Precious Metals

Highlights Portfolio Strategy Execute a long S&P energy/short global gold miners pair trade to take advantage of the liquidity-to-growth handoff. Initiate another new trade, long S&P materials/short S&P utilities, to benefit from a shifting macro landscape. Synchronized global growth and commodity inflation are a boon for materials, but a bane for utilities. Recent Changes Initiate a long S&P energy/short global gold miners pair trade today. Initiate a long S&P materials/short utilities pair trade today. Table 1 Fraying Around The Edges? Fraying Around The Edges? Feature The S&P 500 failed to hold on to gains and drifted lower last week succumbing to Washington-related uncertainty. The transition from liquidity-to-growth remains the dominant macro theme which is prone to bouts of volatility. Nevertheless, a less hawkish Fed should, at the margin, underpin equities with easy monetary and financial conditions complementing the goldilocks equity backdrop (Chart 1). In fact, the St. Louis Fed Financial Stress Index (comprising "18 weekly data series: seven interest rate series, six yield spreads and five other indicators"1) is probing multi-decade lows. This primarily bond market-dependent indicator, has historically done an excellent job in leading the S&P 500 at major turning points at both peaks and troughs (Chart 2A). Recently, it has been more of a coincident indicator with equities, and currently waves the all-clear sign (St. Louis Fed Financial Stress Index shown inverted, Chart 2B). Chart 1Timid Fed Is Supportive Timid Fed Is Supportive Timid Fed Is Supportive Chart 2AExcellent Leading Properties Excellent Leading Properties Excellent Leading Properties Chart 2BAll Clear All Clear All Clear Nevertheless, we do not want to sound too complacent and following up from last week's brief discussion of rising geopolitical uncertainty and equity market performance, we are examining key post-WWII geopolitical events in more detail. The first three columns of Table 2, courtesy of BCA's Geopolitical Strategy Service2, update these episodes to mid-2017. While the S&P 500's drawdown from the three-month peak prior to the event to the three-month trough following the event averages out to roughly 10%, drilling beneath the surface is instructive. Table 2Geopolitical Crises And SPX Returns Fraying Around The Edges? Fraying Around The Edges? On average, broad equity market returns are muted one and three months post the event. Interestingly, on a six- and twelve-month horizon following the geopolitical incident, the S&P clearly shoots higher rising on average 5% and 8%, respectively (Table 2). Chart 3 shows the average profile of the S&P 500's returns during all of these post-WWII events, three months prior to the incident up to one year forward. Chart 3Geopolitical Opportunity? Geopolitical Opportunity? Geopolitical Opportunity? Two key takeaways stand out from this analysis. First, the coming quarter will likely prove volatile as the dust has yet to settle from the recent North Korea escalation. As a result, tactically buying some portfolio protection when the market is near all-time highs, as we cautioned last week3, is prudent and in order, especially given the seasonally challenging months of September and October. Second, on a cyclical horizon, the S&P 500 will likely resume its advance, ceteris paribus. Thus, if history at least rhymes and barring another major flare up of geopolitical risk, the path of least resistance will be higher for the overall equity market into mid-2018. This week we are executing two market neutral pair trades, one levered to the liquidity-to-growth handoff and the other to the synchronized global growth theme. Liquidity-To-Growth Handoff: Buy Energy/Sell Gold Producers A market-neutral way to benefit from the ongoing equity overshoot phase is to go long U.S. energy stocks/short global gold miners (Chart 4). This high-octane trade would benefit most from the handoff of global liquidity to economic growth. Relative share prices have plummeted since the mid-December 2016 peak, collapsing 34%. The selloff in oil prices along with a more accommodative Fed have propelled global gold miners and punished U.S. energy stocks. More recently, increasing geopolitical risks have also boosted flows into bullion and gold-related equities. However, if our thesis that growth will trump liquidity - posited three weeks ago4 - pans out in the coming months, then relative share prices should reverse. Gold prices serve as a global fear proxy, while energy prices move with the ebb and flow of global growth. Importantly, the oil/gold ratio (OGR) hit all-time lows in early 2016 and subsequently enjoyed a V-shaped recovery. But, year-to-date the OGR has relapsed on the back of rising policy uncertainty (policy uncertainty shown inverted, Chart 5). If this geopolitical uncertainty recedes, the upshot is that the OGR will rise in response. Chart 4Ready For A Bounce Ready For A Bounce Ready For A Bounce Chart 5Prefer Black Gold To Bullion Prefer Black Gold To Bullion Prefer Black Gold To Bullion Importantly, global trade is reaccelerating, also suggesting that the OGR should resume its advance (Chart 5). Chart 6 shows a simple growth/liquidity gauge using BCA's Global Synchronicity Indicator. Historically, this metric has been closely correlated with relative share price momentum, and the current message is to expect a sharp turn in oversold relative share prices. Moreover, were the liquidity thrust to convert into significantly higher output, then real interest rates should begin to reflect better growth prospects, and further boost the allure of the pair trade. As with bullion, the relative share price ratio is also overly sensitive to changes in real rates. In fact the 10-year TIPS yield does an excellent job in explaining relative share price fluctuation. Even a modest upturn in real interest rates will go a long way for relative share prices (Chart 7). Chart 6Ample Catch Up Space Ample Catch Up Space Ample Catch Up Space Chart 7Liquidity-To-Growth Beneficiary Liquidity-To-Growth Beneficiary Liquidity-To-Growth Beneficiary Meanwhile, on the relative operating front, the tide is also turning, favoring energy stocks versus gold miners. The oil and gas rig count has recovered smartly from the depths of the global manufacturing recession of late 2015/early-2016. On the flip side, demand for safe haven assets should ebb and further weigh on global gold ETF flows. Additional capital inflows into gold ETF funds from current levels would require either a sizable flare up in global geopolitical risk or another downdraft in global growth. Taken together, this relative demand indicator has surged, signaling that a catch up phase looms for the relative share price ratio (bottom panel, Chart 8). Similarly, relative pricing power is on the verge of climbing into expansionary territory. Extremely depressed pricing power for oil & gas field machinery is unlikely to deflate further, as recent anecdotes of new capital expenditure projects provide some glimmers of light for utilization rates. Conversely, bullion prices are pushing $1,300/oz. near the upper bound of the four year trading range, warning that at least a digestion phase lies ahead. The middle panel of Chart 8 shows that relative pricing power has been an excellent leading indicator of relative earnings. Our relative EPS models do an excellent job in capturing all of these different macro forces, and at the current juncture emit an unambiguously bullish signal: energy EPS will outshine gold producers' profits as the year draws to a close (Chart 9). Finally, relative valuations and technicals are both flashing a green light. Relative value is as compelling as it was during the depths of the Great Recession (middle panel, Chart 10), while our Technical Indicator is one standard deviation below the historical mean. Every time such extreme oversold levels are hit, relative performance has catapulted higher in the subsequent 3-6 months. Chart 8Relative Demand And Price Outlooks##br##Favor Energy Stocks Over Gold Miners Relative Demand And Price Outlooks Favor Energy Stocks Over Gold Miners Relative Demand And Price Outlooks Favor Energy Stocks Over Gold Miners Chart 9Earnings-Led##br## Outperformance Looms Earnings-Led Outperformance Looms Earnings-Led Outperformance Looms Chart 10Unloved ##br##And Oversold Unloved And Oversold Unloved And Oversold Bottom Line: Initiate a long S&P energy/short global gold miners pair trade to benefit from the passing of the baton from liquidity to growth. For investors seeking an alternative way to express this trade opportunity levered to the liquidity-to-growth theme, going long the S&P 1500 metals and mining index instead of the S&P energy sector would also produce similar results (bottom panel, Chart 9). New Pair Trade: Materials Vs. Utilities Macro conditions are ripe to initiate a market neutral trade: long materials/short utilities. This trade provides exposure to the budding shift in underlying portfolio strength away from defensives toward cyclicals5 and also from domestic to global-exposed market areas. In fact, our relative Cyclical Macro Indicators capture the shifting macro backdrop favoring a more cyclical portfolio tilt (Chart 11). The balance of macro evidence is skewing increasingly toward robust manufacturing growth at home and abroad. The ISM manufacturing and global PMI indexes have maintained their recent gains, signaling that the path of least resistance for the relative share price ratio is higher (Chart 12). Chart 11Reflation Trade Reflation Trade Reflation Trade Chart 12U.S. And... U.S. And... U.S. And... Reviving global growth is typically synonymous with rising inflation expectations and bond yields. BCA's view remains that a selloff in the bond markets is the most likely scenario in the coming months. The third panel of Chart 11 shows that relative share price momentum and the bond market are joined at the hip. This makes sense as materials stocks are reflationary beneficiaries, whereas the utilities sector acts as a fixed-income proxy. Not only does the pair trade benefit from rising bond yields in isolation, but also when the stock-to-bond (S/B) ratio is on fire. Currently, a wide gap has opened between the S/B and the materials/utilities ratios that will likely narrow via a catch up phase in the latter. Synchronized global growth suggests that a relative earnings-led recovery will buttress this pair trade higher. Chart 13 highlights four different ways of depicting coordinated EM and DM economic growth, giving us confidence that materials profits will outshine utilities EPS. Materials manufacturers have a sizable export component driving both the top and bottom line. In contrast, utilities are a domestic-only play. As a result, revving global trade and the significant fall in the trade-weighted U.S. dollar will buttress relative EPS prospects (Chart 14). In fact, irrespective of where the greenback ends the year, materials profits will get a lagged bump from a positive FX translation in the back half of the year. Chart 13...Global Growth Favor ##br##Materials Over Utilities ...Global Growth Favor Materials Over Utilities ...Global Growth Favor Materials Over Utilities Chart 14Cheapened Greenback = ##br##Buy Materials At The Expense Of Utilities Cheapened Greenback = Buy Materials At The Expense Of Utilities Cheapened Greenback = Buy Materials At The Expense Of Utilities The depreciating U.S. dollar is also a boon for commodity prices in general and base metals prices in particular. While natural gas prices are the marginal price setter for utilities pricing power, they represent an input feedstock cost to chemicals producers that dominate the materials sector. Taken together, a relative pricing power proxy suggests that materials stocks have the upper hand (bottom panel, Chart 14). Relative valuations and technical conditions also wave the green flag. Our valuation indicator has corrected back to the neutral zone and the technical indicator has unwound overbought conditions, offering a compelling entry point to the pair trade (Chart 15). Finally, our newly introduced relative EPS models encapsulate all of these diverging forces. Currently, the relative profit models signal that materials earnings are on track to outpace utilities profit generation for the remainder of the year (Chart 16). Chart 15Compelling Entry Point Compelling Entry Point Compelling Entry Point Chart 16Heed The Relative Profit Model Message Heed The Relative Profit Model Message Heed The Relative Profit Model Message Consequently, there is an opportunity to execute a long materials/short utilities pair trade in order to benefit from synchronized global growth and looming bond market selloff, and softening U.S. dollar and related commodity inflation. Bottom Line: Initiate a long S&P materials/short S&P utilities pair trade today. Anastasios Avgeriou, Vice President U.S. Equity Strategy & Global Alpha Sector Strategy anastasios@bcaresearch.com 1 https://www.stlouisfed.org/news-releases/st-louis-fed-financial-stress-index/stlfsi-key 2 Please see the August 16, 2017 Geopolitical Strategy Weekly Report titled "Can Pyongyang Derail The Bull Market?", available at gps.bcaresearch.com. 3 Please see the August 14, 2017 U.S. Equity Strategy Weekly Report titled "Three Risks", available at uses.bcaresearch.com. 4 Please see July 31, 2017 U.S. Equity Strategy Weekly Report titled "Growth Trumps Liquidity", available at uses.bcaresearch.com. 5 Please see the August 14, 2017 U.S. Equity Strategy Weekly Report titled "Three Risks" for a recap of our major portfolio moves since May 1, available at uses.bcaresearch.com. Current Recommendations Current Trades Size And Style Views Favor small over large caps and stay neutral growth over value.

Highlights Washington must establish a "credible threat" if it is to convince Pyongyang that negotiations offer the superior outcome; The process of establishing such a credible threat is volatile; U.S. Treasurys, along with Swiss and Japanese government bonds have been consistent safe haven assets; The risk of a U.S. attack against North Korea is a red herring, while the crisis itself is not; We suggest that investors hedge the risk with an equally-weighted basket of Swiss bonds and gold. Feature Brinkmanship between Pyongyang and Washington, D.C. has roiled markets over the past week. The uptick in rhetoric has not come as a surprise. Since last year, BCA's Geopolitical Strategy has stressed that souring Sino-American relations were the premier geopolitical risk to investors and that China's periphery, especially the Korean peninsula, would be the "decisive" factor for markets.1 North Korea's nuclear ambitions - which could be snuffed out immediately by a concerted and coordinated effort by China and the U.S. - are a derivative of the broader U.S.-China dynamic. The U.S. is unlikely to use military force to resolve its standoff with North Korea. There are long-standing constraints to war, ones that all of the interested parties know only too well from their experience in the Korean War of 1950-53. The first of these is that war is likely to bring a high death toll: Pyongyang can inflict massive civilian casualties in Seoul with a conventional artillery barrage; U.S. troops and Japanese troops and civilians would also likely suffer. Second, China is unlikely to remain neutral, given its behavior in the 1950s, its persistent strategic interest in the peninsula, and its huge increase in military strength relative to both the past and to the United States. However, the process by which the U.S. establishes a "credible threat" of military action is volatile.2 Such a credible threat is necessary if Washington is to convince Pyongyang that negotiations offer a superior outcome to the belligerent status quo. Viewed from this perspective - which is informed by game theory -President Donald Trump has not committed any grave mistakes so far, but has rather shrewdly manipulated the world's perception that he is mentally unhinged in order to enhance his negotiating leverage. It is unclear how long it will take Trump to convince North Korea that the threat of a U.S. preemptive strike is "credible." As such, it is unclear how long the current standoff will persist. From an investor perspective, it will be difficult to gauge whether the brinkmanship and military posturing are part of this "territorial threat display" or evidence of real preparations for an actual attack. As such, further volatility is likely. The ongoing crisis in North Korea is neither the first nor the last geopolitical crisis the world will face in today's era of paradigm shifts.3 We have long identified East Asia as the cauldron of investment-relevant geopolitical risks.4 This is a dynamic produced by the multipolar global context and the geopolitical disequilibrium in the Sino-American relationship. For now, investors have been able to ignore the rising global tensions (Chart 1) due to the ample liquidity emanating from central banks, but the day of reckoning is nigh (Chart 2). Chart 1Multipolarity Increases Conflict Frequency Multipolarity Increases Conflict Frequency Multipolarity Increases Conflict Frequency Chart 2Day Of Reckoning? Day Of Reckoning? Day Of Reckoning? Q&A On North Korea Back on April 19, we wrote a Special Report, "North Korea: Beyond Satire," which argued that North Korea had at last become a market-relevant geopolitical risk after decades of limited impact (Chart 3).5 Chart 3North Korean Provocations Rarely Affect Markets For Long North Korean Provocations Rarely Affect Markets For Long North Korean Provocations Rarely Affect Markets For Long Looking to the next steps, we introduced the "arc of diplomacy," a framework comparable to the U.S.-Iran nuclear negotiations from 2010-15 (Chart 4). We predicted that the U.S. would ultimately ramp up threats for the purpose of achieving a diplomatic solution. The U.S. was constrained and would only go to war if an act of war were committed, or appeared imminent.6 Chart 4Arc Of Diplomacy: Tensions Ramp Up As Nuclear Negotiations Begin Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? This assessment is now playing out. But not all clients are convinced of our logic, as we have found in our travels throughout Asia Pacific and elsewhere this month. Below we offer a short Q&A based on questions we have received from clients: Q: Diplomacy has already been tried, so why won't the U.S. attack? A: The U.S. public has less appetite for war, especially a preemptive strike, in the wake of the Iraq War, and has not suffered a 9/11 or Pearl Harbor-type catalyst. The U.S. will exhaust diplomatic options before joining a catastrophic second Korean War. And the diplomatic options are far from exhausted. The latest round of sanctions are tighter and more serious than past ones, but still leave categories untouched (like fuel supplies to the North) and are still very hard to enforce (like cutting illegal North Korean labor remittances). Enforcement is always difficult, and the U.S. is currently attempting to ensure that its allies enforce the sanctions strictly, not to mention its rivals (i.e. Russia and China). While we do not think China will ever impose crippling sanctions, we do think it can tighten them up considerably, which could be enough to change the North's behavior. Q: Why doesn't China just take North Korea out? A: China is a formal political, military, and ideological ally of North Korea, and has a strategic interest in maintaining a buffer space on the Korean peninsula - which it defended at enormous human cost in the Korean War. This interest remains in place. China is far more likely to aid and abet a nuclear-armed ally in North Korea than it is to endorse (much less participate in) regime change. The fallout from a new war, such as North Korean refugees flooding into China, is extremely undesirable for China, though it could handle the problem ruthlessly. China would also prefer not to have to occupy a collapsing North, which would be an extensive and dangerous entanglement. Therefore, expect China to twist Pyongyang's arm but not to break its legs. On a more topical note, China is consumed with domestic politics ahead of the nineteenth National Party Congress. It is perhaps more likely to take action after the congress in October-November. Q: Will U.S. allies cooperate with Trump? Why not bandwagon with China to gain economic benefit? A: South Korea is the best litmus test for whether Trump is causing U.S. allies to drift. The new South Korean President Moon Jae-In, who is politically left-of-center, has played his cards very carefully and started out on good footing with President Trump. A disagreement appears to be a likely consequence of Moon's agenda, which calls for extensive engagement with the North and a review of the U.S. THAAD missile defense deployment in Korea. So far, however, Moon is reaffirming the alliance, in his own way, and Trump has not (yet) expressed misgivings about him. If this changes significantly - as in, South Korea joining with China to give North Korea significant economic aid in defiance of U.S. sanctions efforts - then it would be a sign of division among the allies that would benefit North Korea and could even increase the risk of the U.S. taking unilateral action. The odds of that are still low, however. We have been short the Korean won versus the Thai baht since March 1, and the trade is up 6.03%. We also expect greater volatility and higher prices of credit default swaps to plague South Korea while the crisis continues over the coming months. We are closing our long Korean consumer stocks trade versus Taiwanese exporters for a loss of 4.24%. Q: What is Japan's role in the current crisis? What is the impact on Japan? A: Japan is one of the few countries whose relations with the U.S. have benefited under the Trump administration. The Japanese are in lock-step so far in reacting to North Korea. The government has been sounding louder alarms about North Korea for the past year, including by conducting evacuation drills in the case of attack. Japan has long been within range of North Korea's missiles, but its successes in nuclear miniaturization pose a much greater threat. Not only does North Korea pose a legitimate security risk, but Japanese Prime Minister Shinzo Abe also stands to benefit at least marginally in terms of popular support and support for his controversial constitutional revision. This will, in turn, feed into the region's insecurities. Yen strength as a result of the crisis, however, would be a headwind to Japan's economic growth. Thus Abe has a tightrope to walk. We expect him to take actions to ensure the economy continues to reflate. Q: Is Trump rational? How do we know he won't push the nuclear button? A: Ultimately this is unknowable. It also involves one's philosophical outlook. Josef Stalin and Mao Zedong both committed atrocities by the tens of millions but did not use nuclear weapons. Nikita Khrushchev practically wrote the playbook that North Korea's Kim dynasty has used in making its belligerent nuclear threats. Yet Khrushchev ultimately agreed to détente. Kim Jong Un makes Trump look calm. The combination of Kim and Trump is worrisome; but so was the combination of Eisenhower and Khrushchev, one believing nuclear weapons should be used if needed, the other threatening wildly to use them. It may be the case that the threat of an atrocity, or (in Kim's case) of total annihilation, is enough to keep decisions restrained. As we go to press, Kim has ostensibly suspended his plan to fire missiles around Guam and U.S. officials have repeatedly stated that they would not attack unless attacked. Stairway To (Safe) Haven Revisited In expectation of increased frequency of geopolitical risks, BCA's Geopolitical Strategy has produced two quantitative analyses of safe haven assets over the past two years. The first, "Geopolitics And Safe Havens," unequivocally crowned gold as the ultimate safe haven (Table 1), while showing that the USD is not much of a defense against geopolitical events (Chart 5).7 Table 1Safe-Haven Demand Rises During Crises Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? Table 1Safe-Haven Demand Rises During Crises Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? As such, investors should fade the narrative that the failure of the USD to appreciate amidst the latest North Korean imbroglio is a sign of some structural weakness. The greenback continues to underperform due to weak inflation in the U.S., a fleeting condition that our macro-economist colleagues expect to reverse. Mathieu Savary, BCA's currency strategist, believes that more upside exists for the USD regardless of the geopolitical outcome: Chart 5Gold Loves Geopolitical Crises Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? Chart 6DXY Is Cheap... DXY Is Cheap... DXY Is Cheap... Chart 7...But The Euro Is Not ...But The Euro Is Not ...But The Euro Is Not First, the dollar is currently trading at its deepest discount to the BCA Foreign Exchange Service augmented interest rate parity model since 2010 (Chart 6). The euro, which accounts for 58% of the DXY index, is its mirror image, being now overvalued by two sigma, the most since 2010 (Chart 7). Second, bullish euro bets will dissipate as Europe's economic outperformance versus the U.S. fades. Financial conditions have massively eased in the U.S., while they have tightened in Europe, resulting in the biggest upswing on euro area growth relative to the U.S. in over two years (Chart 8). Such an economic outperformance by the U.S. should lead to a strengthening greenback (Chart 9).8 Chart 8Easing Versus Tightening FCI Easing Versus Tightening FCI Easing Versus Tightening FCI Chart 9PMIs Point To USD Rally PMIs Point To USD Rally PMIs Point To USD Rally Our second attempt to quantify safe-haven assets, "Stairway To (Safe) Haven: Investing In Times Of Crisis," concluded that U.S. Treasurys, Swiss bonds, and Japanese bonds are the best performers in times of crisis.9 We considered 65 assets10 (Table 2) with five different methodologies and back-tested them empirically within the context of 25 financial and geopolitical events since January 1988. Some of these assets have been proven to perform as safe havens by previous academic research, some are commonly utilized in investment strategies, and others could provide alternatives (see Box 1 for further details). Table 2Scrutinizing The World For Safe Havens Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? This report demystifies four key issues related to safe havens: Part I identifies what qualifies as a safe-haven asset. Unsurprisingly, the best performers are U.S. Treasurys along with Swiss and Japanese bonds due to their currency effects. Part II examines if safe havens change over time. We find that gold and Treasurys have changed places as safe havens, and that JGBs and Swiss bonds have a long history as portfolio protectors. Part III breaks down safe havens through an event analysis. We look at the country of origin, the nature of the crisis, and whether the risk is a "black swan" or "red herring" - two classifications of events that BCA's Geopolitical Strategy has established - all of which have an impact on their performance. But red herrings or black swans are only defined after the fact, thus requiring geopolitical analysis or market timing indicators to be able to act on them. Part IV demonstrates that timing plays a crucial part when investing in safe havens as their performance is coincident with that of equities. Box 1 Safe Havens - A Literature Review In a previous Geopolitical Strategy Special Report published in November 2015, it was established that shifts in economic and political regimes alter investors' preferences for safe-haven assets, and that Swiss bonds and U.S. 10-year Treasurys were at the top of that list.11 Also, statistical methods were used to demonstrate that gold had acted as a safe haven from the 1970s to the early 90s, but has since lost its status due in part to a new era of looming deflationary risks. Li and Lucey (2013) have identified a pattern in precious metals, through a series of quarterly rolling regressions testing the significance of the 1st, 5th and 10th percentile movements in U.S. equity movements against safe-haven assets, catching extreme negative events. For instance, the 1st percentile captures the very worst corrections that have occurred, the one that represent the bottom 1% of the equity performances. The 5th and 10th percentiles represent the 5% and 10% lowest returns for equities, respectively. The authors demonstrated that silver, platinum and palladium act as safe havens when gold does not.12 Similarly, Bauer and McDermott (2013) examined the 1st, 5th and 10th percentile movements in U.S. equity movements and proved that both gold and U.S. Treasurys can serve as safe havens, but that gold has the best record in times of extreme financial stress.13 Baele et al. (2015) concentrated on flight-to-safety episodes, which they characterized as events in which the VIX, TED spreads and a basket of CHF, JPY, and USD all increased drastically.14 They found that during flight-to-safety episodes, large cap stocks outperform small caps, precious metal and gold prices (measured in dollars) increase slightly, while bond returns exceed those of the equity market by 2.5-4 percentage points. Baur and Glover (2012) provide further evidence that gold can no longer be utilized as a safe haven due to increased speculation and hedging. Their main finding is that gold cannot be both an investment and a safe-haven asset. That is, gold can only be effective as a safe haven if the periods prior to the event had not generated significant investment demand for gold.15 Using high-frequency exchange rate data, Ranaldo and Soederlind (2010) conclude that the CHF, EUR and JPY have significant safe-haven characteristics, but not the GBP.16 The strongest safe havens are identified as the CHF and JPY, but the returns are partly reversed after a day of safe-haven protection. They also find that the nature of the crisis has a significant effect on safe-haven properties. For instance, a financial crisis and a natural disaster produced drastically different outcomes for the yen. Part I - Safety In Numbers Our first step in identifying safe-haven assets was to review each asset's performance against equities in times of crisis. As such, we conducted a series of threshold regressions to generate a list of true safe-haven assets - assets that have a statistically significant positive performance in times of turmoil. Our method is explained as follows: Step 1 - Percentile Dummies: Following methods from Li and Lucey (2013) and Bauer and McDermott (2013), we created dummy variables for the 1st, 5th and 10th percentile of the S&P 500 daily total returns since 1988. We then multiplied each of these dummies by their corresponding stock returns (see Box 1 for further detail). Step 2 - Regressions: Using the 64 potential safe-haven assets, we ran a series of regressions both in USD and the local currency, testing each asset's returns explained by the three percentile dummies.17 Step 3 - Identifying Safe Havens: We then quantified strong safe-havens as assets having significant coefficients for all three return thresholds (1st, 5th and 10th percentile of the S&P 500 daily total returns). Results - Seek Refuge In Currencies And Government Bonds: Our quantitative results are mainly consistent with what others have found in the past: the Japanese yen and most G10 government bonds are safe havens. Table 3 shows the safe-haven assets that generated negative coefficients versus equities for all three threshold percentiles. Table 3Seeking Protection Against Corrections Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? In our threshold regressions expressed in USD terms, we found that the Japanese yen, Quality Stocks,18 and Japanese, Swiss and U.S. bonds acted as strong safe havens. Currencies play a crucial part in the performance of safe havens. In fact, in local-currency terms, a series of G10 government bonds (U.S., Canada, Belgium, France, Germany, Netherlands, Sweden, Switzerland, and the U.K.) proved to be the most useful safe havens. In sum, true or strong safe havens are government bonds that have currencies that add to positive returns during times of crisis. Unsurprisingly, this select group of strong safe-haven assets is comprised of U.S., Japanese, and Swiss government bonds. Quality Stocks did provide positive and statistically significant results, but the returns were very low - for this reason, we excluded them from our basket of strong safe havens. While gold, the Swiss franc, and the U.S. dollar did generate positive returns during times of crisis, they failed to generate statistically significant results at all three thresholds. Bottom Line: Based on our econometric work, most G10 government bonds can act as safe havens. But due to strong currency effects, our models favor what are already commonly known as safe havens: U.S., Japanese, and Swiss government bonds. Simply put, the difference between this select group and other G10 bonds is that their currencies rise or are stable during turmoil, while the currencies of the other G10 bonds do not. Part II - Are Safe Havens Like Fine Wines? U.S., Japanese, and Swiss government bonds were not always the top assets providing protection against the downside in equities, however. To determine whether safe-haven properties change, we examined the evolution of the relationship between safe havens and U.S. equity markets over time with the following model: Step 1 - Rolling Regressions: Considering the results obtained in Part I, we restricted our sample to G10 governments in USD and local-currency terms, Quality Stocks, gold, JPY, EUR, and USD for this statistical procedure. We put these remaining assets, both in USD and local-currency terms, through a series of 1-year rolling regressions.19 Step 2 - Identifying Trends: Each regression generated a coefficient that explained the relationship between equities and safe havens (B1). We created a new time series by collecting the coefficients for each data point and smoothing them using a five-year moving average, thus depicting a long-term pattern in the evolution of safe havens. Results - A Regime Shift In Gold And Treasurys: Our findings show that safe-haven assets fall in and out of favor through time (Charts 10A, B & C). Most striking are the changes in U.S. Treasurys and gold. Only after 2000 did Treasurys start providing a good hedge for equity corrections. The contrary is true for gold - it acted as one of the most secure investments during corrections until that time, but has since become correlated with S&P 500 total returns. That said, gold's coefficient has been falling closer to zero lately, illustrating that it could soon resurface as a proper safe haven, especially if deflation risks begin to dissipate. Given that this is precisely the conclusion stated by our colleague Peter Berezin - BCA's Chief Global Strategist - and our own political analysis, we suspect that gold may be resurrected as a safe haven very soon.20 Chart 10ASafe Havens Don't Necessarily Age Well Safe Havens Don't Necessarily Age Well Safe Havens Don't Necessarily Age Well Chart 10BSafe Havens Don't Necessarily Age Well Safe Havens Don't Necessarily Age Well Safe Havens Don't Necessarily Age Well Chart 10CSafe Havens Don't Necessarily Age Well Safe Havens Don't Necessarily Age Well Safe Havens Don't Necessarily Age Well Another important finding is that the currency effect plays a key role during recent risk-off periods (Charts 11A & B). The best protector currencies are the ones that are negatively correlated with equity returns. According to our results, the CHF and the JPY have generally been risk-off currencies, while the USD has only been one since 2007, switching places with the euro. This reinforces the case for U.S., Japanese, and Swiss government bonds, which are supported by risk-off currencies. Chart 11ACurrencies Are Difference Makers Currencies Are Difference Makers Currencies Are Difference Makers Chart 11BCurrencies Are Difference Makers Currencies Are Difference Makers Currencies Are Difference Makers Bottom Line: Safe havens change over time. Gold fell out of favor due to global deflationary dynamics. With inflation on the horizon, we will keep monitoring the relationship between gold and equities for a possible return of the yellow metal as a safe haven. Since the July 4 North Korean ICBM test, for example, gold has rallied 4.8%. Part III - Red Herrings And Black Swans Since 1988, we identified 25 economic and (geo)political events that generated instant panic or acute uncertainty in the media and financial markets.21 We analyzed the short-term reactions of the safe-haven assets, both in USD and local-currency terms. This methodology allowed for the deconstruction of the impact of the events by the following factors: Country of origin of the crisis, the nature of the crisis, and whether the event was a "red herring" or a "black swan." Generally speaking, a red herring event is a crisis of some sort with little lasting financial impact. A black swan, on the other hand, is an event that has a very low probability of occurring but has a pronounced market impact if it does. Quantitatively, our definition of a black swan is an event that produces an immediate negative response in the S&P 500 below -1%, while creating a rise in either U.S., Japanese, or Swiss government bonds above 0% (Table 4). Of course, determining which event is a red herring or a black swan is only obvious post-facto and thus requires thorough geopolitical analysis. Table 4Understanding The Crises Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? Results - Red Herrings And Black Swans Matter: Our event analysis solidifies our findings with regards to U.S., Japanese, and Swiss government bonds, but also builds a case for some European bonds as well as gold during black swan events. Our main findings can be summarized as follows. Fade The Red Herrings: Out of the sixteen geopolitical events, ten were identified as red herrings, in which safe havens underperformed the equity market. This, then, suggests that it is not always beneficial to buy safe-haven assets when tensions are rising. What is interpreted as a major geopolitical crisis - say, Ukraine in 2014 or Greece in 2015 - often ends up being a "red herring." Geopolitical Risk = Gold: Geopolitical black swan events, on the other hand, have a significant, negative impact on the market. During these events, gold emerges as the strongest hedge against a downturn in equities. U.S. Treasurys And The Swiss Franc Provide A Baseline: Under all black swan events considered - geopolitical and non-geopolitical - U.S. Treasurys and the Swiss franc had the strongest performance, generating positive returns on the day of the stock market crash in 85% of the cases. G10 Government Bonds Will Also Do: German, Dutch, Swiss and Swedish government bonds also provided protection during black swan events in local and common-currency terms, albeit to a lesser extent. U.S. And Swiss Bonds Outperform During Financial Episodes: During black swan financial crises, Swiss and U.S. government bonds stand out as the best safe havens due to their capacity to generate positive returns both in USD and local-currency terms in eight out of the nine examined crashes. Other findings that are interesting, yet less robust due to a limited sample size, include: When the crisis originated on U.S. soil, U.S. Treasurys and the dollar performed relatively poorly compared to other safe-haven assets. This is a somewhat surprising finding, as most investors believe that U.S. assets rally even at a time of U.S.-based crises, such as the 2011 budget crisis. We show that they may perform well, but in USD, non-U.S. based assets do better. When the crisis originated in Europe, European bonds performed very well both in USD and local-currency terms. When the crisis originated in Europe, Swiss and U.K. government bonds performed poorly in USD terms, but offered strong protection in local-currency terms. When the crisis originated in Russia, precious metals acted as a poor hedge. Bottom Line: It is crucial to gain an understanding of the nature of any potential crisis. Red herrings should always be faded, not hedged against, as they produce poor results in safe-haven assets. U.S. Treasurys, Swiss and Japanese government bonds have been very consistent safe-haven assets during previous periods of acute risk. Part IV: Timing Is Everything As a final step in our quantitative approach, we put our results through numerous timing exercises to test how the assets would perform in real time. Based on our Risk Asset Spectrum (Diagram 1), which summarizes our findings, one could argue that investing in times of crisis simply boils down to buying an equal-weighted basket of U.S. Treasurys, Swiss, and Japanese government bonds. Although this is technically true, such a strategy would require perfect foresight, unparalleled timing, or dumb luck - since black swan events are, by definition, very difficult to predict. Diagram 1Risk Asset Spectrum Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? Proof Of The Ultimate Safe Haven: The first experiment we conducted was to illustrate how powerful safe havens can be when timed perfectly in a trading strategy. We started off by comparing two baskets. The first was a benchmark portfolio comprised of 60% U.S. equities and 40% U.S. bonds. The other contained the same two assets, but with 100% allocated to a basket comprised of U.S. Treasurys, Swiss, and Japanese government bonds during times of negative returns for equities. Of course, this strategy is not realistic and would be impossible to implement, since the trading rule depends on future events. But as Chart 12 shows, if one were able to predict every single period of negative returns for global equities and hold safe-haven assets instead, the trading rule would outperform almost 10-fold. Chart 12Safe Havens Work Wonders With Perfect Information... Safe Havens Work Wonders With Perfect Information... Safe Havens Work Wonders With Perfect Information... One-Month Lag Is Already Too Late: Repeating the same exercise, but with a one-month lag in the execution, produces drastically different results. More specifically, whenever the previous month's equity return is negative (t=0), the portfolio allocates 100% to a single safe-haven asset for the current month (t=1), otherwise it keeps the allocation identical to that of the benchmark. The rationale for using such a simple rule is that average investors are generally late in identifying a crisis and only react once they have validation that the market is in a correction. Chart 13 shows that being late by one month changes the performance of the safe haven basket from astronomically outperforming the benchmark to underperforming it. Chart 13... But Timing Is Everything ... But Timing Is Everything ... But Timing Is Everything Reaction Is Key: As a final timing exercise, we analyzed the reaction function of our assets to see how quickly they react after the correction in equities begins (Chart 14). Unsurprisingly, the top assets that we identified start appreciating as soon as the crisis hits (t=0). Gold is, on average, the quickest asset to react from investors seeking refuge. Swiss bonds come in as a close second, almost mirroring gold during the first few days of the correction. But both assets start to flatten out and even roll over after a few days. Japanese bonds react slightly later than gold and Swiss bonds, but keep increasing for a longer period of time and start plateauing around the 30th day after the crisis. U.S. Treasurys and Quality Stocks, on the other hand, remain rather flat and constant over the short term. These results attest to the importance of timing the crisis using the best safe-haven assets. Chart 14Safe Havens React Instantly Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? Bottom Line: Timing plays a crucial part in investing in safe-haven assets, as their performance is coincident to that of equities. Investment Implications: Is Pyongyang A Red Herring Or A Black Swan? The results of our quantitative analysis are clear: hedging geopolitical risk depends on whether it is persistent or fleeting. So, is Pyongyang a red herring or a black swan? From our geopolitical analysis we make three key conclusions: The U.S. is not likely to preemptively attack North Korea; However, the U.S. has an interest in signaling that it may conduct precisely such an attack; Brinkmanship could last for a long time. Even if the risk of a U.S. attack against North Korea itself is a red herring, the crisis itself is not. In fact, between now and when a negotiated solution emerges, investors may face several new crises, which may include limited military attacks or skirmishes. While markets have faded such North Korean provocations in the past, the current context is clearly different. As such, we would suggest that investors hedge the risk with an equally-weighted basket of Swiss bonds and gold. Even though a "buy and hold" strategy with such a "Doomsday Basket" will likely underperform the market if tensions with North Korea subside, we are betting that it may take time for the U.S. and North Korea to get to the negotiating table. Marko Papic, Senior Vice President Chief Geopolitical Strategist marko@bcaresearch.com Matt Gertken, Associate Vice President Geopolitical Strategy mattg@bcaresearch.com David Boucher, Associate Vice President Quantitative Strategist davidb@bcaresearch.com 1 Please see BCA Geopolitical Strategy and Global Investment Strategy Special Report, "The Geopolitics Of Trump," dated December 6, 2016, available at gis.bcaresearch.com. We upgraded North Korea to the status of a genuine market-relevant risk in "North Korea: A Red Herring No More?" in Geopolitical Strategy Monthly Report, "Partem Mirabilis," dated April 13, 2016, available at gps.bcaresearch.com. 2 Please see BCA Geopolitical Strategy Client Note, "Trump Re-Establishes America's 'Credible Threat'," dated April 7, 2017, available at gps.bcaresearch.com. 3 Please see BCA Geopolitical Strategy Strategic Outlook, "Strategic Outlook 2017: We Are All Geopolitical Strategists Now," dated December 14, 2017 available at gps.bcaresearch.com. 4 Please see BCA Geopolitical Strategy Special Report, "Power And Politics In East Asia: Cold War 2.0," dated September 25, 2012, available at gps.bcaresearch.com. 5 Please see BCA Geopolitical Strategy Special Report, "North Korea: Beyond Satire," dated April 19, 2017, available at gps.bcaresearch.com. In particular, we argued, "the current saber-rattling is carefully orchestrated. But North Korea can no longer be consigned to the realm of satire. The very fact that the U.S. administration is adopting greater pressure tactics makes this year a heightened risk period. Investors should be especially wary of any missile tests that reveal North Korean long-range capabilities to be substantially better than is known to be the case today." Then, on May 13 and July 4, North Korea conducted its first ICBM launches; the UN Security Council agreed to a new round of even tighter economic sanctions on August 5; and the U.S. and North Korea engaged in an alarming war of words. 6 Specifically, we wrote: "Diplomacy is the only real option. And in fact it is already taking shape. The theatrics of the past few weeks mark the opening gestures. And theatrics are a crucial part of any foreign policy. The international context is looking remarkably similar to the lead-up to the new round of Iranian negotiations in 2012. The United States pounded the war drums and built up the potential for war before coordinating a large, multilateral sanctions-regime and then engaging in talks with real willingness to compromise." 7 Please see BCA Geopolitical Strategy Special Report, "Geopolitics And Safe Havens," dated November 11, 2015, available at gps.bcaresearch.com. 8 Please see BCA Foreign Exchange Strategy Weekly Report, "Fade North Korea, And Sell The Yen," dated August 11, 2017, available at fes.bcaresearch.com. 9 Please see BCA Bank Credit Analyst Special Report, "Stairway To (Safe) Haven: Investing In Times Of Crisis," dated August 25, 2016, available at bca.bcaresearch.com. 10 Forty-one assets were denominated in USD only, while G10 bonds, Credit Suisse Swiss Real Estate Fund, and European 600 real estate were used both in local-currency terms and USD, for a total of 65 assets. 11 Please see Geopolitical Strategy Special Report, "Geopolitics And Safe Havens," dated November 11, 2015, available at gps.bcaresearch.com. 12 Sile Li and Brian M. Lucey, "What precious metals act as safe havens, and when? Some U.S. evidence," Applied Economic Letters, 2013. 13 Dirk G. Bauer and Thomas K.J. McDermott, "Financial Turmoil and Safe Haven Assets," 2013. 14 Lieven Baele, Geer Bekaert, Koen Inghelbrecht and Min Wei, "Flights to Safety," National Bank of Belgium Working Paper No. 230, 2015. 15 Dirk G. Baur and Kristoffer J. Glover, "The Destruction of Safe Haven Asset?,"2012. 16 Angelo Ranaldo and Paul Soederlind, "Safe Haven Currencies," Review of Finance, Vol. 10, pp. 385-407, 2010. Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? 18 Quality stocks are defensive equity plays with high, steady earnings with an elevated return on investments. They are estimated by Deutsche Bank's Factor Index Equity Quality Excess Return in USD. Can Pyongyang Derail The Bull Market? Can Pyongyang Derail The Bull Market? 20 Please see BCA Geopolitical Strategy Monthly Report, "Throwing The Baby (Globalization) Out With The Bath Water (Deflation)," dated July 13, 2016, available at gps.bcaresearch.com, and BCA Global Investment Strategy Weekly Report, "A Secular Bottom In Inflation," dated July 28, 2017, available at gis.bcaresearch.com. 21 Since we were interested in the immediate, often unexpected, response to the event, we did not include economic recessions in our event analysis.

Highlights Strengthening income growth is apparent in DM and EM trade volumes, real wages in the U.S., and industrial commodity prices, chiefly oil and copper. This indicates inflation at the consumer level will move higher in the near future, most likely in 2H2018. We believe 10-year U.S. Treasury Inflation-Indexed securities (TIPS) trading below 0.52 do not reflect the risk of higher inflation and are, therefore, going long at tonight's close. Energy: Overweight. Crude oil prices rallied 4.6% this week, following the OPEC 2.0 meeting in St. Petersburg. Although ministers did not announce additional cuts to the 1.8mm b/d agreed at the end of last year, Saudi Energy Minister Khalid al-Falih said the Kingdom would reduce August exports to 6.6mm b/d, which is more than 300k b/d below May's level, the latest month for which data are available from JODI. Given strong global demand, if this export reduction persists - and if others join the Kingdom - it would speed the drawdown in global inventories. Base Metals: Neutral. Copper pushed through $2.80/lb on the COMEX, a level not seen since May 2015. Underlying strength in EM economic activity - seen most recently in global trading activity (discussed below) - and a weaker USD are supporting base metals. Precious Metals: Neutral. Gold fell below $1,257/oz earlier this week, and was trading ~ $1,250/oz going to press Wednesday. We remain long gold as a portfolio hedge; the position is up 1.7% since it was initiated on May 4, 2017. Ags/Softs: Underweight. Harsh weather is impacting grains. The USDA rated 62% of the U.S. corn crop in the 18 states comprising 92% of total output good or excellent last week, down from 76% in 2016. For beans, the split was 58% last week vs. 71% last year. Feature The expansion in global trade that began toward the end of last year continues, which, based on our modeling, indicates inflation at the consumer level likely will move higher in the short run (Chart of the Week). Trade expansion, particularly in EM economies, is consistent with rising incomes, which, all else equal, will keep industrial commodities - oil and copper, in particular - well supported, given income and demand for these commodities are closely aligned.1 These fundamentals dovetail with other indications of stronger growth, particularly in DM economies, where trade volumes also are growing (Chart 2). In the U.S., for example, wage growth continues to outpace inflation, and monetary conditions remain benign (Chart 3). Our colleagues at BCA Research's Global Investment Strategy believe the Fed actually may be behind the curve in reacting to nascent inflationary pressures emerging in the U.S.2 Chart of the WeekRising EM Trade Volumes Consistent##BR##With Higher U.S. CPI Inflation Rising EM Trade Volumes Consistent With Higher U.S. CPI Inflation Rising EM Trade Volumes Consistent With Higher U.S. CPI Inflation Chart 2DM Trade Volumes Are Expanding##BR##At ~ 5% Pace ... DM Trade Volumes Are Expanding At ~ 5% Pace ... DM Trade Volumes Are Expanding At ~ 5% Pace ... Chart 3U.S. Labor Market Tightening,##BR##Financial Conditions Remain Loose U.S. Labor Market Tightening, Financial Conditions Remain Loose U.S. Labor Market Tightening, Financial Conditions Remain Loose Trade Growth Supports Higher Inflation U.S. CPI is highly correlated with EM trade volumes (imports and exports) as shown in the Chart of the Week. In recent research into inflation and trade, we also showed EM oil demand and world base metals demand are highly correlated with EM trade volumes.3 Chart 4EM Trade Volumes##BR##Continue To Strengthen Growth EM Trade Volumes Continue To Strengthen Growth EM Trade Volumes Continue To Strengthen Growth EM import growth continues to expand at a faster pace than DM growth (Chart 4). Year-on-year (yoy) EM import growth came in at 7.7%, a full 2 percentage points above DM growth. This is not to minimize DM growth - it finally broke out of its lethargy in May with a sharp advance of close to 6%, which will lift the trend rate of growth (the 12-month moving average, or 12mma) higher going forward. EM export growth in May was only slightly above DM growth for the month - 5.4% yoy vs. 5.2% yoy. These stout monthly trade performances will, in the next few months, offset the lethargic growth seen in EM and DM prior to the expansion begun at the end of 2016, as weaker monthly performance falls off the trend calculations. Over the year ended in May, within EM markets the annual trend in imports (the 12mma to May 2017) has barely grown more than 1% yoy, dragged down by a 6% contraction in the Middle East and Africa (MEA) and a 2.1% contraction in Latin American growth. The trend in EM - Asia's imports is up, rising 3.2% over the same period. For the year ended in May, imports into central and Eastern Europe were mostly flat; however, since November 2016, the trend turned sharply positive with 3.3% yoy growth. The trend in export volumes is expanding for in MEA and Latin America economies - 3.5% yoy trend growth (12mma) in MEA, and 4.4% growth in Latin America, which is slightly higher than the overall 2.2% rate of trend growth in EM exports. Still, lower oil and commodity prices, along with reduced volumes are curtailing an income recovery in these regions. Central and Eastern Europe's rate of export expansion leads EM generally at close to 4% yoy trend growth. Favor Gold And TIPS Ahead Of Higher Inflation As the labor market tightens and real-wage growth continues to outpace productivity growth, we expect U.S. inflation to pick up. Growth in trade volumes also will support growth in EM oil demand and world base metal demand, as noted above. This will feed into U.S. core PCE, the Fed's preferred inflation gauge (Chart 5). As we've highlighted in the past, there is very strong co-movement among these variables: We've found that, all else equal, a 1% increase in the non-OECD oil demand implies an increase in the core PCE of slightly less than 50bp. If the trend in overall EM trade volumes persists, the likelihood we will be increasing our estimate of non-OECD oil consumption for 2H17 and 2018 increases. U.S. CPI and EM trade volumes show similar co-movement properties, as the Chart of the Week shows. A 1% increase in EM import volumes translates into a 0.53% increase in the U.S. CPI, while a 1% increase in EM export volumes implies a 0.49% increase in the CPI. EM import volumes over the January - May 2017 interval have been growing at slightly more than 8% yoy, while exports have been growing at slightly more than 3%. Continued strength in the EM trade data implies U.S. CPI could grow well above what's currently being priced in inflation markets and by Fed policymakers. This leads us to favour gold and TIPS as inflation hedges. If we do get a larger-than-expected move in the U.S. CPI, gold should respond well. The modelling depicted in Chart 6 shows a 1% increase in the CPI translates into a 4.1% increase in gold. Chart 5Core PCE Will Pick Up##BR##As Commodity Demand Grows Core PCE Will Pick Up As Commodity Demand Grows Core PCE Will Pick Up As Commodity Demand Grows Chart 6Gold Will Pick Up##BR##Larger-Than-Expected CPI Moves Gold Will Pick Up Larger-Than-Expected CPI Moves Gold Will Pick Up Larger-Than-Expected CPI Moves For this reason we recommend getting long U.S. Treasury Inflation-Protected Securities (TIPS), which will appreciate as the U.S. CPI moves higher.4 We will be getting long as of tonight's close. We remain long low-risk calls spreads in Dec/17 WTI and Brent - long $50/bbl strikes vs. short $55/bbl strikes. We are up 39.3% and 32.9% on the Brent and WTI positions, respectively, from last week, and 47.2% and 89.2% since inception. U.S. Monetary Policy Remains A Huge Risk To EM Trade As we've noted in the past, U.S. monetary policy can have an outsized effect on EM trade volumes. In an update of an earlier model using U.S. M2 and the broad trade-weighted USD (TWIB), we find a 1% increase in the broad trade-weighted USD translates into a 1.1% drop in EM imports, while a 1% increase in U.S. M2 (broad money) implies an 85bp increase in EM imports (Chart 7).5 Chart 7EM Trade Volumes Highly Sensitive##BR##To U.S. Monetary Policy EM Trade Volumes Highly Sensitive To U.S. Monetary Policy EM Trade Volumes Highly Sensitive To U.S. Monetary Policy This demonstrates the feedback loop we've identified between U.S. monetary policy and EM trade. EM trade volumes affect inflation at a global level. We've found inflation in the U.S., EU and China to be co-integrated - i.e., these price gauges all follow the same long-term trend. Inflation and inflation expectations drive Fed policy, which drives the price formation of the USD - i.e., the FX rates included in the USD TWIB - and affect Fed policy on M2. These U.S. monetary variables, in turn, affect EM trade volumes. And so it goes ... Too-aggressive a tightening by the Fed as it normalizes its interest-rate policy regime could destabilize EM economies - either via too-sharp an appreciation in the USD TWIB, a larger-than-expected deceleration in M2 growth, or both - and negatively affect trade flows. At the end of the day, this would redound to the detriment of the U.S. economy, as the different feedback mechanisms kick in. This says the Fed's policy doesn't just affect the U.S. economy, or that EM economies essentially are on their own in the policy tools they deploy to adjust to Fed innovations. Like it or not, the Fed has to consider these types of feedback loops in its decision-making, since the Open Market Committee will be dealing with the fallout of its earlier policies. Bottom Line: EM trade volumes continue to grow yoy, continuing the trend that began at the end of last year. This performance, coupled with a tightening labor market in the U.S. and a still-loose financial backdrop, raises the odds inflation will exceed what's currently priced into market and Fed expectations. We are getting long U.S. 10-year TIPS at tonight's close, and remain long gold as a strategic portfolio hedge. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com 1 The income elasticity for industrial commodities in EM economies is ~ 1.0, according to the OECD. Please see "The Price of Oil - Will It Start Rising Again?" OECD Economics Department Working Paper No. 1031, p. 6 (2013). 2 Please see BCA's Global Investment Strategy Weekly Report titled "Are Central Banks Behind The Curve Or Ahead Of It?," published on July 21, 2017. It is available at gis.bcaresearch.com. Among other things, the Global Investment Strategy team notes labor-market slack is dissipating, real wages are increasing, and easier financial conditions are spurring credit growth. Our colleagues note, "The prospect of stronger growth over the next few quarters implies that the unemployment rate is likely to fall below 4% early next year, possibly breaking through the 2000 low of 3.8%." BCA's Global Investment Strategy believes U.S. inflation could move higher by 2H18. 3 Please see BCA Commodity & Energy Strategy Weekly Reports titled "EM Trade Volumes Continue Trending Higher, Supporting Metals" and "Strong EM Trade Volumes Will Support Oil," published June 29, and June 8, 2017. Both are available at ces.bcaresearch.com. 4 U.S. TIPS increase in value as the Consumer Price Index (CPI) rises, and fall in value as the index declines. Please see "TIPS: Rates & Terms" on the UST's TreasuryDirect web page (https://www.treasurydirect.gov/indiv/research/indepth/tips/res_tips_rates.htm). 5 This model covers 2000 to the present, using monthly data. The R2 for the cointegrating regression is 0.96. These variables do not explain EM exports, which are not cointegrated with U.S. monetary variables. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Trade And Commodity Data Point To Higher Inflation Trade And Commodity Data Point To Higher Inflation Commodity Prices and Plays Reference Table Trade And Commodity Data Point To Higher Inflation Trade And Commodity Data Point To Higher Inflation Trades Closed in 2017 Summary of Trades Closed in 2016

Highlights Expanding trade volumes - led by EM growth - will continue to support commodity demand, particularly for base metals. In the first four months of this year, EM import growth averaged 8.4% year-on-year (yoy), led by an expansion of almost 13% in EM Asia. This compares with yoy growth averaging just 0.3% in the DM imports over the January - April period. EM exports grew 5.1% yoy in this interval vs. 2.7% for DM outbound trade. Overall, EM growth led world trade volumes 4% higher yoy, versus 0.8% growth over the January - April interval last year. We expect trade volumes to continue to grow as long as the Fed doesn't tighten monetary conditions too much in the U.S. Energy: Overweight. Benchmark crude oil prices continue their lackluster performance, as high-frequency inventory data in the U.S. fail to convince markets OPEC 2.0 cuts are succeeding in draining global inventories. We expect this to reverse, and remain long Dec/17 $50 vs. $55/bbl call spreads in WTI and Brent. Base Metals: Neutral. Expanded global trade, led by EM Asia, will be supportive of base metals prices. However, we do not expect higher trade volumes to prompt a surge in base metals. We remain neutral base metals generally. Precious Metals: Neutral. Palladium has consistently outperformed platinum in post-GFC markets, as has gold (see below). We remain neutral the precious-metals complex, but are keeping our long gold portfolio hedge in place. Ags/Softs: Underweight. The USDA's crop report will be released Friday. Weather-related crop distress in the grains could start showing up in the data. We remain bearish, but recommend staying on the sidelines. Feature Chart of the WeekStrong Growth In Global Trade Volumes##BR##Will Be Supportive Of Base Metals EM Trade Volumes Continue Trending Higher, Supporting Metals EM Trade Volumes Continue Trending Higher, Supporting Metals Growth in EM imports and exports continues to lead the expansion of global trade volumes. This is important, as the growth in trade supports EM income growth, which, in turn, supports commodity demand. EM growth is the principal source of commodity demand growth globally, particularly for oil and base metals. Global trade volumes expanded yoy in April, with imports up 3.7% and exports up 3.2%, down slightly from the pace in 1Q17, according to the CPB World Trade Monitor (Chart of the Week). The notional value of trade for the year ended in April was $16.3 trillion. The uptrend in global trade begun in 4Q16 continues, however, which we noted earlier this month. As was the case with oil, this expansion of global trade volumes, particularly out of the EM economies, will be supportive of base metals demand generally.1 Similar to EM oil demand, we find EM exports and imports are highly correlated with world base metals demand post-GFC (Chart 2). This is not unexpected, given the prominence of Chinese base metals demand, which accounts for roughly half of base metals demand globally. Given the low level of growth in DM imports and exports, we conclude that the bulk of the increase in global trading volumes is increasingly accounted for by trade within the EM economies with each other. This can be seen in Chart 3, which shows growth of EM imports and exports surpasses DM trade performance, shown in Chart 4. Chart 2World Base Metals Demand,##BR##Highly Correlated With EM Trade Volumes World Base Metals Demand, Highly Correlated With EM Trade Volumes World Base Metals Demand, Highly Correlated With EM Trade Volumes Chart 3Increased Trade Within##BR##EM Economies Powers Global Trade Growth Increased Trade Within EM Economies Powers Global Trade Growth Increased Trade Within EM Economies Powers Global Trade Growth Chart 4DM Growth Not##BR##Keeping Up With EM Growth DM Growth Not Keeping Up With EM Growth DM Growth Not Keeping Up With EM Growth Going Through The Trade Numbers For the year ended in April 2017, yoy world import levels grew 2.5% on average each month. DM imports averaged 1.8% yoy growth, while EM imports grew at twice that level. The notional value of DM imports was $9.6 trillion for the year ended in April. EM notional imports were $6.7 trillion, with EM Asia accounting for $4.7 trillion of this. For exports, world trade volumes grew at an average rate of 2.3% yoy each month for the 12 months ending in April, with DM growth coming in at 1.6%, and EM growth clocking in at 3.1% on average, just shy of double the rate of DM growth. The notional value of DM exports was $8.8 trillion for the year ended in April. EM notional exports were $7.4 trillion, with EM Asia responsible for $4.9 trillion of this. For April alone, DM imports were up 1.8% yoy, while EM imports were up 6.5%, down from a 9.1% average rate in 1Q17. DM exports in April were up 1.7% yoy, down from a 3% average rate in 1Q17, while EM exports rose 5%, equal to the 1Q17 average rate. Import volumes for EM Asia led global growth by a wide margin vs. other EM markets, particularly in Latin America and the Middle East (Chart 5). Average yoy import growth for the year ended in April was 6.6% for EM Asia, with yoy growth for April alone registering 10.4%. EM Asia also leads export volume growth (Chart 6), with average yoy outbound trade up 7.1% yoy. Chart 5EM Asia Dominates Import Growth YoY... EM Asia Dominates Import Growth YoY... EM Asia Dominates Import Growth YoY... Chart 6...And EM Export Growth ...And EM Export Growth ...And EM Export Growth As always, the evolution of China's economy will have an outsized influence on trade and EM growth. We continue to expect China's growth to moderate but not slow sharply, in line with our colleagues at our sister publication China Investment Strategy. The recent credit tightening likely will abate, given the central bank has reversed its credit contraction and injected liquidity into the interbank system in recent weeks, according to BCA's China Investment Strategy service.2 We agree with China Investment Strategy's assessment that, "The Chinese economy will likely continue to moderate, but the downside risk appears low at the moment and overall business activity will remain buoyant."3 Update On Global Inflation Vs. EM Trade Volumes World base metals demand is highly correlated with global consumer price inflation. In fact, these variables are cointegrated, meaning they share a common long-term trend. A 1% increase in world base metals demand can be expected to produce a 0.32% increase in U.S. CPI, a 0.25% increase in the Euro Area Harmonized CPI, and a 0.43% increase in China's CPI. Like the relationships between EM oil demand and EM trade volumes, which we presented earlier this month, the relationships between world base metal demand and EM trade volumes also allows us to track EM income levels. This is because the income elasticity of base metals demand also is ~ 1.0 for EM economies, according to the OECD, meaning a 1% increase in EM income can be expected to produce an increase in base metals demand of ~ 1%.4 Likewise, consumer inflation worldwide also is highly correlated with EM trade volumes post-GFC.5 In the regressions we ran for U.S., Euro Area and China CPI as a function of EM trade volumes, we find a 1% increase in EM imports can be expected to produce a 0.51% increase in the U.S. CPI, a 0.41% increase in the Euro Area harmonized CPI, and a 0.67% increase in the China CPI. A 1% increase in EM exports produces increases of 0.47%, 0.35% and 0.65%, respectively. These relationships can be seen in Charts 7, 8, and 9. Chart 7U.S. CPI Highly Correlated##BR##With EM Trade Volumes... U.S. CPI Highly Correlated With EM Trade Volumes... U.S. CPI Highly Correlated With EM Trade Volumes... Chart 8...Along With The Euro Area##BR##Harmonized CPI... ...Along With The EMU Harmonized CPI... ...Along With The EMU Harmonized CPI... Chart 9...And##BR##China's CPI ...And China's CPI ...And China's CPI Bottom Line: World base metals demand will continue to be supported by continued growth in EM trade volumes this year. While these volumes are up nicely, the rate of growth is moderating somewhat, suggesting global base metals demand will hold up this year, but won't surge ahead. We certainly do not see base metals prices falling precipitously this year, given the growth in EM imports and exports, which started to revive toward the end of last year. We remain neutral base metals, but will be watching the interplay between base metals demand and EM trade volumes for any sign global demand is being re-ignited. Inflation appears to be quiescent globally, but we would expect it to start ticking up if we see an uptick in base metals demand and EM trade volumes. Precious Metals Update PGM Notes Price relationships within the precious-metals complex, particularly vis-à-vis Platinum Group Metals (PGMs), have undergone profound transformations since the end of the Global Financial Crisis (GFC). These changes have been bolstered by technology shifts in the automotive sector as well. Two trading relationships - palladium's relationship to platinum, and platinum's relationship to gold - best illustrate these changing fundamentals. Unlike gold and platinum prices, palladium is heavily influenced by automotive sales, in this case, gasoline-powered automobile sales. Gasoline-powered cars use palladium in their pollution-control catalysts. Such usage was up 5% last year to 7.4mm oz, according to Thomson Reuters GMFS data.6 Autocatalyst demand accounts for slightly more than three-quarters of palladium demand, according to GFMS data.7 Importantly, the two largest car markets in the world - the U.S. and China - are predominantly gasoline-powered, and sales in both have been strong, although the rate of growth has slowed (Chart 10). This is supportive of palladium prices, particularly as the metal registered a 1.2mm physical deficit last year. There is an increase in Chinese platinum-based auto catalyst demand for diesel cars, due to tightening regulation on emissions, but still is a small share of the total demand for platinum. Post-GFC, the value of palladium relative to platinum has consistently strengthened (Chart 11). Chart 10U.S. Sales Growth Down,##BR##But China Remains Strong U.S. Sales Growth Down, But China Remains Strong U.S. Sales Growth Down, But China Remains Strong Chart 11Platinum Eclipsed By##BR##Palladium And Gold Platinum Eclipsed By Palladium And Gold Platinum Eclipsed By Palladium And Gold Platinum's Discount To Gold Endures Platinum traded premium to gold until the GFC (Chart 11). Since then, gold has behaved more like a currency, with its price mostly dependent on financial variables (USD, real U.S. interest rates, and equity risk premium). Importantly, the yellow metal has traded premium to platinum since the GFC ended. While platinum prices are somewhat sensitive to the same financial factors as gold, and also can be modeled as a function of these financial variables, the metal also has a real-demand driver: diesel-powered car sales. These vehicles use platinum in their pollution-control catalysts. Autocatalysts accounted for 3.3mm oz of platinum sales last year, or 42% of demand, according to Thomson Reuters GFMS. Most of this goes to diesel catalysts, which are mainly sold in Europe. Diesel-powered car sales have been trending lower (Chart 12) in Europe, where they are the dominant type of car sold. The second largest demand segment for platinum is jewelry sales, which fell 12% last year to 2.2mm oz, following a 3.6% decline the prior year. Both gold and platinum will be responsive to the same set of financial variables, meaning a stronger USD along with higher real rates will be bearish for both, and vice versa. However, given platinum is also sensitive to the diesel-powered auto market, its price evolution has a component strongly influenced by physical platinum demand and supply. Supply comes from mines and recycling, which increases when steel prices rise. Sales of diesel-powered cars are falling in Europe partly due to the Volkswagen emissions-testing scandal and a longer-lasting trend of cities attempting to lower pollution by restricting where diesel-powered vehicles can drive (e.g., Athens, Madrid, Paris and Mexico City are eliminating diesel traffic by 2025).8 In addition, high steel prices will increase platinum recycling volumes this year (people scrap their cars more when steel prices are high). High steel prices also incentivize the scraping of gasoline-powered vehicles, which use palladium in their pollution-control catalysts. Platinum competes at the margin in the pollution-control catalyst market with palladium. The ratio between palladium and platinum is at its highest level since 2002 (Chart 11). The premium of platinum against palladium (platinum minus palladium) went from $1200/oz. in 2010 to close to parity recently (Chart 13). Chart 12EU Sales Still Growing,##BR##But Diesel Loses Share EU Sales Still Growing, But Diesel Loses Share EU Sales Still Growing, But Diesel Loses Share Chart 13Platinum's Premium To##BR##Palladium Disappears Platinum's Premium To Palladium Disappears Platinum's Premium To Palladium Disappears Continue To Favor Palladium We are more favorably disposed toward palladium than platinum. Given palladium's price is dominated by sales of gasoline-powered cars, which should, all else equal, do well relative to diesel-powered auto sales, even with a globally synchronized economic upturn. With the U.S. Fed expected to continue tightening, gold and platinum will face financial headwinds that will restrain price appreciation. Palladium, on the other hand, will be less sensitive to these headwinds, although higher interest rates in the U.S. relative to the rest of the world will restrain demand for goods purchased on credit like autos. While we remain neutral the precious metals complex generally, we recently recommended a long spot-gold position as a portfolio hedge against rising inflation and inflation expectations.9 Even though inflation has remained quiescent, and markets are trading as if the odds of a return of inflation are extremely low, BCA's Global Investment Strategy argues "the combination of faster growth and dwindling spare capacity will cause inflation to rise. This is particularly the case for the U.S., where the economy has already reached full employment."10 We believe the strengthening of household incomes resulting from the tight U.S. labor market likely will keep the Fed on track to continue with its rates-normalization policy, vs. market expectations of a mere 21 basis points in cumulative Fed rate hikes over the next 12 months. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com 1 Please see BCA Research's Commodity & Energy Strategy Weekly Report "Strong EM Trade Volumes Will Support Oil," published June 8, 2017, available at ces.bcaresearch.com. The CPB World Trade Monitor is published monthly by the CPB Netherlands Bureau for Economic Policy Analysis. Please see https://www.cpb.nl/en/worldtrademonitor for data and documentation. We use CPB's volumetric data for imports and exports in our analysis, which are indexed to 2010 = 100; we converted these data to USD values to see how the composition of imports and exports is changing so as to better see how the relative shares of EM and DM are evolving. 2 Please see BCA Research's China Investment Strategy Weekly Report "Chinese Financial Tightening: Passing The Phase Of Maximum Strength," published by on June 22, 2017, available at cis.bcaresearch.com. 3 Please see BCA Research's China Investment Strategy Weekly Report "A Chinese Slowdown: How Much Downside," published on June 8, 2017, available at cis.bcaresearch.com. 4 As we noted in our research earlier this month, the read-through on this is EM trade volumes are closely tied to income levels, given this income elasticity in non-OECD economies. Please see "The Price of Oil - Will It Start Rising Again?" OECD Economics Department Working Paper No. 1031, p. 6 (2013). In our modeling, we assume the GFC ended in 2010. Our oil results vis-à-vis EM income elasticities can be found in BCA Research's Commodity & Energy Strategy Weekly Report "Strong EM Trade Volumes Will Support Oil," published June 8, 2017, available at ces.bcaresearch.com. 5 We originally published these results for EM oil demand vs. EM trade volumes in the June 8, 2017 article referenced above in footnote 1, in BCA Research's Commodity & Energy Strategy Weekly Report "Strong EM Trade Volumes Will Support Oil," available at ces.bcaresearch.com. The average R2 coefficient of determination for the regressions on imports was 0.89, while the average for the regressions on exports was 0.89. 6 Palladium supply totalled 8.6mm oz, while demand came in at 9.8mm oz, according to the Thomson Reuters - GFMS Platinum Group Metals Survey 2017. 7 In our modelling, we treat palladium as an industrial metal, given the overwhelming influence auto demand - particularly gasoline-powered vehicles - has on its price. Please see BCA Research's Commodity & Energy Strategy Weekly Report "2016 Commodity Outlook: Precious Metals," published by December 3, 2015, available at ces.bcaresearch.com. 8 A number of cities are looking to ban diesel cars entirely from their central districts. Please see https://www.theguardian.com/environment/2016/dec/02/four-of-worlds-biggest-cities-to-ban-diesel-cars-from-their-centres. 9 Please see BCA Research's Commodity & Energy Strategy Weekly Report "Go Long Gold As A Strategic Portfolio Hedge," published May 4, 2017, available at ces.bcaresearch.com. 10 Please see BCA Research's Global Investment Strategy Weekly Report "Stocks Are From Mars, Bonds Are From Venus?," published June 23, 2017, available at gis.bcaresearch.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed In 2017 Summary of Trades Closed in 2016 EM Trade Volumes Continue Trending Higher, Supporting Metals EM Trade Volumes Continue Trending Higher, Supporting Metals

Highlights Business capital spending is still trending up, adding another layer of support for the U.S. economy in the next 6-12 months. Profit growth has accelerated at a faster pace than our top-down model had projected and we expect growth to accelerate further into year end. We estimate that the delayed pass-through of previous dollar strength will remain a slight drag on U.S. EPS growth over rest of 2017. Our tactical view on gold remains bearish, but the BCA Commodity & Energy Strategy service sees strategic value in gold as a hedge. Feature The S&P 500 is attempting to break through the 2400 barrier as we go to press. This is impressive given that the flagging relative performance of infrastructure stocks and highly-taxed companies suggests that investors have given up hope of ever seeing significant tax cuts, infrastructure spending and incentives for capital spending. As we discuss below, disappointment on the policy front has thankfully been offset by solid corporate earnings figures. We believe that investors have gone too far in pricing out tax reform. True, the growing number of White House scandals will serve to delay the GOP's market-friendly policy agenda. Nonetheless, the President desperately needs a win ahead of mid-term elections, and tax reform and deregulation are two key areas where the President and congressional Republicans are on the same page. Capital spending is the part of the economy that could benefit the most from tax reform. Surprising Support From Capex Business capital spending is still trending up, adding another layer of support for the U.S. economy in the next 6-12 months. The post-election rollover in C&I loan growth worried investors that rising rates and election-related uncertainty had cut the flow of credit to the business sector, thus putting capex at risk (Chart 1, top panel). That concern was overdone, as we pointed out in a recent report.1 Business expenditures on plant, equipment and software were a surprising source of strength in first-quarter GDP, and bank lending has stabilized in the past six weeks. The FOMC minutes of the May 2-3 meeting noted that "financing conditions for large nonfinancial firms stayed accommodative." The minutes also stated that, while there was weaker demand for C&I loans in April, the weakness "pertained to customers' reduced needs for financing." The reduced need likely reflected a preference to issue corporate bonds. Chart 1Outlook For Capex Looks Solid Outlook For Capex Looks Solid Outlook For Capex Looks Solid Our BCA Capex indicator for business investment points to solid business spending in the next few quarters. (Chart 1, bottom panel) Our past research shows that sustainable capital spending cycles only get underway when businesses see evidence that consumer final demand is on the upswing. While consumer expenditures were quite soft (+0.3% annualized gain) in Q1, our view is that the weakness was transitory.2 This view was confirmed by the FOMC minutes. A rebound in consumer spending in the second quarter will boost CEO confidence that increased capital spending will be justified in terms of future sales. Our base case is that at least some tax cuts will be enacted by year end, but the risk is that political turmoil further delays a fiscal package or even totally derails the GOP legislative agenda. This scenario would be negative for stocks temporarily, but could end up being positive over the medium term by extending the expansion in the economy and corporate profits. U.S. Profits, Beats And Misses Profit growth has accelerated at a faster pace than our top-down model had projected earlier this year (Chart 2). On a 4-quarter moving total basis, S&P 500 earnings-per-share were up by more than 13% in the first quarter (84% reporting). We expect growth to accelerate further into year end, peaking at just under 20%, before moderating in 2018. The favorable profit picture reflects two key factors. First, profits are rebounding from a poor showing in 2016, when EPS was dragged down by the collapse in oil prices and a global manufacturing recession. Oil prices have since rebounded and global industrial production is recovering as expected (Chart 3). Earnings are of course leveraged to corporate sales, helping to explain why profits are highly correlated with industrial production in the major countries. BCA's U.S. Equity Strategy service estimates that operating leverage for the S&P 500 is 1.4x.3 Chart 2Impact Of Stronger Dollar Is Fading Impact Of Stronger Dollar Is Fading Impact Of Stronger Dollar Is Fading Chart 3IP On The Rebound Globally IP On The Rebound Globally IP On The Rebound Globally Second, the corporate sectors in the major economies are still in a sweet spot in which the top line is growing but there is no major wage cost pressure evident yet. This is the case even in the U.S., where labor market slack has largely been absorbed. Indeed, margins rose in Q1 2017 for the third quarter in a row. Our indicators suggest that the corporate sector has gained some pricing power at a time when wage gains are taking a breather.4 The hiatus of wage pressure may not last long, but for now our short-term EPS growth model remains upbeat for the next 3-6 months (not shown). What About The Dollar? We estimate that the delayed pass-through of previous dollar strength will remain a slight drag on U.S. EPS growth of about one percentage point for the remainder of this year, assuming no change in the dollar from today's level (Chart 2, second and third panels). However, our base case remains that the dollar will appreciate by another 10% in trade-weighted terms. A 10% appreciation would trim EPS growth by roughly 2½ percentage points, although most of this would occur in 2018 due to lagged effects. The key point is that another upleg in the dollar, on its own, should not provide a major headwind for the stock market. Indeed, the dollar would only be rising in the context of robust U.S. economic growth and an expanding corporate top line. Even though the message from our EPS model is upbeat, it still falls short of bottom-up estimates for 2017. Is this a risk for the equity market, especially since valuations are stretched? Investors are well aware that bottom-up estimates are perennially optimistic. Table 1 compares the beginning-of-year EPS growth estimate with the actual end-of-year outcome for 2007-2016. Not surprisingly, bottom-up analysts massively missed the mark in 2008, which was a recession year. But even outside of the recession, analysts significantly over-estimated earnings in seven out of nine years. Despite this, the S&P 500 rose sharply in most cases. One exception was 2015, when the S&P 500 fell by 0.7%. Plunging oil and material prices contributed to an EPS growth "miss" of seven percentage points. Chart 4 highlights that the level of the 12-month forward EPS estimate fell that year, unlike in the other years considered. Valuations are more demanding today than in the past, but the message is that attaining bottom-up EPS year-end estimates is less important for the broad market than the direction of 12-month forward estimates (which remains up at the moment). Table 1Bottom Up Estimates Are##BR##Always Too Optimistic Corporate Earnings Versus Trump Turbulence Corporate Earnings Versus Trump Turbulence Chart 4Oil Related##BR##Dip In 2015 Oil Related Dip In 2015 Oil Related Dip In 2015 The bottom line is that the backdrop is constructive for equities even if the Republicans are unable to push through any fiscal stimulus. In fact, it may be better for the stock market in the medium term if the GOP fails to pass any meaningful legislation. The U.S. economy does not need any demand stimulus at the moment (although measures to boost the supply side of the economy would help lift profits over the long term). The current long-in-the-tooth expansion is likely to stretch further in the absence of stimulus, extending the moderate growth/low inflation/low interest rate backdrop that has been positive for risk assets in recent years. Gold Update Our tactical view on gold remains bearish, but the BCA Commodity & Energy Strategy service sees strategic value in gold as a hedge against rising inflation and inflation expectations, geopolitical risk and increased equity volatility.5 Chart 5A shows that the price of gold in real terms is still very expensive. On a nominal basis, gold is at the top end of a trading channel that it has been in since early 2012 (Chart 5B). There has been a big gap between the model value and the actual price of gold for the past three years. The real price of gold remains elevated despite the fact that inflation has been well contained.6 Chart 5AModel Suggests Gold Is Overvalued Model Suggests Gold Is Overvalued Model Suggests Gold Is Overvalued Chart 5BIn A Downward Channel Since 2012 In A Downward Channel Since 2012 In A Downward Channel Since 2012 Our 6-12 month view on gold is that it will take its cues from Fed policy and policy expectations. The Fed is not behind the curve on inflation, and inflation expectations and measured inflation remain low. Our CPI and PCE models (Chart 6) show only a modest acceleration in inflation by year end, just enough to keep the Fed on track this year as it begins to shrink its balance sheet and raise rates two more times. Thus, we do not see a great need to hold gold as a hedge against inflation over the next year. Nonetheless, for those investors concerned about a pullback that turns into a correction or a bear market, we mention that gold has a 33% weight in our Protector Portfolio.7 Chart 6Core Inflation To Stay Near##BR##Fed's Target This Year Core Inflation To Stay Near Fed's Target This Year Core Inflation To Stay Near Fed's Target This Year Bottom Line: Gold is expensive in real terms relative to a set of fundamentals that have explained its real price since 1970. However, the yellow metal may have value on a strategic basis or as part of a portfolio designed to protect against falling equity prices. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com 1 Please see U.S. Investment Strategy Weekly Report, "Earnings Rebound Will Earn Some Respect", April 10, 2017. Available at usis.bcaresearch.com. 2 Please see U.S. Investment Strategy Weekly Report, "Growth, Inflation And The Fed", May 8, 2017. Available at usis.bcaresearch.com. 3 Please see U.S. Equity Strategy Weekly Report, "Operating Leverage To The Rescue?," published April 17, 2017. Available at uses.bcaresearch.com. 4 Please see U.S. Investment Strategy Weekly Report, "Spring Snapback?," published April 24, 2017. Available at usis.bcaresearch.com. 5 Please see Commodity & Energy Strategy Weekly Report, "Go Long Gold As A Strategic Portfolio Hedge," published May 4, 2017. Available at ces.bcaresearch.com. 6 Please see U.S. Investment Strategy Weekly Report, "Gold: The Asset Allocation Dilemma," published August 1, 2011. Available at usis.bcaresearch.com. 7 Please see U.S. Investment Strategy Weekly Report, "Still Awaiting The Next Pullback," published May 15, 2017. Available at usis.bcaresearch.com.

Highlights We are going long spot gold at tonight's closing price, given our view that inflation and inflation expectations will continue to move higher going into 2018. In the U.S., we expect higher fiscal spending and tax cuts hitting the economy next year to have a significant effect on an economy already at or very close to full employment, boosting real wages and inflationary pressures. As a safe-haven, gold also is well suited to hedging geopolitical risks, which also are rising. Lastly, gold exposure has the added benefit of providing a hedge to equity positions. Energy: Overweight. The ~ 10% correction in benchmark crude oil prices from 1Q17 levels likely has run its course, as representatives of key states that are party to the November 2016 production cut deal signal it will be extended at the upcoming May 25 meeting in Vienna. We remain long Dec/17 Brent $65/bbl calls vs. short the Dec/17 Brent $45/bbl puts, which is down $0.88/bbl, and will be getting long Dec/17 Brent $55/bbl calls vs. Dec/17 $60/bbl calls at tonight's close. We expect Dec/17 Brent to reach $60/bbl by year-end, with WTI trading ~ $2.00/bbl lower. Base Metals: Neutral. Indonesia's state mining company PT Aneka Tambang is expected to resume nickel exports, reversing a three-year ban on outgoing trade. We remain neutral base metals. Precious Metals: Neutral. We are recommending an allocation to gold outright as a strategic hedge against higher inflation, particularly emanating from the U.S., and geopolitical risk in Europe (see below). Underweight. Markets remain well stocked with indications stocks-to-use data will continue to weigh on prices. We remain bearish. Feature Recent indications inflation and inflation expectations are ticking higher will persist into 2018 (Chart of the Week). U.S. fiscal spending and tax cuts expected next year will lift real wages and boost spending power. The American economy already is at or very close to full employment, and U.S. rate hikes are lagging wage growth, which will, all else equal, boost inflation and inflation expectations (Chart 2). Although we expect the Fed to raise rates at least two more times this year - perhaps three - we believe the central bank will continue to keep rate hikes behind wage growth, and will not try to get out in front of inflation (Chart 3). Chart Of The WeekGlobal CPI Inflation Continues To Percolate Global CPI Inflation Continues To Percolate Global CPI Inflation Continues To Percolate Chart 2Rate Hikes Lagging Wage Growth Rate Hikes Lagging Wage Growth Rate Hikes Lagging Wage Growth Chart 3Fed Likely Won't Get Ahead Of Inflation Fed Likely Won't Get Ahead Of Inflation Fed Likely Won't Get Ahead Of Inflation On the political and geopolitical fronts, looming Italian elections are a risk that is all but being ignored by financial markets. Our colleague Marko Papic, head of BCA's Geopolitical Strategy service, identifies next February's Italian elections as "the highest probability risk to European integration at the moment," given its potential to "reignite Euro Area breakup risk."1 Political risks dog the DM economies: falling support for globalization, which will undermine the benefits of sourcing low-cost inputs (labor and capital) worldwide; tighter immigration policies, which go hand-in-hand with falling support for globalization; a predisposition to monetize debt via higher money supply; and higher minimum-wage demands as income inequality increases all raise inflation and inflation expectations in DM economies.2 This financial and political backdrop again points us toward gold in an attempt to identify safe-haven assets and hedges against the increasing likelihood of renewed inflation. In addition, while our House view does not include a marked equities correction in the near term, it is worthwhile pointing out that gold does hedge equities when they are selling off, and in bear markets generally. A corollary to this property is that in equity bull markets, gold tends to hold value, even if it underperforms stocks in absolute terms. These are powerful properties, which increase the stability of investors' portfolios. Before proceeding, it is useful to distinguish between the specifications mentioned above:3 A safe-haven asset refers to an asset that is negatively correlated (or uncorrelated) with other assets that lose value in times of financial stress. An important feature of a safe-haven asset is that it only exhibit low or negative correlation with financial assets (e.g., equities) in extremely negative market conditions, without specifying any particular behavior when markets are not under stress. In other words, both assets could be positively correlated in bull markets, as long as the correlation turns negative when financial-market conditions deteriorate. We make a distinction between the weak and strong form of safe-havens: The weak form represents an asset that is uncorrelated with the reference asset, while the strong form is negatively correlated.4 A hedge is an asset that is negatively correlated (or uncorrelated) with another asset, on average, over the time interval being examined in a particular analysis. As with safe-haven assets, there is a similar distinction between weak- and strong-form hedges. A diversifier refers to an asset that is positively, but imperfectly, correlated with another asset on average during the period of analysis. Gold Vs. Inflation During inflationary periods, assets that generate returns for investors that offset purchasing-power losses experienced by other assets in their portfolio - i.e., a store of value - traditionally have been preferred. Gold has been used as a store of value during inflationary episodes, and for this reason is viewed as a safe haven. Fundamentally, gold's supply is relatively inelastic, and consists of above-ground physical stocks comprising public and private holdings. The world gold council estimates physical gold stocks were ~ 4570.8t at the end of 2016, up 5.8% since 2010. Demand for gold was estimated at 4249.1t at the end of 2016, versus 3281t at the end of 2000. The inelasticity of gold supply makes it difficult to respond to changes in inflation - or to any shocks to the economy, for that matter - by increasing the supply over the short term, as it would be the case with any fiat currencies and other assets. For this reason, price allocates limited supply. During inflationary periods and during a macroeconomic shock, gold's price is bid up, which is the source of returns for holding gold.5 Gold often is seen as a currency; however, it lacks a central bank that can increase its supply via turning up the printing press. This makes the precious metal a so-called "hard currency," and endows it with the ability to maintain its purchasing power during periods of inflation. In addition, it is an asset that is accepted as collateral to support bank lending and margining by the BIS and numerous banks.6 In Table 1, we look at the correlation between year-on-year gold return and U.S. CPI inflation.7 We used a sample period from 1985 to now.8 On average, during the entire sample, we obtained a correlation of 26%. Within the sub-periods gold provides a hedge against inflation, but how much of a hedge depends on other financial factors - chiefly the broad USD TWI and real U.S. interest rates - affecting its performance (Chart 4). We examine these below. Table 1Gold Vs. U.S.##BR##And EU Inflation Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge Chart 4Gold's Inflation-Hedging Properties##BR##Affected By Monetary Conditions Gold's Inflation-Hedging Properties Affected By Monetary Conditions Gold's Inflation-Hedging Properties Affected By Monetary Conditions The hedging relationship between gold returns and the CPI inflation rates does not consistently hold up in all bear markets - e.g., the GFC, when global assets became highly correlated and lost significant value. It is possible, though, that in times of financial stress or downturn, gold's ability to act as a hedge asset to U.S. equities might sometime dominates its ability to hedge inflation, leading to an ambiguous relationship with inflation during bear markets. We delve further into this below. Gold, Inflation And U.S. Monetary Conditions We typically model gold as a function of financial variables, which are sensitive to inflation and inflation expectations and to Fed policy shifts. Given our preference for modeling gold's price evolution as a function of U.S. financial variables - the broad trade-weighted (TWI) USD and real rates, in particular - we looked further into this (Chart 5). The impact of inflation on gold prices is stronger when the dollar experiences large negative shocks and depreciates, and weaker when the USD appreciates (i.e., a large positive shock).9 So, when the USD broad TWI is falling, gold is an effective hedge. When the greenback is appreciating, it is less effective. Next, we examined the ability of gold to hedge inflation risk when U.S. real rates are high and low. To do this, we used 10-year real rates and cut a long-term sample from 1990 to now into two different sub-periods: a high-rate period from 1990 to 2003, and a low-rate period from 2003 to now (Chart 6).10 Chart 5USD's Evolution Is Important To Gold,##BR##As Are U.S. Real Rates USD's Evolution Is Important To Gold, As Are U.S. Real Rates USD's Evolution Is Important To Gold, As Are U.S. Real Rates Chart 6U.S. 10-Year##BR##Real Rates U.S. 10-year Real Rates U.S. 10-year Real Rates During the high-real-rate period, the correlation between gold and inflation is close to zero (0), meaning gold did not act as a strong hedge against inflation, but still could have been acting as a weak hedge (meaning it's uncorrelated). Gold's hedging ability increased significantly in the low-real-rate period (Table 2). Again, this supports our theory that gold's hedging ability depends on U.S. monetary conditions, and that during periods of low real U.S. interest rates gold is an effective hedge against inflation. Table 2Gold Vs. CPI Inflation In High- And Low-Real Rate Environments Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge Gold Vs. U.S. Equities Cutting right to the chase, gold can be used to hedge equities exposure in portfolios, as the correlation analysis in Table 3 demonstrates. Here, we are examining the hedging ability of gold relative to the U.S. stock market (proxied by the S&P 500 Total Return (TR) index). Table 3Gold's Hedging Properties Vs. Equities Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge In our analysis, we find gold and U.S. equities are negatively correlated, on average, over the entire sample (correlation coefficient -0.19). We also tested for time-varying correlation by looking at the correlation separately in different bull- and bear-market sub-periods. Bull (bear) markets are defined as periods in which the U.S. stock index has a positive (negative) move of more than 15% and that lasts for at least 3 months.11 During both bear markets, gold's annualized compound returns were up when the S&P 500 returns were negative (Table 4). This strongly suggests gold is a safe-haven asset in time of extended weakness for equities, all else equal (i.e., we don't have a 100-year global meltdown that takes all correlations to 1.00). Interestingly, the relationship is unclear for bull markets which reflects the non-linearity in gold's hedging ability. We can conclude that during bull markets, gold tends to underperform equity markets; however, this does not imply that holding gold will lead to negative returns. Hence, gold offers protection against bear markets that offsets the costs in terms of returns during bull markets.12 Table 4Gold Hedges U.S. Equities Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge The correlation between month-on-month gold and S&P 500TR returns corroborate the earlier finding. We find that gold is negatively correlated with U.S. equities during equity bear markets, and that it is ambiguous in equity bull markets. Bottom Line: We find gold is a good hedge during inflationary periods, particularly when the USD TWI is weak and real rates are low. We also show gold has excellent safe-haven and hedging properties versus equities (using the S&P 500TR index as a proxy). Based on this analysis, we are recommending a strategic allocation to gold, and will get long at tonight's close. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com Hugo Bélanger, Research Assistant Commodity & Energy Strategy hugob@bcaresearch.com 1 Please see "Political Risks Are Understated in 2018," published on April 12, 2017, by BCA Research's Geopolitical Strategy. It is available at gps.bcaresearch.com. 2 Please see "The End Of the Anglo-Saxon Economy?" published April 13, 2016, by BCA Research's Geopolitical Strategy. It is available at gps.bcresearch.com. 3 Baur, Dirk G.; Brian M. Lucey (2010), "Is Gold a Hedge or a Safe Haven? An Analysis of Stocks, Bonds and Gold". The Financial Review 45, 217-229. 4 Baur, Dirk G.; Thomas K.J. McDermott (2010), "Is Gold a Safe Haven? International Evidence", Journal of Banking & Finance 34, 1886-1898. 5 We would note that the real price of gold increased during the Great Depression, which indicated gold's value during a period of significant deflation appears to increase, perhaps as investors fear the debasement of their currencies and the subsequent loss of purchasing power. 6 Please see Section 4 of "Basel III counterparty credit risk and exposures to central counterparties - Frequently asked questions," published by the BIS December 2012. 7 We use CPI here because it drives the payout of inflation-linked securities in the U.S. 8 We begin our analysis in 1990 for consistency throughout. We also note that several papers take note of an important structural break in U.S. inflation around 1984. Please see Batten, Jonathan A.; Cetin Ciner; Brian M. Lucey (2014), "On The Economic Determinants Of The Gold-Inflation Relation", Resources Policy 41, 101-108; and Stock, James H.; Mark W. Watson (2007), "Why Has U.S. Inflation Become Harder to Forecast?", Journal of Money, Credit and Banking 39 (supplement). For the selection of bear and bull markets, please see "Monthly Economic Report" published on April 2017, by Mackenzie investments. 9 We did this by estimating a regression to see how gold responds when the broad trade-weighted USD is trading in the 5% and 90% quantile of year-on-year U.S. dollar variation over the period 1995 to present. We did this using dummy variables to represent the impact of U.S. inflation in periods of large dollar appreciation and dollar depreciation. The model's adj-R2 is 0.45, and all coefficients are significant below 5%. 10 The mean for the high-rates period is 3.77%; for the low-rates period it is 1.07%. These rates are statistically different between these two sub-periods (using a two-tailed t-test). 11 The selection of bull and bear markets is based on Mackenzie investment analysis. Please see "Monthly Economic Report" published on April 2017, by Mackenzie investments. 12 Our results were supported by further econometric analysis of the variance properties using GARCH modeling. These results are available upon request. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed In 2017 Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge Go Long Gold As A Strategic Portfolio Hedge Summary Of Trades Closed In 2016

Feature Table 1Recommended Allocation Monthly Portfolio Update Monthly Portfolio Update Don't Worry About The Tepid Data Risk assets are likely to continue to grind higher. Two of the catalysts we cited for this in our most recent Quarterly1 have half happened: European political risk is lifting now that Marine Le Pen looks most unlikely to win in the second round of the French presidential election (polls give her less than 40% of the vote); and the Trump administration announced its tax cut plan (which, though details are still sparse, we expect to be passed in some form this year). As a result, the MSCI All Country World Index hit a record high in late April and the S&P 500 is only 1% below its high. But both growth and inflation have surprised somewhat to the downside in the past couple of months. The Citi Economic Surprise Index for the U.S. has fallen sharply, though surprises remain fairly positive elsewhere (Chart 1).Q1 U.S. real GDP growth came in at an annualized rate of only 0.7%. This has pushed bond yields down (with the US Treasury 10-year yield falling back to 2.2%), consequently weakening the dollar. We are not unduly worried about the tepid data. It is mainly due to technical factors. Corporate loan growth in the U.S., for example (Chart 2), mostly reflects just the lagged effect of last year's slowdown on banks' willingness to lend, as well as energy companies repaying credit lines they tapped in early 2016 when short of working capital. The weakness in auto sales (Chart 3) is most likely caused by the end of the car replacement cycle which began in 2010, rather than reflecting any generalized deterioration in consumer behavior. Moreover, there seem to be problems with seasonal adjustment of data caused by the extreme swings in the economy in 2008 and 2009: Q1 has been the weakest quarter for U.S. GDP in six out of the past 10 years, and has on average been 2.3 ppts lower than Q2.2 There were no such distortions prior to 1996. Chart 1U.S. Growth Has Surprised To The Downside U.S. Growth Has Surprised To The Downside U.S. Growth Has Surprised To The Downside Chart 2Weaker Loan Growth Is Mostly Technical... Weaker Loan Growth Is Mostly Technical... Weaker Loan Growth Is Mostly Technical... Chart 3...And The Slowdown In Autos Is Just The End Of A Replacement Cycle ...And The Slowdown In Autos Is Just The End Of A Replacement Cycle ...And The Slowdown In Autos Is Just The End Of A Replacement Cycle A consequence of the wobbly data is that markets have become too complacent about the Fed raising rates, with futures markets now projecting only about 40 bps of hikes over the next 12 months (Chart 4). Our view is that wages will gradually move up this year, pushing core PCE inflation to 2% by year end, which will cause the Fed to raise rates twice before end-2017 and once early in 2018 (though the latter rise could be postponed if the Fed starts to reduce its balance-sheet and forgoes one quarter's hike to judge the impact of this on the market). By contrast, we do not see the ECB hiking before 2019 at the earliest, with ECB President Draghi reiterating that he sees core inflation staying low and remains concerned about the fragile banking systems in peripheral European markets and about Italian politics. We also believe Bank of Japan governor Kuroda when he says he has no plans to change the BoJ's 0% target for the 10-year JGB yield. All this implies that the dollar is likely to appreciate further in the next 12 months as interest rate spreads widen (Chart 5). Chart 4Fed Is Likely To Hike Faster Than This Fed Is Likely To Hike Faster Than This Fed Is Likely To Hike Faster Than This Chart 5Interest Differentials Suggest Further Dollar Strength Interest Differentials Suggest Further Dollar Strength Interest Differentials Suggest Further Dollar Strength The next catalyst for equities to rise further could be earnings. Q1 U.S. earnings are surprising significantly on the upside, with EPS growth of 11.7% year on year and 75% of companies beating analysts' estimates.3 BCA's proprietary model suggests that S&P 500 operating earnings this year could grow by over 20% (Chart 6). If anything, upside surprises to earnings have been even stronger in the euro zone and Japan. With none of the standard indicators signaling any risk of recession over the next 12 months (Chart 7), we remain overweight equities versus bonds. We continue to warn, though, that the Goldilocks scenario of healthy growth and stable inflation may not last for long. A combination of tax cuts, wage growth accelerating as labor participation hits a ceiling, and the Fed falling behind the curve (perhaps when President Trump - given that he recently confessed "I do like a low interest rate policy" - appoints a dovish replacement for Janet Yellen as Fed Chair) could cause inflation to rise unexpectedly next year, forcing the Fed to raise rates sharply, triggering a recession in 2019. Chart 6U.S. Earnings Could Grow 20% This Year U.S. Earnings Could Grow 20% This Year U.S. Earnings Could Grow 20% This Year Chart 7No Sign Of A Recession On The Horizon No Sign Of A Recession On The Horizon No Sign Of A Recession On The Horizon Equities: In a risk-on environment, euro zone equities should continue to outperform, due to their higher beta (averaging 1.3 against global equities over the past 20 years, compared to 0.9 for the U.S.), more cyclical earnings, and modestly cheaper valuations (forward PE is at a 18.9% discount to the U.S.). Japanese equities should also do well as interest rates rise again globally (except in Japan where the BoJ will stick to its 0% yield target on 10-year bonds), which should push down the yen and boost earnings. We remain overweight Japanese equities on a currency-hedged basis. We are underweight EM equities, which are likely to be weighed down over the next 12 months by the stronger dollar, and by a slowdown in China which should cause commodity prices to fall. Fixed Income: We expect the 10-year U.S. Treasury yield to reach 3% by year-end: a pickup in real growth, slightly higher inflation and two more Fed hikes can easily add 70 bps to the yield over the next eight months. Euro zone yields will also rise, though not by as much. This implies a negative return from G7 sovereign bonds for the first time since 1994. We continue to prefer corporate credit, with a preference for U.S. investment-grade debt over high-yield bonds (which have stretched valuations) and over European corporate debt (which will be negatively affected by the tapering of ECB purchases next year). Currencies: As described above, we do not believe that the dollar appreciation which began in 2014 is over, due to divergences in monetary policy. We would look for a further 5-10% appreciation of the dollar over the coming 12 months, though the rise is likely to be bigger against the yen and emerging market currencies than against the euro. Commodity currencies such as the Australian dollar also look vulnerable and overvalued. The British pound will be driven by the vicissitudes of the Brexit negotiations in the short-run but looks undervalued in the long run if, as we expect, the EU eventually agrees a moderately satisfactory trade deal with the U.K. Commodities: We continue to believe that the equilibrium level for oil is $55 a barrel, and that an extension of the OPEC production agreement beyond June and a drawdown in inventories in the second half will bring WTI crude back to that level - with the risk of even $60-65 temporarily if there are any unforeseen supply disruptions. We remain more cautious on industrial commodities, which will be hurt by a mild withdrawal of monetary and fiscal stimulus in China. Following its 6.9% GDP print in Q1, Chinese growth is likely to slow moderately. However, with the Party Congress coming up in the fall, growth will not be allowed to slow excessively - and, indeed, there are signs that central government spending has begun to accelerate recently (Chart 8). We remain positive on gold as a long-term hedge against the tail risk of inflation. As our recent Special Report on Safe Havens demonstrated,4 gold has historically provided good returns during recessions, particularly those associated with high inflation (Chart 9). Chart 8China Is Withdrawing Stimulus - Or Is It? China Is Withdrawing Stimulus - Or Is It? China Is Withdrawing Stimulus - Or Is It? Chart 9Gold Glisters When Inflation Rises Gold Glisters When Inflation Rises Gold Glisters When Inflation Rises Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com 1 Please see Global Asset Allocation, "Quarterly Portfolio Outlook: No Reasons To Turn Cautious," dated 3 April 2017, available at gaa.research.com 2 For detailed analysis of the problems with seasonal adjustment, please see U.S. Investment Strategy, "Spring Snapback?" dated April 24, 2017, available at usis.bcaresearch.com 3 So far about half of U.S. companies have reported. 4 Please see Global Asset Allocation, "Safe Havens: Where To Hide Next Time?" dated April 21, 2017, available at gaa.bcaresearch.com. Recommended Asset Allocation

Highlights Safe-haven assets do not simply outperform equities on a relative basis during bear markets. In fact, the average return of nine safe-haven assets has been positive in every bear market since 1972. A safe haven should serve two purposes. First, it should have a negative correlation with equities during bear markets, not necessarily in all markets. Second, it should have an insurance-like payoff, surging during systemic crashes. Low intra-correlations between safe-haven assets, and substantial absolute differences between individual returns and the overall group average suggest that selection adds significant alpha. In the next bear market, we recommend positions in CHF over USD and JPY, due to its greater consistency as a safe-haven asset and more attractive valuations. Favor gold over farmland and TIPS, as gold offers a better hedge against political risks while still protecting against rising inflation. Overweight Treasuries relative to Bunds given a more appealing return distribution and high spreads. Feature Feature ChartSafe Haven Performance Safe Haven Performance Safe Haven Performance As the economic expansion approaches its 100th month, far longer than 38.7 month average1 of cycles starting from 1854, concerns continue to mount over the next recession and equity market crash. Memories of over 50% losses in stocks during the subprime crisis are still ingrained in investors' minds and the importance of capital preservation and safe-haven assets cannot be stressed enough. Safe-haven assets do not simply outperform equities on a relative basis during bear markets. In fact, during the subprime crisis, an equal-weighted portfolio of nine safe-haven assets actually increased in absolute value by 12% (Feature Chart)! This has held consistent through every bear market since 1972 and we expect the next crisis to be the same. While we do not expect a bear market in the next 12 months, we do stress the importance of being prepared and tactically flexible given the substantial relative and absolute performance of safe-haven assets. In this Special Report, we analyze behaviors of safe havens during past bear markets in order to recommend tilts to outperform during the next major equity selloff. Historical Perspective For our analysis, we used monthly return data to more accurately compare across asset classes. We used the following nine safe-haven assets: U.S. Dollar - As the world's reserve currency, the U.S. dollar benefits from massive trade volumes. Japanese Yen - Japan is still the world's 3rd largest economy and runs a current account surplus. Investors' perceptions of safety are intact and the currency benefits from unwinding of carry trades during risk-off environments. Swiss Franc - Switzerland has built a reputation for its international banking prowess, political neutrality and economic stability. U.S. Farmland - Farmland differs from the others in that it is a tangible, hard asset. With finite supply and an increasing population leading to higher needs for farming and food, demand will remain robust. U.S. Treasuries - Treasuries have essentially no default risk. Since its formation in 1776, the U.S. has never failed to pay back its debt. German Bunds - Germany benefits from being economically and politically stable. Bunds are extremely liquid and could receive capital inflows in the event of euro area disintegration. Gold - Gold has a longstanding history as a safe-haven asset, protecting against inflation, currency debasement and geopolitical risks. U.S. TIPS - TIPS are the purest inflation hedge; their historical performance has held a very tight correlation with realized changes in consumer prices. Hedge Funds - Hedge funds are attractive given their lack of restrictions and ability to short. We classified an equity bear market as a decline in the S&P 500, from peak to trough, larger than 19%.2 Using this definition, we recorded eight separate instances since 1972 (See Appendix). On average, these episodes lasted about 14 months and equity prices experienced declines of 34%. We examined returns, correlations and recession characteristics in order to draw conclusions about potential future behavior. Key Findings: During bear markets, the value of these nine safe havens increased on average by 9.2% (Table 1). This certainly does not offset the 34% average decline in equities, but it does provide a considerable buffer, particularly if allocators tilt asset class weightings. However, there is concern that safe havens as a whole will not provide as much protection in the next downturn as they have in the past, given weak equity inflows and still-considerable cash on the sidelines (Chart 2). The average absolute spread between the returns of the nine safe havens and their overall average return was 12.3%. While the correlations between financial assets tend to spike upwards during bear markets, they actually remain very low between safe-haven assets. This indicates a significant opportunity for alpha generation during equity downturns. The region from which a crisis stems has little impact on which safe haven outperforms. For example, U.S. Treasuries and the U.S. dollar both increased in value during the past two recessions, despite the tech bubble and subprime crisis originating from the U.S. (Chart 3). Capital inflows into those assets remained robust given their reputation for safety and quality. U.S. Treasuries and the Swiss franc always had positive absolute returns during the eight bear markets, and therefore have always had a negative correlation with equities (Table 2). These two assets have very stable reputations for safety. Nevertheless, other safe havens, such as gold, USD, JPY and Bunds, still maintained negative correlations with equities during most bear markets. U.S. farmland and U.S. TIPS also had positive returns in the three bear markets since their starting dates. Hedge funds, while known to outperform equities during bear markets, did not provide positive absolute returns in any of the four equity downturns since the index began. Table 1Bear Market Performance Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? Chart 2Safe Havens: Less Protection Next Time? Safe Havens: Less Protection Next Time? Safe Havens: Less Protection Next Time? Chart 3Location Doesn't Matter Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? Table 2Correlation With Equities Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? Investment Implications Chart 4A Near-term Bear Market Is Unlikely A Near-term Bear Market Is Unlikely A Near-term Bear Market Is Unlikely It is crucial to understand the purpose of a safe-haven asset as it pertains to portfolio management. First, a safe-haven asset should have a negative correlation with equities during bear markets, not necessarily in all environments. Secondly, and more importantly, a safe-haven asset should have an insurance-like payoff, surging during systemic crashes. As safe havens naturally receive a smaller allocation in typical portfolios due to their underperformance versus equities in most years, it is imperative that relatively smaller weightings and minor tilts offset large declines in equity prices. It is important, however to note that we view the probability of a bear market as highly unlikely over the next twelve months (Chart 4). First, substantial stock price declines are not very common outside of recessions. As our colleague Martin Barnes points out, the yield curve is not inverted, there are no serious financial imbalances, and the leading economic indicator remains in an uptrend.3 Monetary conditions are still stimulative, and it generally requires Fed tightening to surpass equilibrium before recessions occur. Massive average absolute deviations for each individual safe haven from the overall group average and low intra-correlations suggest that selection adds significant alpha (Chart 5). Unlike most financial assets, intra-correlations between safe havens actually decline during bear markets. In order to best compare and contrast safe havens, we divided the assets into three buckets: currencies, inflation hedges and fixed income. Below, we recommend tilts within these buckets and will revisit these recommendations closer to the next bear market. Chart 5Intra-correlations Remain Low In Bear Markets Intra-correlations Remain Low In Bear Markets Intra-correlations Remain Low In Bear Markets Currencies: Overweight CHF relative to USD and JPY. As a zero-sum game, currency selection offers a critical avenue for alpha generation. As global growth continues to improve and capital flows to more cyclical currencies, or to the USD where policymakers are tightening, the Swiss franc should become even more attractively valued. The franc's considerable excess kurtosis, indicating higher likelihood of outsized returns, best fits the insurance-like payoff quality (Chart 6). It is the only currency to have outperformed, and therefore held a negative correlation with equities, during each of the eight recessions, indicating high reliability as a safe-haven asset. Going forward, we see no reason for Switzerland's reputation for economic stability or political neutrality to be compromised. The biggest risk to this view would be if the Swiss National Bank were to stick stubbornly to its peg of the CHF to the EUR during the next recession, thereby dampening the franc's risk-off properties. The USD has historically been able to outperform even when the crisis originated in the U.S. Historical bear market performance was greatest, however, following sharp Fed tightening such as the Volker crash, when the Fed increased rates in response to high inflation, or in the subprime crisis, when the Fed increased rates to slow growth (Chart 7). While we expect inflation and growth to grind upward over the cyclical horizon, our base case is not for a surge in consumer prices or for economic growth to expand significantly above trend. Chart 6Return Distributions Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? Chart 7Fed Tightening = USD Outperformance Fed Tightening = USD Outperformance Fed Tightening = USD Outperformance In the next bear market, the JPY will likely benefit from cheap starting valuations as the BoJ is currently aggressively easing, and its current account surplus raises its fair value. Nevertheless, the yen's returns during equity downturns have not always been consistent with its safe haven reputation. Of the three currencies, since 1970, it has had the lowest probability for large returns. Inflation Hedges: Overweight Gold relative to TIPS and Farmland. Over most of the time frames we tested, gold had the highest correlation with both headline and core inflation (Tables 3 & 4). Table 3Correlation With Core Inflation Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? Table 4Correlation With Headline Inflation Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? The main differentiating factor with gold is its ability to hedge against political risk. Our geopolitical strategists found that of all of the safe-haven assets, gold offered the best protection against political shocks4 (Chart 8). As mentioned in one of our recent Special Reports,5 we believe that stagnation in median wages and wealth inequality will continue to fuel the rise in populism and social unrest. Chart 8Gold Is Best At Hedging Political Risk Safe Havens: Where To Hide Next Time? Safe Havens: Where To Hide Next Time? Farmland has historically offered decent inflation protection, but its history is limited, supply is scarce and the massive runup in prices is a cause for concern. While we currently favor TIPS over nominal bonds, their negative skew and excess kurtosis suggest that they are vulnerable to large negative returns, making them a less-than-ideal safe-haven asset. Fixed Income: Overweight Treasuries relative to Bunds. Concerns that, because government yields are starting at very low levels, bonds will not provide safety in the next bear market, are overblown. Recent history proves that yields can reach negative territory, and historical performance for government fixed income has been robust in almost every significant equity decline. Additionally, the end of the 35-year decline in interest rates should not negatively affect the protection capabilities of Treasuries. Yields actually rose leading up to, and during, the 1972 and 1980 bear markets, and Treasuries still provided positive absolute returns (Chart 9). One caveat is that starting yields are much lower today. If yields were to rise during the next recession, they may not achieve positive absolute returns, though government bonds would still certainly outperform equities by a wide margin. Overall, Treasuries have held a more negative correlation with equities during bear markets, spreads over Bunds will likely continue to rise given diverging monetary policy, and they have historically been more prone to outsized positive returns during crisis periods (Chart 10). Bunds are currently benefitting from flight-to-quality flows resulting from political and policy issues originating in the periphery. However, at some point, concerns that the euro crisis will spread to Germany may eliminate this advantage. Chart 9Rising Yields Were Not A Problem Rising Yields Were Not A Problem Rising Yields Were Not A Problem Chart 10Relative Treasury Valuations Will Become More Attractive Relative Treasury Valuations Will Become More Attractive Relative Treasury Valuations Will Become More Attractive Patrick Trinh, Associate Editor patrick@bcaresearch.com 1 http://www.nber.org/cycles.html. 2 While a 20% decline may be a more widely-used measure for bear markets, there have been three instances of 19% declines since 1972, one of which was a recession. We decided to include these in our analysis to increase the number of observations and improve the reliability of our analysis. 3 Please see The Bank Credit Analyst Special Report, "Beware The 2019 Trump Recession," dated 7 March 2017, available at bca.bcaresearch.com. 4 Please see Geopolitical Strategy Special Report, "Geopolitics and Safe Havens" dated November 11, 2015, available at gps.bcaresearch.com. 5 Please see Global Asset Allocation Special Report, "Refreshing Our Long-Term Themes," dated 5 December 2016, available at gaa.bcaresearch.com.

Global gold mining stocks are breaking out on the back of the bid in gold prices. The confluence of spiking geopolitical uncertainty, both in the U.S. and globally, rise in equity volatility and safe haven bid in U.S. Treasurys have raised the profile of gold-related hedges. Importantly, the top panel of the chart shows that the surge in the economic policy uncertainty index heralds additional relative gains for gold-related shares. Similarly, easing real interest rates are raising the profile of gold prices. This portfolio hedge has served our portfolio well since the broad market's peak in early-March and more gains are in store until this consolidation/correction phase fully runs its course. Bottom Line: We are reiterating our overweight stance in the global gold mining index via the GDX:US exchange traded fund. A Gold Hedge Still Makes Sense A Gold Hedge Still Makes Sense

Highlights The Fed's evident desire to lift its policy rate next week - presumably to get out ahead of inflation that has yet to show up in its preferred gauge - will weigh on gold. Oil ... not so much. This is because fundamentals once again are asserting themselves in the evolution of oil prices, something that has been evident even before markets balanced last year. Gold, meanwhile, remains exquisitely sensitive to Fed policy expectations and their effects on the USD and real rates, as with other currencies. Energy: Overweight. We are looking to re-establish our long WTI Dec/17 vs. short Dec/18 spread if it trades in contango again, i.e., if Dec/17 is less than Dec/18. We believe the combination of OPEC and non-OPEC adherence to their production Agreement will remain high, and demand likely will remain stout. Base Metals: Neutral. Spot copper is down ~ $0.10/lb on COMEX over the past week. We expect transitory supply issues in Chile and Indonesia to be resolved, and reflationary stimulus in China to wane going into the 19th National Congress of the Communist Party in the autumn, and, with it, copper demand. We remain neutral. Precious Metals: Neutral. Gold is weakening as the Fed's March meeting approaches next week, given the overwhelming expectation for a 25bp rate hike. We remain long volatility, expecting fiscal-policy uncertainty in the U.S. to be resolved over the next few months, and Fed policy drivers to become more focused. Ags/Softs: Underweight. We are not expecting significant changes in the USDA's estimates of stocks globally, and therefore remain underweight. Feature The choreographed messaging of voting and non-voting FOMC members asserting the need for a policy-rate hike over the past two weeks succeeded in pushing markets' expectations for such action to 88.6% as of Tuesday's close, up from 44.6% at the end of February. This despite the fact that the Fed's preferred inflation gauge - core PCE - has yet to show any sign of pushing up and thru the Fed's target of 2% growth yoy (Chart of the Week). Nor, for that matter, has core PCE shown any tendency to remain above 2% yoy growth over the past two decades (Chart 2). Chart of the WeekThe Fed's Preferred Inflation ##br##Gauge Still Quiescent The Fed's Preferred Inflation Gauge Still Quiescent The Fed's Preferred Inflation Gauge Still Quiescent Chart 2Core PCE Has Been ##br##Quiescent For Decades Core PCE Has Been Quiescent For Decades Core PCE Has Been Quiescent For Decades Between mid-December 2016 and the end of last month, gold prices rallied ~11.3% largely on the expectation the Fed would not raise rates until at least June, and, even then, would be constrained by uncertainty over what Congress and the Trump Administration would offer up in terms of fiscal policy later this year. Now, with the Fed succeeding in raising the market's expectation of a March rate hike, gold markets are left to re-calibrate the number of hikes to expect this year, and the likely implications for the USD and real rates. We believe the Fed will execute three rate hikes this year, but this will be highly dependent on how markets react to the now fully priced-in hike markets expect next week. Synchronized Growth, Inflation And Feedback Loops It is likely the Fed feels confident accelerating its rates normalization because, for the first time since the Global Financial crisis, we are getting a globally synchronized recovery in GDP. All else equal, this will give the U.S. central bank a bit of headroom to experiment with an earlier-than-expected rate hike. This synchronized growth also will provide a positive backdrop for commodity demand this year and next (Chart 3). The possibility of highly stimulative - or even just moderately stimulative - fiscal policy in the U.S. at a time when the economy is apparently at or close to full employment, will be positive for aggregate demand, and could be inflationary if its principal result is to lift real wages in the U.S. In addition to synchronized growth, we also are seeing evidence of synchronized inflation in the largest economies in the world (Chart 4). Chart 3Synchronized Global Growth ##br##Could Embolden The Fed Synchronized Global Growth Could Embolden The Fed Synchronized Global Growth Could Embolden The Fed Chart 4Synchronized Inflation Globally ##br##Likely Caught The Fed's Attention Synchronized Inflation Globally Likely Caught The Fed's Attention Synchronized Inflation Globally Likely Caught The Fed's Attention This synchronized growth and inflation is, we believe, important to the Fed, in that its effects constitute something of a global feedback loop. As we have noted in earlier research, the Fed is much more sensitive to how its policy actions affect other economies, given the deepening of global supply chains over the past two decades or so. Equally, policymakers are well aware the evolution of monetary policy and economic growth in other economies affects the U.S. growth and policy variables important to the Fed.1 Absent a policy shock in the U.S., Europe or China, the backdrop for EM growth should remain positive for at least 2017, even with reflationary stimulus waning in China, a left-tail risk to commodity prices that we identified in last week's publication.2 We expect the Fed's policy normalization to be tempered by continued monetary accommodation globally, which will be supportive of growth at the margin. This will keep global oil demand growth on track to average 1.50 - 1.60mm b/d this year and next, and, importantly for inflation and inflation expectations, keep EM oil demand growing. The income elasticity of per-capita oil consumption in EM economies typically is ~ 1.0, meaning a 1% increase in EM incomes is associated with a 1% increase in EM oil demand.3 EM growth accounts for close to 85% of the growth we expect in global oil demand this year. This is important, given EM oil demand, which we proxy with the U.S. EIA's non-OECD oil consumption time series, to be a common factor that explains the evolution of the CPI series shown above (Chart 5). EM oil demand is able to explain the synchronization of inflation in the three largest economies in the world is because incremental growth is occurring in the EM economies, and this is driving global growth. We continue to expect high compliance in the OPEC - non-OPEC production deal negotiated by the Kingdom of Saudi Arabia (KSA) and Russia at the end of last year, which will, against the backdrop of continued global growth, cause inventories to fall and for markets to backwardate. We believe last week's increase in U.S. crude oil inventories to be the last big build, and expect the decline to begin later this month. On average vessels leaving the Persian Gulf destined for the U.S. have a 45- to 50-day sailing period depending on multiple factors such route, weather and sea conditions. Therefore, the recent increase in U.S. crude oil inventories can be linked to the arrival of the final fleet of vessels in concert with the pre-OPEC agreement production surge undertaken by the GCC. Evidence of this phenomenon is apparent in the ~500k b/d increase in U.S. crude oil imports (374k b/d coming from Iraq) over the prior week. We expect OECD oil stocks to start declining this month and fall some 300mm bbl before the end of 2017. This supply-demand dynamic will continue to dominate financial-market influences on oil prices, as we argued in last week's publication (Chart 6).4 Gold, on the other hand, will continue to take its cue from Fed policy and policy expectations, particularly as regards expectations for the USD, which should strengthen at the margin, given the Fed's new-found hawkishness, and real rates, which also should strengthen (Chart 7). Chart 5EM Oil Demand Continues##br## To Drive Inflation EM Oil Demand Continues To Drive Inflation EM Oil Demand Continues To Drive Inflation Chart 6IF KSA And Russia Can ##br##Coordinate Production... IF KSA And Russia Can Coordinate Production... IF KSA And Russia Can Coordinate Production... Chart 7Gold Will Continue To Take##br## Its Cue From Fed Policy Gold Will Continue To Take Its Cue From Fed Policy Gold Will Continue To Take Its Cue From Fed Policy Bottom Line: Oil prices will continue to be dominated by supply-demand-inventory fundamentals, with monetary policy effects on the evolution of prices taking a secondary role. Gold prices will continue to take their cue from Fed policy and policy expectations. We look to re-establish our long Dec/17 WTI vs. short Dec/18 WTI spread if it trades thru flat (i.e., $0.00/bbl). Given our gold view, we remain long volatility via the put spreads and call spreads we recommended February 23 - i.e., long Jun/17 $1,200/oz puts vs. short $1,150/oz puts, and long $1,275/oz calls vs. short $1,325/oz calls. The position was up 15% as of Tuesday's close. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com 1 Please see BCA Research's Commodity & Energy Strategy Weekly Reports "Global Inflation and Commodity Markets," dated August 11, 2016, and "Memo To The Fed: EM Oil, Metals Demand Key To U.S. Inflation," dated August 4, 2016, available at ces.bcaresearch.com. 2 Please see BCA Research's Commodity & Energy Strategy Weekly Report "Gold's Known Unknowns, And Fat Tails," dated February 23, 2017, available at ces.bcaresearch.com. 3 Oil consumption frequently is employed to approximate EM income growth, given the income elasticity of demand for oil is ~ 1.0, meaning a 1% increase in income (GDP) produces an increase in demand for oil of approximately 1.0%. The OECD notes, "Non-OECD countries are found to have a higher income elasticity of oil demand than OECD countries. On average across countries, a one per cent rise in real GDP pushes up oil demand by half a per cent in OECD countries over the medium to long run, whereas the figure is closer to unity for most non-OECD countries." Please see "The Price of Oil - Will It Start Rising Again?" OECD Economics Department Working Paper No. 1031, p. 6 (2013). 4 Please see BCA Research's Commodity & Energy Strategy Weekly Report "Days Of Oil Future's Past: Mean Reversion," dated March 2, 2017, available at ces.bcaresearch.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed In 2017 Summary of Trades Closed in