering a further depreciation of the Chinese yuan, and making life difficult for U.S. companies that manufacture and sell in China (where their sales total USD350 billion). Businesses around the world have woken up to this risk: capex intentions among U.S. companies have slipped recently and, in the Global ZEW survey, future expectations are now the lowest relative to current conditions since 2007, a bearish indicator (Chart 7). Chart 6Fed Is Likely To Hike more Than This Fed Is Likely To Hike more Than This Fed Is Likely To Hike more Than This Chart 7Businesses Expect Things To Get Worse Businesses Expect Things To Get Worse Businesses Expect Things To Get Worse Moreover, we don't see China launching a massive reflationary stimulus, as it did in 2009 and 2015. In the past few weeks, it has announced some minor easing of monetary policy, targeted tax cuts, and an acceleration of this year's fiscal spending. This will be enough to cushion the downside. But interest rates have not fallen anything like as much as in previous episodes (Chart 8). The authorities have reiterated that structural reform remains the priority. Given the significant slowdown in credit growth over the past year, we expect a further deceleration in the Chinese industrial economy (and, therefore, in imports) through the end of the year. If our macro outlook is correct, it is likely to have the following consequences for financial markets: further rises in long-term interest rates (we forecast 3.3-3.5% for the 10-year U.S. Treasury bond yield by early 2019), a further appreciation of the U.S. dollar (as monetary policy divergences with the euro area and Japan widen further), and negative performance for emerging market assets (hurt by higher U.S. rates, the rising USD, and the slowdown in China). This points to small negative returns from global government bonds over the next 12 months. Equities are more complicated. Earnings growth remains strong. If S&P500 companies really achieve the 20% EPS growth this year and 10% next year that analysts (and BCA's models) are forecasting, the forward multiple will fall from 16.5x now to 14.0x by end-2019. We would expect to see low single-digit positive returns from global equities over the rest of the year. We accordingly remain neutral on equities, where we can see both upside and downside risks. One key is the timing of the peak in profit margins. This has typically come a few quarters before the start of a recession. Currently margins continue to improve (Chart 9). They are likely to peak around the end of this year, when wages (and input prices, partly because of higher import tariffs) begin to rise faster than sales. We expect to move underweight equities around that time, when this and other recession indicators start to flash warning signals. Chart 8Not 2015 Redux In China Not 2015 Redux In China Not 2015 Redux In China Chart 9Watch For The Peak In Profit Margins Watch For The Peak In Profit Margins Watch For The Peak In Profit Margins Currencies: The outlook for the USD remains the key to the performance of other asset classes, particularly emerging markets and commodities. We see the risk of a short-term pullback, since long speculative positions in the dollar have recently built up (Chart 10). But differences in growth, inflation, monetary policy, and long-term rates between the U.S. and other developed economies suggest further moderate dollar appreciation over the coming 12 months. We remain very negative on EM currencies. Central banks in many emerging markets have been forced to raise rates sharply in recent weeks to defend their currencies. This is likely to slow growth over coming quarters. Those central banks that have resisted hiking (for example, Turkey and Brazil) are likely to see sharp rises in inflation. Equities: We prefer developed market equities over emerging ones. Our two overweights are the U.S. and Japan. The U.S. is a defensive market, with a beta to global equities of only 0.9 over the past 20 years. But, if there were to be a last-year equity market melt-up (along the lines of 1999), it is likely to be led by internet stocks, in which the U.S. is particularly overweight, and so the U.S. overweight also acts as a hedge against this upside risk. Our overweight in Japan is based on our view that the Bank of Japan will continue its ultra-accommodative monetary policy (bolstered by the recent tweaks to the operation of the policy), even while other DM central banks are moving towards tightening. There are also some signs of wage growth picking up, which should be positive for consumer sectors. Fixed Income: We remain underweight bonds and, within the asset class, are neutral between government bonds and spread product. U.S. junk bonds continue to have some attraction as long as economic growth remains strong (and the oil price does not fall). But junk bonds typically peak one or two quarters before equities. And, in this cycle, U.S. corporate leverage began to rise rather early, which suggests that at the start of the next recession leverage will be worryingly high (Chart 11) and that junk bonds will, therefore, perform particularly poorly. Chart 10Dollar Long Positions Building Up Again Monthly Portfolio Update Monthly Portfolio Update Chart 11Leverage Is High For This Stage Of The Cycle Leverage Is High For This Stage Of The Cycle Leverage Is High For This Stage Of The Cycle Commodities: Oil has become much harder to forecast in recent weeks, with downside risk to the price of crude coming from the recently announced OPEC production increases, but upside risk from Iran (which is threatening to close the straits of Hormuz in the face of renewed U.S. sanctions) and the collapse in Venezuelan production. BCA's energy strategists see Brent falling a little to average USD70 a barrel in 2H, and at USD75 on average next year, with greater risk of upside surprises than downside.2 Industrial metals prices are likely to remain under pressure if the USD appreciates and China slows further, as evidenced by significant downside moves in copper, iron ore and other metals over the past few weeks. Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com 1 Please see Global Asset Allocation Quarterly Portfolio Review, "Lowering Risk Assets To Neutral," dated 2 July 2018, available at gaa.bcaresearch.com 2 Please see Commodity & Energy Strategy Special Report, "U.S., OPEC Talk Oil Prices Down; Gulf Tensions Could Become Kinetic," dated 19 July 2018, available at ces.bcaresearch.com GAA Asset Allocation

Highlights Chart of the WeekTrade Fears Weighing On Ag Complex Trade Fears Weighing On Ag Complex Trade Fears Weighing On Ag Complex Bearish sentiment in ag markets is overdone. We believe prices have bottomed. But we are not yet ready to get bullish, given the elevated trade-policy uncertainty dominating markets at present. The evolution of grains and bean prices from here will depend on whether ongoing trade disputes between the U.S. and some of its largest ag markets are transitory or permanent (Chart of the Week). Highlights Energy: Overweight. We closed our Dec18 Brent $65 vs. $70/bbl call spread last week with a net gain of 80%. We remain long call spreads along the Brent forward curve in 2019, which are down an average 2.7%, and the SP GSCI, which is up 12.1%. Base Metals: Neutral. Aluminum prices are down ~ 1.6% in the past week, following indications from U.S. Treasury Secretary Steven Mnuchin sanctions against Russian aluminum supplier Rusal could be removed. Precious Metals: Neutral. Gold prices recovered slightly over the past week, but remain under pressure, given continued strength in the broad trade-weighted USD and real U.S. interest rates. We remain long gold as a portfolio hedge, nonetheless. Ags/Softs: Underweight. Fundamentals support higher grain and bean prices. However, trade-policy uncertainty - particularly re Sino - U.S. relations - will keep them under pressure (see below). Feature Weather-related uncertainty typically is center stage when it comes to forecasting ag prices during the growing season. This year, trade-policy uncertainty emanating from Washington will contend with weather risk as the dominant influence on prices. We do not expect ag-related trade policies to become more hostile. This means the path of ag prices will be contingent on whether the current trade disputes - primarily between the U.S. and China - are transient or permanent features of international trade. Given what we've seen already, we can expect American farmers will fare poorly in the ongoing trade spats. U.S. agricultural exports have been disproportionately hard hit by tariffs from their most important foreign consumer markets, levied in retaliation against U.S. tariffs (Chart 2). BCA Research's Geopolitical Strategy analysts assign a high probability to the escalation of current tensions into a full-blown trade war.1 Nevertheless, we believe the negative sentiment in ag markets is overdone, and that there is not much further downside from here. It is unsurprising that agriculture is a natural first target in this trade dispute. More than a quarter of U.S. crops are exported, with the share rising above 50% in many cases (Chart 3). This provides foreign consumers with ammunition in the dispute. Furthermore, these exports account for a large chunk of global ag trade, in some cases making American exports price makers in the global market. Importantly, many farmers and farm-belt voters cast ballots for Donald Trump. Chart 2American Ags Hit Hard##BR##By Trade Barriers... Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals Chart 3...Because They Are Exposed##BR##To Foreign Markets Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals The USDA's plans announced earlier this week to spend as much as $12 billion between September and end of harvest to help soften the impact of tariff retaliations against U.S. farm states loyal to Trump are not unexpected. The measures will entail (1) direct payments to soybean, sorghum, cotton, corn, wheat, dairy and pork farmers, (2) the procurement and subsequent re-distribution of ag products to nutrition programs, and (3) working with the private sector to promote trade and develop new export markets.2 Trade Spats Hit Grain Markets Hard Grain markets have been especially hard hit in the cross-fire between the U.S. and some of its key trade partners (Table 1). China's retaliatory tariffs are especially consequential, due to its outsized role as a main ag demand market. Table 1Ags Caught In The Crossfire Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals All in all, the Thomson Reuters Equal Weight Grains & Oilseeds Index is down ~ 10% since end-May on the back of these tariffs. Soybeans lead the decline with a 17% loss. We have been foreshadowing this since the beginning of the year.3 Now that it's played out consistent with our previous expectations, it leaves us wondering "now what?" We see three potential scenarios unfolding in the ongoing trade skirmish: Scenario 1: The current tariffs remain in place with no significant increase in ag-relevant trade barriers.4 Scenario 2: The disputes peak soon, and de-escalate. In this scenario, tariffs imposed since the beginning of the year are reversed, ultimately leading to a free and now-fairer global trade order. Scenario 3: A complete breakdown in global trade. This scenario can take on a soft outcome whereby tariffs are increased, or to a more aggressive scenario, resulting in a seismic collapse in world trade agreements. The first two scenarios are clearly more optimistic. In Scenario 1, near-term downside to prices would be restrained, contingent on the responses of major ag consumers. We discuss their four main options and potential courses of action below. Scenario 2 is the most bullish, with price formation once again a function of supply-demand-inventory fundamentals. In this scenario, exogenous risks primarily stem from weather and U.S. financial variables. However, Scenario 3, in which a prolonged trade war pushes the global economy into a recession, would intensify the pain. This would lead to a contraction in the global flow of goods and services, reducing access to foreign markets. Additionally, it would hurt ag demand through the income channel. Consumption growth of ags is correlated with income growth. If the trade war bears down on incomes, it will reduce per-capita demand for ag commodities, which ultimately depresses prices. This is especially true in the case of lower income and emerging economies, where demand is more elastic. Impact Of Tariffs In face of higher costs brought on by U.S. tariffs, foreign buyers are essentially faced with four options: Reduce imports from the U.S., and opt to purchase more from other major producers; Reduce consumption of particular crops by substituting with others; Consume out of inventory, or Continue purchasing U.S. crops, but at a higher price. Chart 4Soybean Farmers Are Most Vulnerable Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals Given the heightened risks surrounding the Sino-American trade dispute, we analyze these possibilities with reference to China. In addition, since soybeans are the most vulnerable of the crops hit by the trade dispute, we focus on beans, arguing that in most cases similar courses of action can be taken for other crops (Chart 4). Chinese authorities have already communicated that they plan to use options 1 - 3, and, as such, have assessed the impact of these restrictions on Chinese buyers to be minimal. Furthermore, according to a comment earlier this month by Lu Xiaodong, deputy general manager of state stockpile Sinograin, China is capable of fully meeting its needs without importing soybeans from the U.S.5 The extent to which buyers are successful in doing so will ultimately determine the overall impact of the trade dispute on U.S. ag markets. We expect China's solution will be a mélange of these four options. Below we assess these possibilities. Option 1: Chinese Buyers Are Turning To Other Major Producers An oft-noted change in Chinese purchasing behavior in reference to U.S. soybeans has been cited as the rationale for the negative sentiment towards U.S. ags. While it is true that Chinese buyers have been shunning American beans, the conclusion fails to recognize a few key points (Chart 5). Chart 5U.S. Soybean Exports Down On Weak Sales To China Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals First, due to the difference in crop calendars - South American beans are harvested in spring while the U.S. crop is harvested in the fall - there is a clear seasonal pattern in China's purchasing behavior (Chart 6). Thus, greater Chinese imports of Brazilian soybeans are typical for this time of year. In addition, agricultural commodities are fungible, which means a reduction of China's imports of U.S. crops does not mean the U.S. crops will go to waste. While American crops are clearly trading at a disadvantage from the perspective of a Chinese buyer, there are still other foreign markets open to American ag exports. Now that these crops are selling at a discount, they have become much more competitive, incentivizing a shift in trade flows. This has already started - the U.S. has increased exports to consumers such as Egypt and Mexico, and even found soybeans buyers in Argentina and Brazil, both major producers of soybeans (Chart 7)! Chart 6Seasonality Is Partly To Blame Seasonality Is Partly To Blame Seasonality Is Partly To Blame Chart 7New Markets Opening Up For American Beans New Markets Opening Up For American Beans New Markets Opening Up For American Beans Option 2: China Will Adjust Its Feed Recipe China's decision to remove import tariffs on animal feed ingredients from Asian suppliers also highlights another policy route. To the extent possible, Chinese consumers will attempt to find substitutes for the now-more-costly U.S. imports. This includes supplies from alternative producers, and imports of substitute products. The potential from this option depends on the availability of close substitutes to replace ags exports affected by the Sino - U.S. trade dispute. In the case of soybeans, Chinese bean imports are crushed to produce meal and oil. The former is then used as a primary protein in livestock feed, while the latter is refined to be used in foods. Similarly, the majority of corn is also used as a critical ingredient in animal feed. As such, in face of higher costs, bean crushers will likely turn to meal from other protein substitutes such as rapeseed, peanuts and sunflower seeds. Nevertheless, soybean meal remains the optimal source of protein for livestock. Thus, while China will attempt to reduce its consumption of the tariff-laden U.S. ags, alternatives are not perfect substitutes. Consequently, this option does not completely eliminate the need for soybean imports. Option 3: Eat Into Ag Inventories Chart 8Chinese Stocks Will - Partially -##BR##Cushion The Blow Chinese Stocks Will - Partially - Cushion The Blow Chinese Stocks Will - Partially - Cushion The Blow Chinese ag inventories are relatively high and can cushion the blow to supply, at least temporarily (Chart 8). This means we may see a decline in Chinese stocks, on the back of drawdowns to fill in the gap left by lower imports from the U.S. While Beijing's stocks are notoriously large, there are reports that, in some cases, they are of low quality, and are unfit for human and animal consumption. Thus, this policy may appear more feasible on paper than in reality. Without accurate information regarding the size and quality of China's ag inventories, it is impossible to determine the potential of this option. Option 4: Absorb the Price Hike: Continue Importing - Now Pricier - U.S. Ags Chinese buyers likely will attempt to exhaust options 1 - 3 above, before resorting to purchasing now-pricier U.S. grains and beans. Nevertheless, it is inevitable - some U.S. ags will continue to flow to China. The relevant question - admittedly extremely difficult to quantify - is with regards to the magnitude of the impact. This essentially will depend on China's ability to use options 1 - 3, to avoid the now-higher import costs. While in the case of soybeans, U.S. exports have been shunned for now, the true test will come in the fall after the Brazilian harvest is over, and the market is flooded with the American crops. Furthermore, the 25% increase in costs due to the tariffs will, to some extent, be offset by the discount in the price of the American crops. Fundamentals Imply Higher Ag Prices While ag markets have taken several direct hits recently, we believe global fundamentals are not as bearish as current pricing conditions suggest. In the event there is a de-escalation of trade disputes - Scenario 2 above - prices will rebound to levels implied by fundamentals. While soybeans are expected to record a small surplus in the 2018 - 19 crop year, wheat and corn will be in a global deficit (Chart 9). Furthermore, global inventories - measured in stocks-to-use terms - are expected to come down. In the case of corn and soybeans, this will be the second consecutive annual decline (Chart 10). Chart 9Bullish Fundamentals On Back##BR##Of Corn And Wheat Deficits... Bullish Fundamentals On Back Of Corn And Wheat Deficits... Bullish Fundamentals On Back Of Corn And Wheat Deficits... Chart 10...And Falling##BR##Inventories ...And Falling Inventories ...And Falling Inventories In the corn market, the inventory drawdown is , to a large extent, driven by Chinese policy which is incentivizing the consumption of stocks by offering lower subsidies to corn farmers vs. soybeans, and through measures to encourage corn use for ethanol. This is expected to bring stocks down to levels last witnessed in the 1960s! On the other hand, U.S. soybean stocks are expected to continue increasing in line with lower demand for American beans by the world's largest soybean consumer (China). As always, weather is the biggest source of near term supply-side uncertainty. Wheat prices are supported by weather concerns in Europe - particularly the Black Sea region - which is damaging crops there. This is especially important given the expectation of a smaller crop there this year. Some Final Notes A couple of distinctions within the ags space reveals some ags are more vulnerable to the ongoing dispute than others. These are the number of sellers and the number of buyers in these markets. For instance, U.S. soybean exports have fewer foreign markets than corn, making them relatively more susceptible to downward price movements as supplies back up and are forced to find alternative markets. This is especially true since China is the single largest consumer of soybeans (Chart 11). Chart 11Global Wheat Market Relatively Insulated From Trade Frictions Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals On the other hand, the global wheat market resembles a perfectly competitive market. This means that there are many buyers and sellers, each with limited ability to influence prices. Given that both the U.S. and China are price takers in this market, wheat prices will be relatively more insulated from trade headwinds. As such, we favor wheat in the current environment. Bottom Line: American farmers will be the losers in the still-evolving Sino - American trade disputes, as barriers are imposed on their exports, rendering them uncompetitive for their most significant foreign consumer. However, this will open markets for other global producers - most notably Brazil, Argentina, and the Black Sea region - making farmers there the winners in this dispute. Roukaya Ibrahim, Editor/Strategist Commodity & Energy Strategy RoukayaI@bcaresearch.com 1 Please see BCA Research's Geopolitical Strategy Special Report titled "The U.S. And China: Sizing Up The Crisis," dated July 11, 2018, available at gps.bcaresearch.com. 2 Please see "Factbox: USDA's $12 billion farmer relief package," dated July 24, 2018, available at reuters.com. 3 Please see BCA Research's Commodity & Energy Strategy Weekly Reports titled "Ags Could Get Caught In U.S. Tariff Imbroglio," dated March 15, 2018, page 9 from "Oil Price Forecast Steady, But Risks Expand," dated March 22, 2018, and "Ag Price Volatility Will Pick Up," dated May 3, 2018. 4 Our colleagues at BCA's Geopolitical Strategy team expect the trade dispute to intensify, especially before the mid-terms. However, tariffs already have been placed on most ag commodities we follow. This leaves little room for further risk from this direct channel, unless tariff rates are increased. 5 Please see "China does not need U.S. soybeans for state reserves: Sinograin official," dated June 12, 2018, available at reuters.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals Trades Closed in 2018 Summary of Trades Closed in 2017 Policy Uncertainty Could Trump Ag Fundamentals Policy Uncertainty Could Trump Ag Fundamentals