Emerging Markets

Highlights A number forward-looking indicators for EM corporate profits point to a major deceleration in the next several months, and potentially a contraction early next year. The most reliable forward-looking indicators for EM EPS have been EM/China narrow and broad money growth and they herald a bearish outlook for EM EPS. We continue deciphering the differences between China's various money and credit aggregates. Irrespective of which money measure we use, and regardless of their past track record, all of them are currently extremely weak and point to a major and imminent slump in China's growth in the next six to 12 months. We recommend shifting the underweight EM corporate and sovereign credit position versus U.S. high-yield to underweight versus U.S. investment-grade corporate credit. Feature Chart I-1Asian Exports And EM EPS The recovery in EM earnings per share (EPS) has been instrumental to the EM stock rally this year. As such, the equity strategy at the moment hinges on the outlook for corporate profits. In this report, we revisit coincident and leading indicators for EM profits. At the moment, EM corporate profit growth still appears robust, though several forward-looking indicators point to a major deceleration in the next several months, and potentially a contraction early next year. Korean and Taiwanese exports can be used as proxy for global trade. The latest data for July reveal that the sum of Taiwanese exports and Korean total exports excluding vessels has rolled over (Chart I-1). Historically, the U.S. dollar values of both economies' exports have correlated with EM EPS, and Chart I-1 entails that EM EPS growth will roll over very soon. The reason why we exclude vessel exports in the case of Korea is because vessel shipments are one-off occurrences and when they take place, they distort export growth. This was the case in the last several months - vessel (shipbuilding) exports surged by 75% from a year ago, distorting the annual growth rate of total exports. Overall, Korea's and Taiwan's overseas shipments in the past three months have averaged about 10%, which is lower than the mid-teen growth rates recorded earlier this year. In China, export growth is close to 9% in the past three months, and it is also rolling over. On a similar note, Korea's and Taiwanese shipments-to-inventory ratios lead EM EPS cycles, and they are presently sending a downbeat message (Chart I-2). China's import growth has relapsed, as suggested by both Chinese trade data and their counterparties export data to China (Chart I-3). Chart I-2Asia's Shipment-To-Inventory Ratios And EM EPS Chart I-3Exports To China And Chinese Imports The recovery in Chinese imports has been responsible for a considerable part of the recovery in global trade. Importantly, Chinese import cycles correlate very well with EM EPS growth (Chart I-4). The key pillar of our view remains that Chinese imports will contract going forward, which will depress both advanced and developing countries' shipments to China. Exports to China are much more important for EM than DM economies, and deteriorating sales to China will weigh considerably on EM profits and currencies. The most reliable forward-looking indicators for EM EPS have been EM/China narrow and broad money growth. Chart I-5A and Chart I-5B demonstrate that both EM narrow (M1) growth and China's broad money impulse (the second derivative) - herald a major slump in EM EPS. This is the main reason behind our negative stance on EM share prices and other risk assets. Chart I-4Chinese Imports And EM EPS Chart I-5AChina Broad Money Impulse And EM EPS Chart I-5BEM Narrow Money And EM EPS Both narrow and broad money growth in China have already relapsed, and it is a matter of time until economic growth and imports downshift enough to produce a major selloff in EM risk assets. We discuss China's monetary aggregates in the section below. Finally, if Chinese imports and commodities prices relapse, any reasonable strength in DM domestic demand will not be sufficient to preclude a meaningful EM slowdown. The basis is that exports to the U.S. and EU only make up 7% of GDP for China, 8% for Korea and 11% for Taiwan. While exports to China account for 10% of Korean GDP and 15% of Taiwanese GDP. The same holds true for most East Asian countries. With the exceptions of India and Turkey, non-Asian EM countries are primarily commodities producers. These two have their own idiosyncratic problems. Most of our analysis is not applicable to smaller central European economies that are leveraged to the EU business cycle. That said, neither Turkey, India, nor central European markets have large enough financial markets to make a difference in the EM benchmarks. The above is the primary reason behind our bearish view on EM growth and profits. That said, there are a few other interesting considerations regarding EM corporate profits dynamics. First, EM share prices lead EM EPS by six to nine months. Therefore, to be bullish on EM stocks, it is not sufficient to expect EM EPS growth to be robust over the next three months. Rather, to be bullish on EM stocks at the current juncture, one should have a bullish view on EM EPS by the end of this year and into the early part of 2018. Consistently, we believe that EM EPS growth will decelerate materially by the end of this year and shrink in the early part of 2018. Second, the top-line shrinkage in 2015 and the consequent recovery for EM exporters has been mostly driven by prices rather than volumes. Chart I-6A illustrate that Korean, Taiwanese and Chinese manufacturing production growth is rather muted. Chart I-6ACorporate Pricing Power Chart I-6BAsian Manufacturing Production Price fluctuations affect profits much more than output volume changes. Therefore, if global tradable goods prices deflate - at the moment they have rolled over (Chart I-6B) - EM EPS will contract materially. Third, in EM excluding China, Korea and Taiwan, there has been little economic recovery, as evidenced by Chart I-7. Along the same lines, the latest (July) manufacturing PMI for EM ex-China, Korea and Taiwan has dropped below the crucial 50 line (Chart I-7, bottom panel). This and the majority of other economic aggregates we use are equity market-cap weighted averages, so they are relevant to investors. This corroborates the fact that outside China, Korea and Taiwan there has been little genuine growth improvement in EM domestic demand - despite the decent recovery in global trade. This challenges the prevailing widespread consensus of a synchronized global economic recovery/expansion. This is also consistent with the fact that the overwhelming EM profit recovery has occurred in technology and resource sectors while domestic sectors have not seen much of corporate earnings recovery (Chart I-8). Chart I-7EM Ex-China, Korea And Taiwan: ##br##No Strong Recovery Chart I-8EM Sectors' EPS: Exporters ##br##Have Outperformed Domestic Finally, bottom-up equity analysts have recently downgraded their EPS estimates for listed EM companies (Chart I-9). Typically, analysts alter their forecasts simultaneously with swings in share prices. Hence, the latest decoupling is puzzling. Chart I-9EM EPS And Analysts' Net Revisions Notably, EM net EPS revisions have failed to move into positive territory in the past 7 years. This entails that analysts' expectations have been chronically high in recent years, and/or that companies have failed to deliver profits that match these projections. Bottom Line: The EM EPS outlook is downbeat, and listed companies profits will likely contract early next year. Deciphering China's Money Puzzle Based on our assessment of multiple measures, our conclusion with respect to Chinese broad money growth is as follows: Irrespective of which measure we use, and regardless of their individual past track records, all Chinese monetary growth aggregates are currently weak (Chart 10), and point to a major and imminent slump in China's growth in the next six to 12 months. In recent weeks, we have been working to understand differences among various measures of money growth in China. Our motivation is because neither M2 nor total social financing and fiscal spending - variables that we relied on last year - did a good job of forecasting the duration and magnitude of China's economic and profit revival in the past 12 months. In our July 26 report,1 we introduced the concept of broad money calculated using commercial banks' assets. We called it credit-money. This week, we discuss a different broad money calculation based on commercial banks' liabilities, and refer to it as deposit-money. Deposit-money is an aggregate of non-financial companies' time and demand deposits, household deposits, transferable and other deposits, other liabilities, bonds issued and liabilities to non-depository financial corporations. This measure is broader than official broad money (M2) because the latter includes only non-financial companies' time and demand deposits, household deposits and some of liabilities to non-depository financial corporations. In brief, our deposit-money calculation is more comprehensive than the official broad money figures (M2). In turn, banks' credit-money is the sum of commercial banks' claims on companies, households, non-bank financial institutions and all levels of government, as well as banks' foreign assets. Also, we deduct government deposits at the central bank (see July 26 Emerging Markets Strategy report1 for more details). Chart I-10 illustrates the differences between credit-money, deposit-money, total social financing and M2. Based on our calculations, deposit-money grew faster in 2015-'16 than both M2 and total social financing. Yet its current and ongoing slowdown is as bad as that of credit-money or M2. Chart I-10Dichotomy Among Various Money And Credit Aggregates In China The reason why M2 growth has lagged behind deposit-money growth since the middle of 2015 until now is the fact that the latter's components that are not included in the official M2 measure have outpaced M2 growth by a wide margin since late 2015. The main components of deposit-money are shown in Chart I-11. This is one of the main reasons why we missed the latest China-play rally - we relied on the official measure of money and credit published by the PBoC that has been much tamer than the broader money and credit, as banks have originated credit and hence money in a way that official monetary aggregates have not captured. In addition, banks' credit-money and deposit-money measures should theoretically be identical, but this has not been the case in China in recent years. Deposit-money is larger and it may well be more comprehensive than credit-money (Chart I-12). Chart I-11China: Components Of Deposit-Money Aggregate Chart I-12The Outstanding Stock And Flow Of Money Understanding these discrepancies is an ongoing work-in-progress for us, and we will be refining these measures going forward. For now, we would say that these differences are probably due to banks' efforts to misrepresent/hide their assets and liabilities to meet the regulatory ratios and avoid penalties, as well as maximize short-term profits. All that said, the gaps between M2 and deposit-money has recently narrowed: both deposit-money and M2 growth and their impulses are at all-time lows (Chart I-13). Furthermore, we expect deposit-money to slow further because of the lagged impact of higher interest rates and regulatory tightening that is intended to curb commercial banks' ability to originate more money via shadow banking activities. Finally, as can be seen from Chart I-14A, Chart I-14B and Chart I-15, deposit-money's impulse - its second derivative - leads many cyclical economic variables such as nominal GDP, producer prices, freight index, and imports. Chart I-13China: Two Measures Of Broad Money Chart I-14ADeposit-Money Leads Real Business Cycle Chart I-14BDeposit-Money Leads Real Business Cycle There are several other data points from China's real economy that portend developing weakness. Specifically, car sales growth has almost ground to a halt, real estate floor space sold and started are decelerating (Chart I-16). Chart I-15Deposit-Money Leads Metals Prices And Construction Chart I-16China: More Signs Of Slowdown Bottom Line: Regardless of which money measure we use, and regardless of their past track record, all of them are currently weak and point to a major and imminent slump in China's growth in the next six to 12 months. This gives us confidence in reiterating our negative view on China plays (including commodities) and EM. Credit Markets Strategy We have been recommending a strategy of shorting/underweighting EM sovereign and corporate credit versus U.S. high-yield (HY) credit and this strategy has shown strong performance, producing 15% gains with low volatility since August 2011 (Chart I-17). However, today we recommend shifting the underweight EM corporate and sovereign credit position from U.S. HY to U.S. investment grade (IG) corporate credit. The primary reason is that credit spreads are extremely tight and odds favor credit spreads widening in both U.S. and EM. Chart I-18 shows that when U.S. TIPS yields rise U.S. IG usually outperforms U.S. HY on an excess return basis. We expect U.S. Treasurys and TIPS yields to grind higher in the near term because U.S. growth and inflation are much stronger than the bond market is currently pricing in. Chart I-17Book Gains On This Strategy Chart I-18Higher U.S. Bond (TIPS) Yields Warrant Rotation Rising U.S. bond yields also warrants EM credit underperformance versus U.S. IG because the EM credit benchmark is riskier than U.S. IG. While the two segments have similar durations, the duration times spread measure of risk is greater for EM credit. Furthermore, U.S. HY spreads have narrowed versus both EM sovereign and corporate spreads since early 2016 (Chart I-19, top panel). Hence, there is little value favoring the former versus EM credit. In contrast, U.S. IG spreads versus both EM sovereign and corporate credit are appealing historically (Chart I-19, bottom panel). Therefore, there is a valuation aspect to this strategy change. Relative spread differences have historically correlated quite well with the subsequent 12-month return. Given where relative spreads are, the subsequent 12-month return for investing in U.S. IG relative EM credit is positive (Chart I-20, top panel) but it is negative for investing in U.S. HY versus EM credit (Chart I-20, bottom panel). Chart I-19EM Credit Offers Value Relative ##br##To U.S. HY But Not Versus U.S. IG Chart I-20Projected Returns Of EM Credit ##br##To Both U.S. IG And HY As to the rationale of favoring U.S. credit to EM credit, this is consistent with our theme that the growth outlook, corporate leverage, and health of the banking system are in much better shape in the U.S. than in EM. Bottom Line: Book profits on the short EM sovereign and corporate credit / long U.S. HY credit position. Institute a new position: short EM sovereign and corporate credit / long U.S. IG corporate credit. Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com 1 Please refer to the Emerging Markets Strategy Weekly Report, titled "Follow The Money, Not The Crowd", dated July 26, 2017, link available on page 18. Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights Duration: An environment characterized by strong global growth and a weak dollar is very bearish for U.S. bonds. According to our model, fair value for the 10-year Treasury yield is 2.6%. Stay at below-benchmark duration. The Fed & The Dollar: A weak dollar eases financial conditions and supports higher core goods inflation. All else equal, this will strengthen the Fed's hawkish resolve in the near term. However, a rebound in core services (excluding shelter and medical care) inflation will be necessary for core inflation to reach the Fed's target on a sustained basis. USD Sovereigns: USD-denominated sovereigns are not attractive compared to domestic Baa-rated U.S. credit. At the country level, Finland, Mexico and Colombia offer the most attractive spreads and Finnish debt offers the best risk/reward trade-off. Feature Please note there will be no U.S. Bond Strategy report next week. Our regular publishing schedule will resume on August 29, 2017. Chart 1Firm Growth, Despite Weaker $ Escalating tension between the U.S. and North Korea captured the market's attention during the past week, causing investors to ignore what in our view is a more important economic development: Global growth has managed to stay firm even in the face of significant dollar depreciation. Not only does this break the pattern of the past few years when periods of substantial dollar weakness were associated with slowing global growth (Chart 1), but in our view it sends a very bearish signal for U.S. bonds. Above all else, a weak dollar amidst strong global growth suggests that the breadth of the economic recovery is improving. This intuition is confirmed by the fact that our Global Manufacturing PMI Diffusion Index, which measures the net percentage of countries with PMIs above the 50 boom/bust line, is fast approaching 90% (Chart 2). Not only that, but PMIs from the four most important economic blocs are all showing signs of strength. Both the Eurozone and Japanese PMIs are holding firm at high levels, while the U.S. and Chinese PMIs have recently reversed their year-to-date downtrends (Chart 2, bottom two panels). Why is the breadth of the global recovery important? Precisely because a more synchronized recovery prevents the dollar from appreciating too quickly. All else equal, a stronger dollar causes investors to reduce their forecasts for future U.S. growth and inflation. This implies a slower expected pace of rate hikes and lower Treasury yields. Conversely, a weaker dollar causes investors to revise up their growth and inflation forecasts, leading to a quicker expected pace of rate hikes and higher yields. To capture the importance of both global growth and the exchange rate we turn to our 2-factor Treasury model (Chart 3). This is a simple model of the 10-year Treasury yield based on the Global PMI and bullish sentiment toward the dollar. A stronger Global PMI pressures the model's fair value higher, as does increasingly bearish dollar sentiment. Chart 2Synchronized Global Growth Chart 310-Year Treasury Yield Fair Value At present, the model pegs fair value for the 10-year Treasury yield at 2.6%, meaning the current 10-year Treasury yield of 2.22% is 38 bps below fair value. This is the most expensive Treasuries have appeared on our model since the immediate aftermath of last year's Brexit vote. Political Uncertainty & Flights To Quality While our 2-factor model does a good job, there is one important driver of Treasury yields it does not capture. That is the tendency for political events to drive a flight to safety into Treasuries (Chart 4). Typically, if it is possible to identify a purely politically-driven flight to safety - one that is unlikely to exert a meaningful economic impact during the next 6-12 months - then the correct strategy is to heed our model's message and position for higher yields. This strategy worked out perfectly following the Brexit vote, and we anticipate it will work again this time around. Chart 4Policy Uncertainty Is A Driver Of Bond Yields With regards to the catalyst for last week's flight to safety, our Geopolitical Strategy service wrote in a recent Special Report1 that a pre-emptive strike by the U.S. on North Korea is extremely unlikely. The theatrics of the past week demonstrate only that the U.S. needs to establish a "credible threat" if it wants to eventually open a new round of negotiations over North Korea - not unlike the Iranian nuclear negotiations of the past decade. Looking further down the road, if those talks eventually fail then the potential for military conflict is high. We therefore conclude that there is not much potential for U.S. / North Korean tensions to exert a meaningful economic impact during the next 6-12 months, and view the recent bond rally as an opportunity to position for sharply higher yields in the near-term. Bottom Line: An environment characterized by strong global growth and a weak dollar is very bearish for U.S. bonds. According to our model, fair value for the 10-year Treasury yield is 2.6%. Stay at below-benchmark duration. How The Fed Views A Weaker Dollar Financial Conditions Chart 5Weak $ Eases Financial Conditions The Fed views the 7% year-to-date depreciation of the dollar as a significant easing of financial conditions. In fact, most broad indicators of financial conditions have eased this year, even though the Fed has lifted rates by 75 bps since December (Chart 5). In the Fed's framework, this means that the pace of rate hikes might need to increase in order to tighten financial conditions as much as desired. New York Fed President William Dudley summed up this approach in a 2015 speech:2 All else equal, if financial conditions tighten sharply, then we are likely to proceed more slowly. In contrast, if financial conditions were not to tighten at all or only very little, then - assuming the economic outlook hadn't changed significantly - we would likely have to move more quickly. In the end, we will adjust the policy stance to support financial market conditions that we deem are most consistent with our employment and inflation objectives. Of course, all else is not equal. Core inflation has disappointed so far this year and our current assessment of monetary policy is that while the Fed will take action to start shrinking its balance sheet next month, rate hikes are on hold until inflation turns higher. We remain optimistic that inflation will show sufficient strength in time for the Fed to lift rates in December.3 Inflation Chart 6Weak $ = Higher Inflation A weaker dollar also increases the Fed's confidence that inflation will head higher. Although so far we have not seen much evidence that this is occurring. Last Friday's July CPI report showed that core CPI rose only 0.1% month-over-month, while the year-over-year growth rate held flat at 1.7%. However, evidence is mounting that core inflation will soon put in a bottom. Our CPI diffusion index bounced back into positive territory in July (Chart 6) and our PCE diffusion index is at its highest level since last October.4 Both of these measures have excellent track records capturing the near-term swings in core inflation. The year-to-date weakness in the dollar has led to a surge in import prices. Stronger import prices will soon translate into higher core goods inflation (Chart 6, panels 2 and 3). Unfortunately, any increase in core goods inflation is unlikely to be sustained beyond the next 12 months. If the year-to-date dollar weakness starts to reverse, as our currency strategists anticipate,5 then import prices will decline anew. Eventually, this will translate into a deceleration in core goods inflation. For core inflation to sustainably reach the Fed's target, improvement in the lagging core services (excluding shelter and medical care) component will be required. Historically, this component is the most tightly linked to wage growth (Chart 6, bottom panel). A Rising Wage Growth Environment Two related methods do an excellent job predicting the direction of wage growth on a cyclical horizon. First, wages accelerate when the unemployment rate is falling, and second, wages accelerate when the prime-age (25-54) employment-to-population ratio is increasing. The top two panels of Chart 7 show the relationship between wage growth and the unemployment rate. The shaded regions in both panels correspond to periods when the unemployment rate is falling. As can be seen, wage growth always rises during these periods. That being the case, we calculate that non-farm employment needs to grow by more than 125k per month (on average) for the unemployment rate to continue its downtrend, assuming the labor force participation rate remains flat. Chart 7A Rising Wage Environment Of course it is not guaranteed that the labor force participation rate will stay flat. In a recent report we discussed the risk that a large cyclical increase in the participation rate might cause the unemployment rate to rise even as the economy continues to recover.6 This is why we also look at the shaded regions in the bottom two panels of Chart 7 and see that wages always rise during periods when the prime-age employment-to-population ratio is rising. By looking at the employment-to-population ratio instead of the unemployment rate we do not need to make an assumption about the trend in labor force participation. Using this method, we calculate that monthly employment growth must exceed 140k (on average) for the prime-age employment-to-population ratio to keep increasing. Non-farm payroll growth has averaged 184k per month so far in 2017 and averaged 187k per month in 2016. In other words, the U.S. jobs machine is running at a fairly steady pace, well above the thresholds we see as necessary for the recovery in wage growth to continue. Bottom Line: A weak dollar eases financial conditions and supports higher core goods inflation. All else equal, this will strengthen the Fed's hawkish resolve in the near term. However, a rebound in core services (excluding shelter and medical care) inflation will be necessary for core inflation to reach the Fed's target on a sustained basis. Sovereigns Not Buying The Weak Dollar USD-denominated sovereign bonds should benefit from a falling dollar. A weaker U.S. dollar makes the debt obligation cheaper in the issuing nation's local currency. However, the USD Sovereign index has actually underperformed the duration-matched Baa U.S. Credit index during the past six months, despite a depreciating U.S. currency (Chart 8). The duration-matched Baa-rated U.S. Credit index is the closest comparable we can find for the Sovereign index. It matches the Sovereign index in terms of duration and average credit rating, although historically it also delivers less excess return volatility (Chart 8, bottom panel). The two main factors we consider when deciding whether to add USD-denominated sovereigns to our portfolio at the expense of domestic U.S. credit are relative valuation and the outlook for the dollar. Historically, spread differential has been an important driver of relative returns. Attractive starting valuations even allowed sovereigns to outperform credit in 2014 and 2015 despite the dollar's surge. But at the moment, relative value is skewed heavily in favor of domestic U.S. credit (Chart 8, panel 1). Chart 8Sovereigns Too Expensive Added to that, with U.S. growth likely to remain strong and U.S. inflation poised to rebound, we think there is a high likelihood that the Fed will deliver more rate hikes than are currently priced in. This will make it difficult for the dollar to decline further from current levels. Taken together, poor relative valuation and a bullish outlook for the dollar lead us to continue underweighting USD-denominated sovereigns in our portfolio. The Sovereign Index: Country Breakdown Even though the overall index is unappealing, opportunities might still exist at the country level. Chart 9 shows a risk/reward picture for each country in the Bloomberg Barclays Sovereign index. The upper panels show the option-adjusted spread for each country relative to its duration and credit rating. The lower panels show a risk-adjusted spread on the y-axis. This risk-adjusted spread is the excess spread that remains after we adjust for differences in credit rating and duration using a cross-sectional model. What sticks out immediately is that Finland, Colombia and Mexico all offer compelling spreads after adjusting for differences in credit rating and duration. The outlook for each country's currency versus the U.S. dollar is obviously also important. And in fact, the lower-right panel of Chart 9 shows that exchange rate volatility is positively correlated with the risk-adjusted spreads from our cross-sectional model. This implies that the extra compensation available in Mexican and Colombian sovereigns is probably compensation for assuming highly volatile currency risk. By this measure, Finland looks even more attractive given the euro's slightly lower volatility. Chart 9USD Sovereign Index: Country Breakdown Bottom Line: USD-denominated sovereigns are not attractive compared to domestic Baa-rated U.S. credit. Remain underweight. At the country level, Finland, Mexico and Colombia offer the most attractive spreads and Finnish debt offers the best risk/reward trade-off. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see Geopolitical Strategy Special Report, "North Korea: Beyond Satire", dated April 19, 2017, available at gps.bcaresearch.com 2 https://www.newyorkfed.org/newsevents/speeches/2015/dud150605 3 For further details on our outlook for the near-term path of monetary policy please see U.S. Bond Strategy Portfolio Allocation Summary, "On Hold, But Not For Long", dated August 8, 2017, available at usbs.bcaresearch.com 4 For a chart of the PCE diffusion index please see page 11 of U.S. Bond Strategy Portfolio Allocation Summary, "On Hold, But Not For Long", dated August 8, 2017, available at usbs.bcaresearch.com 5 Please see Foreign Exchange Strategy Weekly Report, "Fade North Korea, And Sell The Yen", dated August 11, 2017, available at fes.bcaresearch.com 6 Please see U.S. Bond Strategy Weekly Report, "Three Scenarios For Treasury Yields In 2017", dated June 20, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification

Highlights The GOP can bolster its case for re-election in 2018 by passing tax cuts and rolling back regulation. With U.S. equity valuations stretched, prolonged uncertainty in Northeast Asia may be a catalyst for a pullback. The global economic outlook is brightening and will be a tailwind for U.S. economic growth and equities. Rising wage pressure will be another headwind for EPS growth in 2018, although wages appear quite benign at the moment. Wages are not always a good leading indicator for the inflation cycle. Indeed, sometimes upturns in wage growth lags that of consumer prices. Feature Safe haven assets caught a bid last week while risk assets sold off as investors weighed geopolitical tensions in Northeast Asia and more uncertainty over fiscal policy in Washington. Last week's U.S. economic data highlighted the disconnect between a tighter labor market and a lack of wage pressures. Meanwhile, the data suggest that growth outside the U.S. is accelerating. Nonetheless, history shows that investors should be patient while waiting for an upturn in inflation. Next Up: Tax Cuts The GOP will deliver on tax cuts this year despite disarray at the White House and an incompetent Congress, but fiscal stimulus may fail to live up to its hype. Furthermore, a fiscal lift from infrastructure spending is unlikely anytime soon. Republicans need a win ahead of the 2018 mid-term elections and they have already laid the groundwork for tax reform via the budget reconciliation process. Moreover, cutting taxes is easier to justify politically than removing an entitlement program (i.e. Obamacare). Tax rates probably will not be lowered by as much as originally promised because conservative Republicans in the House will demand "revenue offsets" to pay for tax cuts. Internal GOP battles over how to fund tax cuts could spill over into some tension regarding raising the debt ceiling. However, it is in neither political party's interests to create another "fiscal cliff" out of thin air. The GOP needs Democratic votes to pass this legislation in the Senate and the Democratic leadership has indicated it is willing to support it. At what price? House Minority leader Nancy Pelosi and Senate Minority leader Chuck Schumer may link the debt ceiling and spending bill to tax reform, and push for the tax cuts to extend to the middle class and to be revenue neutral. There is a chance that both parties will agree to temporarily eliminate the debt ceiling, perhaps beyond the 2018 mid-term elections. In any event, we expect a last minute resolution to both the U.S. debt ceiling and the potential government shutdown in September. Thus, there should be no lasting impact on financial markets from the debt ceiling debate. Turning to government regulation, the NFIB survey shows that small businesses are pleased with the Trump administration's attack on red tape. President Trump has made progress on slowing regulation and is on track to enact one-tenth the amount of economically significant regulation1 passed by the Obama administration (Chart 1). By this metric, Trump is even more frugal than Reagan. Trump and the GOP-held Congress have rolled back Obama-era rules and delayed others. Still, regulatory change is slow to impact the economy and it may take years for the regulatory rollback to provide any meaningful lift to growth. Accordingly, the "Trump Put"2 is still in place. U.S. politics will remain a mess for much of the year, delaying any progress on populist economic policies that would have buoyed U.S. nominal GDP growth and given the Fed a reason to hike interest rates more aggressively (Chart 2). Chart 1Trump Has Had Success In Slowing Regulation Chart 2The Trump Put Bottom Line: Trump will not be impeached until after the 2018 mid-term election, and only then if the Democrats manage to take control of the House. The GOP can bolster its case for re-election in 2018 by passing tax cuts and rolling back regulation. The intensifying Mueller investigation and White House incompetence will only fuel the "Trump Put", which has been positive for U.S. equities, neutral for Treasuries, and bad for the dollar, all else equal. A significant uptick in inflation could overwhelm the "Trump Put" and spark a dollar rally. As such, investors should focus on inflation prospects rather than on White House politics. Fire And Fury Investors are on high alert and with the Q2 earnings season over, may look beyond the positive news on corporate profits for direction. Our colleagues in the BCA Geopolitical Strategy service have long maintained that Northeast Asia is ripe for economic/political risk.3 The underlying driver of uncertainty on the Korean Peninsula is the Sino-American rivalry. China is an emerging "great power" that threatens the global dominance of the U.S. and its allies. The immediate consequence is mounting friction in China's periphery. That is why Taiwan, the South China Sea, and North Korea, are all heating up. North Korea's regime is highly unpredictable as evidenced by events in the past few weeks. In that sense, it is more significant than the other "proxy battles" between the U.S. and China. In essence, North Korea is no longer merely an object of satire. A new round of negotiations over North Korea's nuclear and missile programs is about to begin. The potential for a military conflict is high unless diplomacy succeeds in convincing North Korea to freeze its weapons programs. The events on the Korean peninsula are unfolding as we expected they would. North Korea has a history of rational action. It wants a nuclear deterrent and a peace treaty, but not a regime change. The U.S. has forsworn regime change as an intention and China has recommitted to new sanctions. South Korea is pro-engagement. Moreover, we are seeing the U.S. establish a credible military as part of the "arc of diplomacy," comparable to U.S.-Iran relations 2010-15. Bottom Line: We do not expect a pre-emptive strike by the U.S. on North Korea, as the constraints to conflict are extremely high and not all diplomatic options have been exhausted. Nonetheless, with U.S. equity valuations stretched, prolonged uncertainty in the region may be a catalyst for a pullback. A Rosy Global Picture The global economic outlook is brightening and will be a tailwind for U.S. economic growth and equities. Global real GDP estimates continue to move higher, a welcome departure from years past when estimates slid relentlessly lower (Chart 3). Since the start of 2017, global GDP estimates for this year have increased from 2.8% to 3%, while 2018 forecasts have accelerated from 2.7% to 2.9%. This upward trajectory has occurred despite a recalibration by many major central banks away from accommodative policies. Aggressive central bank actions or escalating tensions in Northeast Asia, or both, may halt the improving growth forecasts. Falling oil prices would also challenge a quickening of global growth, but our view is that oil prices will move higher in the coming months.4 Chart 3Global Growth Estimates Accelerating Despite Stalled U.S. Growth Global leading indicators are on the upswing (Chart 4). The BCA Global Leading Indicator Index (excluding the U.S.) in July 2017 was the strongest since 2010 when it slowed after a sharp rebound from the global financial crisis. The increase in growth still has room to run. Admittedly, the LEI's diffusion index has dipped below 50%. It would be a warning sign for global growth if the diffusion index does not soon turn up. Nominal global GDP growth is speeding up, boosted by improving consumer and business confidence, rising capital spending and declining policy uncertainty (Chart 5). The global economic surprise index is also climbing, which provides additional support. Investors may be concerned that the global PMIs have peaked (Chart 6), but they remain at levels consistent with above-trend GDP growth and we see no reason why they should drop below 50. Chart 4LEIs Pointing Higher Chart 5Supports For Global Growth In Place Chart 6Global Economic Activity Brightening Industrial production (IP) overseas is expanding nearly twice as fast as in the U.S. (Chart 5). This suggests that U.S. economic activity will be pulled up by foreign demand. A stronger dollar (as much as a 10% appreciation in the next year) may dampen U.S. exports and earnings, but this will be more a problem for 2018 than 2017. Bottom Line: Improving economic activity outside the U.S. is a tailwind for both U.S. economic growth and profits of U.S. firms with significant business abroad. Solid foreign demand will help the economy hit the Fed's GDP target and also support additional, but gradual, tightening by the central bank. Stay overweight U.S. equities and remain short duration. Waiting For Wages Rising wage pressure will be another headwind for EPS growth in 2018, although wages appear quite benign at the moment. Both primary and secondary indicators point to a tighter U.S. labor market. The July jobs report (released in early August) was yet another sign that the slack in the jobs market is vanishing.5 Data released last week on job openings (JOLTS) and the National Federation of Independent Business (NFIB) further supported this trend, and indicated that the labor market may tighten even more. Job openings rose to a new all-time high along with BCA's quit rate less layoffs indicator (Chart 7). The hire rate remained at a cycle peak. The NFIB data was equally impressive, with hiring plans and job openings surging in July. Small businesses are also finding it increasingly difficult to find quality labor. (Chart 7, panel 4) The strength in the labor market has not yet translated into accelerating wages, but patience is required. The July NFIB survey noted that "while a tight job market may point to higher wages and rising consumer spending down the road, which is also good for small businesses, the current expansion efforts by small business owners are being choked by their difficulties in hiring and keeping workers." The NFIB's compensation plans (Chart 7) provided quantitative support for the group's qualitative assessment. However, the latest readings on labor compensation from the Q2 productivity report, the tepid July average hourly earnings data and the Atlanta Fed wage tracker suggest that the labor market is still not tight enough to generate much wage pressure (Chart 8). Chart 7Widespread Evidence That##BR##Labor Market Is Tightening Chart 8Not Much Wage##BR##Pressure Yet Inflation And Long-Expansion Dynamics That said, wages are not always a good leading indicator for the inflation cycle. Indeed, sometimes upturns in wage growth lag that of consumer prices. In previous research we split U.S. post-1950 economic cycles into three sets based on the length of the expansion phase: short (about 2 years), medium (4-6 years) and long (8-10 years). What distinguishes short from medium and long expansions is the speed at which the most cyclical parts of the economy accelerated, and the time it took unemployment to reach a full employment level. Long expansions were characterized by a drawn-out rise in the cyclical parts of the economy and a very slow return to full employment, similar to what has occurred since the Great Recession. Chart 9 compares the current cycle to the average of two of the long cycles (the 1980s and the 1990s). We excluded the long-running 1960s expansion because the Fed delayed far too long and fell well behind the inflation curve. We define the 'late cycle' phase to be the time period from when the economy first reached full employment to the subsequent recession (shaded portions in Chart 9). The average late-cycle phase for these two expansions lasted almost four years, highlighting that reaching full employment does not necessarily mean that a recession is imminent. Inflation pressures are slower to emerge in 'slow burn' recoveries, allowing the Fed to proceed slowly. The Fed waited an average of 25 months to tighten policy after reaching full employment in these two long expansions, in part because core CPI inflation was roughly flat. The result was an extended late-cycle phase that was very rewarding for equity investors because the economy and earnings continued to grow. Of course, inflation eventually did turn higher, signaling the beginning of the end for the expansion and equity bull phase. In Chart 10, we compare the core PCE inflation rate in the current cycle with the average of the previous two long expansion episodes (the inflection point for inflation in the previous cycles are aligned with June 2017 for comparison purposes). The other panels in the chart highlight that, in the 1980s and 1990s, wage growth gave no warning that an inflation upturn was imminent. Indeed, wages were a lagging indicator of consumer price inflation. Chart 9Labor Market, Inflation And Stocks##BR##In The Long 80's & 90's Expansions Chart 10In The 80's & 90's Wage Growth##BR##Gave No Early Warning On On Inflation Market commentators often assume that inflation is driven exclusively by "cost push" effects, such that the direction of causation runs from wage pressure to price pressure. However, causation runs in the other direction as well. Households see rising prices and then demand better wages to compensate for the added cost of living. Chart 11Leading Indicators Of Inflation##BR##In "Slow Burn" Recoveries This is not to say that we should totally disregard wage information. But it does suggest that we must keep an eye on a wider set of data. Indicators that provided some leading information for inflation in the previous two long cycles are shown in Chart 11. To this list we would also add the St. Louis Fed's Price Pressure index, which is not shown in Chart 11 because it does not have enough history. All of these indicators have moved higher over the past 18 months, after bottoming at extremely low levels in 2015 and early 2016. However, they have all pulled back to some extent in recent months. This year's pipeline inflation "soft patch" continued into July, according to last week's release of the Producer Price Index. The easing in cost pressures at the producer level has been broadly based (i.e. one cannot blame special factors). These indicators suggest that consumer price inflation, according to either the CPI or the PCE, will struggle to rise in the next few months. The July CPI report revealed another tepid 0.1% monthly rise in the core price index, while the year-over-year rate remained at 1.7%. Rising prices for health care goods and services were offset by price declines for new and used cars. The diffusion index for the CPI moved up to the zero line in July, indicating that disinflation was a little less broadly based in the month. Bottom Line: Our base case is that core PCE inflation edges higher in the coming months, which will be enough for the FOMC to justify a rate hike in December. We also expect that inflation will be high enough in 2018 for the Fed to hike rates by more than is discounted in the bond market. Nonetheless, the warning signs of an inflation upturn are mixed at best. It would flatter our stocks-over-bonds recommendation if we are wrong on the inflation outlook, but our short duration stance would not be profitable in this case. 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 Office of Information and Regulatory Affairs (OIRA) of Office of Management and Budget (OMB): https://www.reginfo.gov/public/do/eAgendaMain and https://www.reginfo.gov/public/do/eoCountsSearchInit?action=init 2 Please see Geopolitical Strategy Weekly Report, "How Long Can The Trump Put Last" dated June 14, 2017, available at gps.bcaresearch.com. 3 Please see Geopolitical Strategy Weekly Report, "North Korea: Beyond Satire, dated April 18, 2017, available at gps.bcaresearch.com. 4 Please see Commodity & Energy Strategy Weekly Report, "KSA's Tactics Advance OPEC' 2.0's Agenda," dated August 10, 2017, available at ces.bcaresearch.com. 5 Please see U.S. Investment Strategy Weekly Report, "Stay The Course" dated August 7, 2017, available at usis.bcaresearch.com.

Highlights The rise in the yen sparked by the verbal confrontation between the U.S. and North Korea is creating an opportunity to buy USD/JPY. The DXY is set to stabilize and may even rebound, removing a key support for the yen. The U.S. economy is showing signs of strength, and the bond market is expensive, a backup in yields is likely. Rising U.S. bond yields should be poisonous for the yen Until higher bond yields cause an acute selloff in risks assets, an opportunity to buy USD/JPY is in place for investors. Feature After benefiting from the U.S. dollar's generalized weakness, the yen has received a renewed fillip thanks to the rising tensions between North Korea and the U.S. If the U.S. were indeed to unleash "fire and fury" on North Korea, safe-haven currencies like the yen or Swiss franc would obviously shine. While the verbal saber-rattling will inevitably continue, our colleagues Marko Papic and Matt Gertken - head and Asia specialist respectively of our Geopolitical Strategy service - expect neither the U.S. nor North Korea to go to war. Historically, North Korea has behaved rationally, and it only wants to use the nuclear deterrent as a bargaining chip. Meanwhile, the U.S does not want to invest the time, energy, and money required to enact a regime change in that country. Additionally, China is already imposing sanctions on Pyongyang, and Moon Jae-in, South Korea's new president, wants to appease its northern neighbor. With cooler heads ultimately likely to prevail, will the yen rally peter off, or should investors position themselves for additional USD/JPY weakness? We are inclined to buy USD/JPY at current levels. DXY: Little Downside, Potential Upside Most of the weakness in USD/JPY since July 10 has been a reflection of the 3.7% decline in the DXY between that time and August 2nd. However, the dollar downside is now quite limited and could even reverse, at least temporarily. The dollar is currently trading at its deepest discount since 2010 to our augmented interest rate parity model, based on real interest rate differentials - both at the long and short-end of the curve - as well as global credit spreads and commodity prices (Chart I-1). Crucially, the euro, which accounts for 58% of the dollar index, is its mirror image, being now overvalued by two sigma, the most since 2010 (Chart I-2). Confirming these valuations, investors have now fully purged their long bets on the USD, and are most net-long the euro since 2013. Chart I-1DXY Is Cheap... Chart I-2...But The Euro Is Not Valuations are only an indication of relative upside and downside; the macro economy dictates the directionality. While U.S. financial conditions have eased this year, they have tightened in Europe, resulting in the biggest brake on euro area growth relative to the U.S. in more than two years (Chart I-3). This is why euro area stocks have eradicated their 2017 outperformance against the S&P 500, why PMIs across Europe have begun disappointing, and why the euro area economic surprise index has rolled over - especially when compared to that of the U.S. The improvement in U.S. economic activity generated by easing financial conditions also has implications for the dollar. As Chart I-4 illustrates, the gap between the U.S. ISM manufacturing index and global PMIs has historically led the DXY by six months or so. This gap currently points to a sharp appreciation in the dollar. Chart I-3Easing Versus Tightening FCI Chart I-4PMIs Point To USD Rally If the dollar were indeed to stop falling, let alone appreciate, this would represent a hurdle for the yen to overcome, especially as the outlook for U.S. bond yields is pointing up. Bottom Line: Before North Korea grabbed the headlines, the USD/JPY selloff was powered by a weakening dollar. However, the dollar has limited downside from here. It is trading at a discount to intermediate-term models, while macroeconomic momentum is moving away from the euro area and toward the U.S. - a key consequence of the tightening in European financial conditions vis-à-vis the U.S. Additionally, the strong outperformance of the U.S. ISM relative to the rest of the world highlights that the dollar may even be on the cusp of experiencing significant upside. The Key To A Falling Yen: Treasury Yields Upside An end to the fall in the USD is important to end the downside in USD/JPY. However, rising Treasury yields are the necessary ingredient to actually see a rally in this pair. We are optimistic that U.S. bond yields can rise from current levels. The U.S. job market remains very strong. The JOLTS data this week was unequivocal on that subject. Not only are there now 6.2 million job openings in the U.S., but the ratio of unemployed to openings has hit its lowest level since the BLS began publishing the data, suggesting there is now a limited supply of labor relative to demand. Additionally, the number of unfilled jobs is nearly 30% greater than it was at its 2007 peak, pointing to an increasingly tighter labor market. We could therefore see an acceleration in wage growth going into the remainder of this business cycle, even if structural factors like the "gig-economy", the increasing role of robotics, or even the now-maligned "Amazon" effect limit how high wage growth ultimately rises. The Philips curve, when estimated using the employment cost index and the level of non-employment among prime-age workers, still holds (Chart I-5). Thus, a tight labor market in conjunction with continued job-creation north of 100,000 a month should put upward pressure on wages. Even when it comes to average hourly earnings, glimmers of hope are emerging. Our diffusion index of hourly wages based on the industries covered by the BLS cratered when wage growth slowed over the past year. However, it has hit historical lows and is beginning to rebound - a sign that average hourly earnings should also reaccelerate (Chart I-6). Chart I-5The Philips Curve Still Works Chart I-6Even AHE Are Set To Re-Accelerate The job market is not the only source of optimism, as U.S. capex should continue to be accretive to growth. Despite vanishing hopes of aggressive deregulation, the NFIB small business survey picked up this month. Even more importantly, various capex intention surveys as well as the CEO confidence index point to continued expansion of corporate investment (Chart I-7). Healthy profit growth is providing both the necessary signal and the source of funds to engage in this capex. This will continue to lift the economy. This is essential to our bond and our yen views, as it points to higher U.S. inflation. In itself, economic activity is not enough to generate higher prices. However, when this happens as aggregate capacity utilization in the economy is becoming tight, inflation emerges. As Chart I-8 shows, today, our composite capacity utilization indicator - based on both labor market conditions and the traditional capacity utilization measure published by the Federal Reserve - is in "no-slack" territory, a condition historically marked by bouts of inflation. Chart I-7U.S. Capex To Boost Growth Further Chart I-8No Slack Plus Growth Equals Inflation The recent increase to a three-year high in the "Reported Price Changes" component of the NFIB survey corroborates this picture, also pointing to an acceleration in core inflation (Chart I-9). But to us, the most telling sign that inflation will soon re-emerge is the behavior of the U.S. velocity of money. For the past 20 years, changes in velocity - as measured by the ratio of nominal GDP to the money of zero maturity - have lead gyrations in core inflation, reflecting increasing transaction demand for money. Today, the increase in velocity over the past nine months points to a rebound in core inflation by year-end (Chart I-10). Chart I-9The Pricing Behavior Of Small Businesses ##br##Points To An Inflation Pick Up Chart I-10Reaching Escape ##br##Velocity Expecting higher inflation is not the same thing as expecting higher interest rates and bond yields. However, we believe this time, higher inflation will result in higher yields. First, the Fed wants to push interest rates higher. Fed Chairwoman Janet Yellen and her acolytes have been very clear about this, with the "dot plot" anticipating rates to rise to 2.9% by the end of 2019. While the Fed's preference and reality can be at odds, this is currently not the case. Our Fed monitor continues to be in the "tighter-policy-needed" zone. While it is undeniable that it is doing so by only a small margin, higher inflation - as we expect - would only push this indicator higher. Moreover, the diffusion index of the components of the Fed monitor is already pointing toward an improvement in this policy gauge (Chart I-11). Chart I-11The Fed Monitor Will Pick Up Second, the Fed may have increased rates, and the spread between U.S. policy rates and the rest of the world may have widened, but the dollar has weakened this year. This counterintuitive result highlights that the Fed's effort has had little impact in tightening liquidity conditions. In fact, as we have mentioned, because of the lower dollar and higher asset prices, financial conditions have eased, suggesting liquidity remains plentiful. As such, like in 1987 or 1994, this is only likely to re-invigorate the Fed in its confidence that it can hike rates further, as liquidity conditions remain massively accommodative. Third, beyond the Fed's reaction function, what also matters are investors' expectations. At the time of writing, investors only expect 45 basis points of rate hikes over the upcoming 24 months, which is a reasonable expectation only if inflation does not move back toward the Fed's 2% target. However, our work clearly points toward higher inflation by year end. In a fight between the Fed's "dot plot" and the OIS curve, right now, we would take the side of the Fed. Fourth, it is not just 2-year interest rate expectations that seems mispriced, based on our view on U.S. growth, inflation, and the Fed. U.S. Treasury yields are also trading at a 36 basis points discount to the fair-value model developed by our U.S. Bond Strategy sister service (Chart I-12). Continued good news on the job front and an uptick in inflation would likely do great harm to Treasury holders. Finally, the oversold extreme experienced by the U.S. bond market in the wake of the Trump victory has been purged. While we are not at an oversold extreme, our Composite Technical Indicator never punched much into overbought territory during the Fed tightening cycle from 2004 to 2006 (Chart I-13). Moreover, with no more stale shorts, an upswing in U.S. economic and inflation surprises should help put upward pressure on U.S. bond yields. Confirming the intuition laid out above, the copper-to-gold ratio, a measure of growth expectations relative to reflation, has now broken out - despite the North Korean risks. In the past, such a development signaled higher yields (Chart I-14). With this in mind, let's turn to the yen itself. Chart I-12U.S. Bonds Are##br## Too Expensive Chart I-13Stale Shorts Have Been Purged, ##br##But Overbought Conditions Are Unlikely Chart I-14Where The Copper-To-Gold Ratio Goes, ##br## So Do Bond Yields Bottom Line: The U.S. economy looks healthy. The labor market is strong, and capex continues to offer upside. Because capacity utilization is tight and money velocity is accelerating, inflation should begin surprising to the upside through the remainder of 2017. With the market pricing barely two more hikes over the course of the next 24 months and U.S. bonds trading richly, such an economic backdrop should result in higher U.S. bond yields. Yen At Risk, Even If Volatility Rises JGB yields have historically displayed a low beta to global bond yields. As a result, when global bond yields rise, the yen tends to weaken. USD/JPY is particularly sensitive to yield upswings driven by actions in the Treasury market. This contention is even truer now than it has been. The Bank of Japan is targeting a fixed yield curve slope and does not want to see JGB yields rise much above 10 basis points. With the paucity of inflation experienced by Japan - core-core inflation is in a downtrend, ticking in at zero, courtesy of tightening financial conditions on the back of a stronger yen - this policy remains firmly in place. Emerging signs of weakness in Japan highlight that the BoJ is likely to remain wedded to this policy, even as Shinzo Abe's popularity hits a low for his current premiership. The recent fall in the leading indicator diffusion index suggests that industrial production - which has been a bright spot - is likely to roll over in the coming months (Chart I-15). This means the improvement in capacity utilization will end, entrenching already strong deflationary pressures in Japan. This only reinforces the easing bias of the BoJ, and truncates any downside for Japanese bond prices. Chart I-15The Coming Japanese IP Slowdown In short, while JGB yields might still experience some downside when global yields fall, they will continue to capture none of the potential upside. This makes the yen even more vulnerable to higher Treasury yields than it was before. Hence, based on our view on U.S. inflation and yields, USD/JPY is an attractive buy at current levels. But what if the rise in U.S. bond yields causes a correction in risk assets, especially EM ones? Again, monetary policy differences and the trend in yields will dominate. As Chart I-16 illustrates, USD/JPY has a much stronger correlation with dynamics in the bond markets than it has with EM equity prices. Chart I-16Yen: More Like Bonds Than Anything Else Chart I-17USD/JPY Falls Only When EM Selloffs Are So Acute That They Cause Bond Rallies Moreover, as the experience of the past three years illustrates, only once EM selloffs become particularly acute does USD/JPY weaken (Chart I-17). Essentially, the EM selloff has to be so severe that it threatens the Fed's ability to tighten policy, and therefore causes U.S. bond yields to fall. It is very possible that a rise in Treasury yields will ultimately generate this outcome, but in the meantime the rise in U.S. bond yields should create a tradeable opportunity to buy USD/JPY. Bottom Line: With Japan still in the thralls of deflation and the BoJ committed to fight it, JGB yields have minimal upside. Therefore, higher Treasury yields are likely to do what they do best: cause USD/JPY to rally. This might ultimately lead to a selloff in EM stocks, but in the meanwhile, a playable USD/JPY rally is likely to emerge. Thus, we are opening a long USD/JPY trade this week. Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com Currencies U.S. Dollar Chart II-1USD Technicals 1 Chart II-2USD Technicals 2 The U.S. labor market continues to strengthen, with the JOLTS Survey's Job Openings and Hires both ticking up. The NFIB Survey also shows signs of strength as the Business Optimism Index steadied at lofty levels, coming in at 105.2. Unit labor costs disappointed, but this supports U.S. equities. Nonfarm productivity also outperformed, pointing to improving living standards. U.S. data has turned around, with data surprises improving relative to the euro area. These dynamics are likely to prompt a resumption of the greenback's bull market. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Look Ahead, Not Back - June 9, 2017 The Euro Chart II-3EUR Technicals 1 Chart II-4EUR Technicals 2 Euro area data has been mixed: German current account underperformed, with both exports and imports contracting on a monthly rate, and underperforming expectations. The trade balance, however, outperformed; German industrial production failed to meet expectations, even contracting on a monthly basis; Italian industrial production outperformed both on a monthly and yearly rate, but remains well below capacity European data has begun to show the pain inflicted by tightening financial conditions. Relative to the U.S., the economic surprise index has rolled over. If this trend continues, EUR/USD will struggle to appreciate more this year, and may even weaken if U.S. inflation can improve. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 Who Hikes Next? - June 30, 2017 The Yen Chart II-5JPY Technicals 1 Chart II-6JPY Technicals 2 Recent data has been negative in Japan: Labor cash earnings yearly growth went from 0.6% in May to a contraction of 0.4% in June, underperforming expectations. Machinery orders yearly growth fell down sharply, contracting at a 5.2% rate and underperforming expectations. The Japanese economy continues to show signs of weakness, which means that the Bank of Japan will not let 10-year JGB yields rise above 10 basis points. In an environment of rising U.S. bond yields this will cause the yen to fall. However the question remains: Could a selloff in EM prompted by a rising dollar help the yen? This should not be the case, at least for now, as the yen is much more correlated with U.S. bond yields than it is with EM stock prices. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 A Market Update: June 23, 2017 British Pound Chart II-7GBP Technicals 1 Chart II-8GBP Technicals 2 Recent data in the U.K. has been mixed: BRC like-for-like retail sales yearly growth came in at 0.9%, outperforming expectations. However, the RICS Hosing Price Balance - a crucial bellweather for the British economy - came in at 1%, dramatically underperforming expectations. Also, the trade balance underperformed expectations, falling to a 12 billion pounds deficit for the month of June as exports sagged. As we mentioned on our previous report, we expect the pound to suffer in the short term, as the high inflation produced by the fall in the pound following the Brexit vote is starting to weigh on consumers. Furthermore, house prices are also suffering, and could soon dip into negative territory. All of these factors will keep the BoE off its hawkish rhetoric for longer than priced by the markets. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 Australian Dollar Chart II-9AUD Technicals 1 Chart II-10AUD Technicals 2 AUD gains are reversing as the U.S. dollar rebounds from a crucial support level. This has also occurred due to mixed Chinese and Australian data: Chinese trade balance beat expectations, however, both exports and imports underperformed; Chinese inflation underperformed expectations; Australian Westpac Consumer Confidence fell to -1.2% from 0.4% in August; This is largely in line with our view that the rally in AUD was would only create a better shorting opportunity. Underlying structural and fundamental issues will remain a headwind for the AUD for the remainder of the year. Iron ore inventories in China are also at an all-time high, which paints a dim picture for Australian mining and exports going forward. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 Who Hikes Next? - June 30, 2017 New Zealand Dollar Chart II-11NZD Technicals 1 Chart II-12NZD Technicals 2 On Wednesday, the RBNZ left their Official Cash Rate unchanged at 1.75%. Overall, the bank signaled that it will continue its accommodative monetary policy for "a considerable period of time". Furthermore the RBNZ's outlook for inflation, specifically tradables inflation, remains weak. Finally, the bank also showed concern for the rise in the kiwi, stating that "A lower New Zealand Dollar is needed to increase tradables inflation and help deliver more balanced growth". Overall, we continue to be positive on the kiwi against the AUD. While the outlook for tradable-goods inflation might be poor, this is a variable determined by the global industrial cycle.. Being a metal producer, Australia is much more exposed to these dynamics than New Zealand, a food producer. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 Who Hikes Next? - June 30, 2017 Canadian Dollar Chart II-13CAD Technicals 1 Chart II-14CAD Technicals 2 Data continues to look positive for Canada: Housing Starts increased by 222,300, beating expectations; Building permits also increased at a monthly pace of 2.5%, also beating expectations. CAD has experienced some downside as the stretched long positioning that emerged in the wake of the BoC's newfound hawkishness are being corrected. While we expect the CAD to outperform other commodity currencies, based on rate differentials and oil outperformance, USD/CAD should is likely to trend higher as U.S. inflation bottoms. EUR/CAD should trend lower by the end of this year as euro positioning reverts. As a mirror image, CAD/SEK may appreciate based on the same dynamics. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Bad Breadth - July 7, 2017 Who Hikes Next? - June 30, 2017 Swiss Franc Chart II-15CHF Technicals 1 Chart II-16CHF Technicals 2 Last week we highlighted the possibility of a correction in EUR/CHF, given that it had reached highly overbought levels. This prediction turned out to be accurate, as EUR/CHF fell by almost 2% this week, as tensions between North Korea and the United States continue to escalate. Meanwhile on the economic front, Switzerland continues to show a tepid recovery: Headline inflation went from 0.2% in June to 0.3% in July, just in line with expectations. The unemployment rate continues to be very low at 3.2%, also coming in according to expectations. Inflation, house prices and various economic indicators are all ticking up, however, the economic recovery is still too weak to cause a major shift in monetary policy. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 Norwegian Krone Chart II-17NOK Technicals 1 Chart II-18NOK Technicals 2 The krone has fallen this week against the U.S. dollar, even as oil prices have remained relatively flat. This highlights a key theme we have mentioned before: USD/NOK is more sensitive to rate differentials than it is to oil prices. We expect these rate differentials to continue to widen, as the Norwegian economy remains weak, and inflation will likely remain below the Norges Bank target in the coming years. On the other hand, U.S. yields are set to rise, as a tight labor market will eventually lift wages higher and thus increase rate expectations. Meanwhile EUR/NOK, which is much more sensitive to oil prices than USD/NOK, will keep going down, as inventory drawdowns caused by the OPEC cuts should continue pushing up Brent prices. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 A Market Update: June 23, 2017 Swedish Krona Chart II-19SEK Technicals 1 Chart II-20SEK Technicals 2 Data in Sweden was mixed: New Orders Manufacturing yearly growth fell from 7.3% to 4.4%. Industrial production yearly growth increased from 7.5% in May to 8.5% in June, outperforming expectations. The Swedish economy continues to exhibit signs of strong inflationary pressures. Overall we continue to be bullish on the krona, particularly against the euro, as the exit of Stefan Ingves at the end of this year should give way for a more hawkish governor, who would respond to the strength in the economy with a more hawkish stance. Report Links: Balance Of Payments Across The G10 - August 4, 2017 Who Hikes Next? - June 30, 2017 Bloody Potomac - May 19, 2017Xx Trades & Forecasts Forecast Summary Core Portfolio Closed Trades

Highlights Strong corporate earnings growth will drown out worries about North Korea. Stay cyclically overweight global equities. Underlying wage growth in the U.S. is stronger than the official data suggest. Surveys point to a further acceleration in U.S. wages, as do pay gains at the lower end of the income distribution. Labor's share of income will resume its cyclical recovery. This will lead to more consumer spending, and ultimately, higher price inflation. Wage growth elsewhere in the world will also pick up as labor slack declines. Global fixed-income investors should underweight duration and increase exposure to inflation-linked securities. Feature Focus On Corporate Earnings, Not Korea Chart 1EPS Estimates Have Remained ##br##Resilient This Year Global equities dropped over the past few days on the back of rising risks of conflict in the Korean peninsula. Our geopolitical strategists believe that neither the U.S. nor North Korea will launch a preemptive strike.1 Despite its bluster, North Korea has a history of rational action. It wants a nuclear deterrent and a peace treaty. The U.S. has forsworn regime change as a policy goal. China has recommitted to new sanctions and the South is pro-engagement. This raises the likelihood that a diplomatic solution will be found. Unfortunately, getting from here (open hostilities) to there (negotiated solution) will take time, which leaves the door open to increased market volatility. Nevertheless, we expect any selloff to be short-lived, owing to the positive earnings picture. More than anything else, strong profit growth has underpinned the cyclical bull market in stocks, and we expect this to remain the case over the coming months. More than 80% of S&P 500 companies have reported Q2 results. Based on these preliminary numbers, EPS appears to have increased by 11% over the previous year, marking the fourth consecutive quarter of margin expansion. The strength has been broad based, with all eleven sectors reporting positive growth. U.S. earnings estimates for both 2017 and 2018 have remained steady since January, bucking the historic pattern of downward revisions throughout the course of the year (Chart 1). The picture is even more impressive outside the U.S., where earnings estimates continue to move higher. The Euro STOXX 600 is now expected to deliver EPS growth of 12.6% this year. EPS of stocks listed on the Japanese Topix is expected to rise 14.8% this year and 7.3% next year, giving them an attractive 2018E P/E of 13.6. We recommend overweighting euro area and Japanese stocks over their U.S. counterparts in currency-hedged terms. EM stocks have seen the strongest positive earnings revisions this year. We continue to worry about some of the structural headwinds facing emerging markets (high debt levels, poor governance, etc.). However, the cyclical picture remains more upbeat. Chinese H-shares remain our favorite EM market, trading at just 7.5 times 2017 earnings estimates. The U.S. Labor Market Gets A JOLT, But Where's The Wage Growth? The Job Openings and Labor Turnover Survey (JOLTS) released on Tuesday provided more good news about the state of the U.S. labor market (Chart 2). The number of job openings rose to 6.2 million in June. There are now 28% more unfilled jobs in the U.S. than at the prior peak in April 2007. The number of unemployed workers per job opening fell to 1.1, the lowest level in the history of the series. One might think that with numbers like these, wage growth would be skyrocketing. Yet, it is not. While monthly average hourly wages did surprise to the upside in the June payrolls report, the year-over-year change remained stuck at 2.5%. This week's productivity report showed that compensation per hour increased by only 1% in Q2 relative to the same period in 2016. Other measures of wage growth generally point to some softening this year (Chart 3). Chart 2More Good News For The U.S. Labor Market Chart 3U.S. Wage Growth Remains Soft Many commentators regard the lackluster pace of wage inflation - coming at a time when the unemployment rate has fallen below its 2007 lows - as a "mystery" that needs to be solved. As we argue in this report, there is less to this mystery than meets the eye. Properly measured, underlying wage growth in the U.S. has been rising for some time, and may actually be stronger than the "fundamentals" warrant. Wage inflation elsewhere in the world is more subdued. However, this is largely because progress towards restoring full employment has been slower outside the U.S. Is Wage Growth Being Mismeasured? How can U.S. wage growth be characterized as "strong" when it is still so weak by historic standards? Part of the answer has to do with that old bugbear: measurement error. Low-skilled workers have been re-entering the labor force en masse over the past few years, after having deserted it during the Great Recession. This has put downward pressure on average wages, arithmetically leading to slower wage growth. Most of the official wage series, including the Employment Cost Index, do not adjust for this statistical bias.2 In a recent research report, economists at the San Francisco Fed concluded that "correcting for worker composition changes, wages are consistent with a strong labor market that is drawing low-wage workers into full-time employment."3 In addition to cyclical factors, demographic shifts have depressed official measures of wage inflation. Historically, population aging has pushed up average wages because older workers tend to earn more than younger ones. The retirement of millions of well-paid baby boomers over the past few years has reversed this trend, at least temporarily. Chart 4 shows that the median age of employed workers has fallen for the past three years, the first time this has happened since the 1970s. Weak Productivity Growth Dragging Down Wages Unfortunately, there is more to the story than measurement error. Today's young workers are not better skilled or educated than those of previous generations. This, along with other factors that we have discussed extensively in past reports, has dragged down productivity growth.4 Nonfarm productivity has increased at an average annualized pace of less than 1% over the past few years, down from 3% in the early 2000s (Chart 5). Slower productivity growth gives firms less scope to raise wages. In fact, for all the talk about how wages are stagnant, real wages have risen by more than productivity since 2014. This has pushed labor's share of income off its post-recession lows. Chart 4Median Age Of Workers No Longer Rising Chart 5Real Wages Have Increased Faster ##br##Than Productivity Over The Past Few Years It remains to be seen whether the structural downtrend in the share of income going to labor will be reversed. One can make compelling arguments for both sides of the issue.5 But over a cyclical horizon of one-to-two years, it is highly likely that labor's share will rise. Labor's share of income is fairly procyclical. It increased significantly in the late 1990s and rose again in the years leading up to the Great Recession. Considering how low unemployment is today, it is not unreasonable to assume that it will maintain its cyclical uptrend. If so, this will lead to more consumer spending, and ultimately, higher inflation. Surveys Point To Faster Wage Growth... Surveys such as those conducted by the National Federation of Independent Business, Duke University/CFO Institute, National Association for Business Economics, and various regional Federal Reserve banks suggest that employers are becoming increasingly willing to raise compensation in order to fill vacancies (Chart 6). Workers, in turn, are becoming more choosy. This can be seen in an improving assessment of job availability and a rising quits rate. Both of these measures lead wage growth (Chart 7). Chart 6ASurveys Show Employers More Willing To Raise Compensation Chart 6BSurveys Show Employers More Willing To Raise Compensation Chart 7Workers Are Feeling More Confident ...As Do Wage Gains Among Low-Income Workers Median weekly earnings of low-income workers have accelerated this year, even as wage gains among higher-income workers have hit an air pocket (Chart 8). For example, restaurant workers have seen pay hikes of nearly 5% this year, up from 1% in 2014. Wage growth among lower-income workers tends to be less noisy than for higher-income workers. The incomes of better-paid workers are often influenced by bonuses and other variables that may be driven more by industry-specific or economy-wide profit trends rather than labor slack per se. Less-skilled workers are usually the first to get fired and the last to get hired. Thus, wage pressures at the lower end of the skill distribution often coincide with an overheated labor market. This makes the trend in lower-income wages a more reliable gauge of underlying labor market slack. Wage Inflation Will Slowly Pick Up As Global Slack Diminishes We expect U.S. wage growth to rise over the next few quarters by enough to allow the Fed to raise rates in line with the dots. However, a more rapid acceleration - one that forces the Fed to raise rates aggressively - is improbable, at least over the next 12 months. This is mainly because the relationship between domestic labor market slack and wage growth is not as tight as it once was. Trade unions have less clout these days, which means it takes longer for a tight labor market to produce larger negotiated pay hikes. The labor market has also become less fluid, as evidenced by the structural decline in both the rate of job creation and job destruction (Chart 9). Wages tend to adjust more slowly when there is less hiring and firing going on. Chart 8Better Pay For Low-Wage Earners: ##br##A Sign Of A Tighter Labor Market Chart 9Structural Declines In Job Creation##br## And Destruction Perhaps most importantly, an increasingly globalized workforce has given firms the ability to move production abroad in response to rising wages at home. This suggests that wage growth in the U.S. is unlikely to increase significantly until falling unemployment begins to push up wages abroad. Wage Growth Around The World For now, wage growth in America's trading partners remains subdued. Euro area wage inflation is stuck between 1% and 1.5%, although with important regional variations (Chart 10). Wage inflation has accelerated to over 2% in Germany, but is still close to zero in Italy and Spain. Considering that unemployment in both countries remains well above pre-recession levels, it will be difficult for the ECB to tighten monetary policy to any great degree over the next few years. Japanese wage growth has picked up since 2010, but is still below the level consistent with the BoJ's 2% inflation target (Chart 11). Wage inflation is likely to ratchet higher over the next few years, now that the ratio of job openings-to-applicants has risen to the highest level since 1974 (Chart 12). In a sign of the times, Yamato Transport, Japan's largest parcel delivery company, recently told Amazon that it would not be able to make same-day deliveries due to a shortage of available drivers. Chart 10Euro Area Wage Growth Remains ##br##Weak Outside Of Germany Chart 11Modest Pickup In Japanese Wages Wage growth in Canada has actually declined since 2014. However, that is likely to change given that the unemployment rate has fallen close to nine-year lows. Falling unemployment rates should also boost wage inflation in the U.K., Australia, and New Zealand. Chinese wage growth also remains brisk. Chart 13 shows that urban household future income confidence has picked up notably of late, as growth has improved and the labor market has tightened. Chart 12Job Openings Ratio Will Push Wages Higher Chart 13Optimism Over The Labor Market In China Faster Wage Growth Will Ultimately Lead To Higher Inflation Chart 14The Decline In Inflation Expectations ##br##Have Weighed On Wage Growth Going forward, the combination of falling labor slack abroad and an overheated labor market at home will cause U.S. wage inflation to increase more rapidly starting in the second half of 2018. This will be a break from the past. Lower longer-term inflation expectations have tempered nominal wage growth over the past eight years (Chart 14). Both market-based inflation expectations and inflation expectations 5-to-10 years out in the University of Michigan's survey have fallen by about half a point since the financial crisis. The recent decline in headline CPI inflation from 2.7% in February to 1.6% in June may also explain why wage growth has dipped this year even as payroll gains have rebounded. Rising wage growth could begin to feed on itself. As we have discussed before, the Phillips curve tends to steepen once an economy reaches full employment (Chart 15). If the unemployment rate falls from 7% to 6%, this is unlikely to have a huge effect on wages. But if it falls from 4.5% to 3.5%, the effect could be substantial. A recent Fed paper concluded that "evidence strongly suggests a non-linear effect of slack on wage growth and core PCE price inflation that becomes much larger after labor markets tighten beyond a certain point."6 The implication is that once inflation does start rising, it could rise more quickly than investors (or the Fed) expect. Concluding Thoughts The past three U.S. recessions were all caused by the unravelling of financial sector and asset market excesses: The housing bust lay the groundwork for the Great Recession; the collapse of dotcom stocks ushered in the 2001 recession; and the failure of hundreds of banks during the Savings and Loan crisis paved the way for the 1990-91 recession. Unlike the last few recessions, the next one may end up being more akin to those of 1960s, 70s, and 80s. Those earlier recessions were generally triggered by aggressive Fed rate hikes in the face of an overheated economy and rising inflation (Chart 16). Chart 15The Phillips Curve Appears To Be Non-Linear Chart 16Are We Heading Towards A "Retro-Recession"? The good news is that neither wage nor price inflation is likely to soar over the next 12 months. This means that the bull market in global equities can continue for a while longer. The bad news is that complacency about inflation risk is liable to cause central bankers to fall increasingly behind the curve. Rising inflation will force the Fed to pick up the pace of rate hikes in the second half of 2018. This is likely to lead to a stronger dollar and higher Treasury yields. The resulting tightening in U.S. financial conditions could trigger a recession in 2019 or 2020. Investors should remain overweight risk assets for now, but prepare to scale back exposure next summer. Peter Berezin, Global Chief Strategist Global Investment Strategy peterb@bcaresearch.com 1 Please see Geopolitical Strategy Special Report titled "North Korea: Beyond Satire," dated April 19, 2017. 2 Unlike the widely followed average hourly wage series published every month in the payrolls report, the quarterly Employment Cost Index (ECI) does control for shifts in the weights of different industries in total employment. Thus, an increase in the relative number of low-paid hospitality workers would depress average hourly wages, but would not affect the ECI. Nevertheless, the ECI does not control for the possibility that the composition of the workforce within industries may change over time. The Atlanta Fed's Wage Tracker does overcome this bias because it uses the same sample of workers from one period to the next. However it, too, is subject to a number of methodological problems. 3 Mary C. Daly, Bart Hobijn, and Benjamin Pyle, "What's Up with Wage Growth?" FRBSF Economic Letter 2016-07 (March 7, 2016). 4 Please see Global Investment Strategy Special Report, "Weak Productivity Growth: Don't Blame The Statisticians," dated March 25, 2016; and The Bank Credit Analyst Special Report, "Taking Off The Rose-Colored Glasses: Education and Growth In The 21st Century," February 24, 2011. 5 Please see Global Investment Strategy Special Report, "Is Slow Productivity Growth Good Or Bad For Bonds?" dated May 31, 2017; and The Bank Credit Analyst Special Report, "Rage Against The Machines: Is Technology Exacerbating Inequality?" dated June, 2014. 6 Jeremy Nalewaik, "Non-Linear Phillips Curves With Inflation Regime-Switching," Federal Reserve Board, Finance and Economics Discussion Series 2016-078 (August 2016). Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Special Report

Feature There have been two major milestones in China's financial market liberalization in recent months. In June, MSCI Inc. moved to include Chinese domestic A shares in its widely followed world and emerging market equity indices. In July, regulators in Hong Kong and on the Mainland jointly launched the "bond connect" program, allowing foreign investors easier access to China's massive onshore bond market.1 The immediate impact of these measures will likely be muted, but they mark China's continued efforts to deregulate capital account transactions, opening up Chinese domestic financial assets that a mere few years ago were still completely isolated from the rest of the world. Over the years, we have published and periodically updated our Research Note, "China Shop," as a practical guide for investors looking for exposure to Chinese assets. The guide has come a long way since its first edition more than a decade ago, when investing in China was extremely difficult and very limited for foreigners, and we were struggling to find the best "China play" proxies. Over the years, various indexes, tracker funds and derivatives have been established outside China, making investing in Chinese equities a lot easier and more straightforward. The China ETF universe not only covers broad market indexes but also specific sectors and different market caps, allowing for discretionary sector allocations and investment styles for China-focused portfolios (Box 1). Box 1 A Primer On Chinese Stocks A shares are stocks traded on the Shanghai and Shenzhen stock exchanges. These shares are denominated and traded in RMB, and are restricted to local investors and Qualified Foreign Institutional Investors (QFII). B shares are Chinese companies traded on the Shanghai and Shenzhen stock exchanges. This equity class was originally open to foreign investors only, but was made available to domestic investors in 2001. These stocks are denominated in the Chinese currency but traded in U.S. dollars on the Shanghai Stock Exchange and in Hong Kong dollars on the Shenzhen Stock Exchange. H shares are mainland-registered state-owned companies listed in Hong Kong and denominated in Hong Kong dollars. The term N shares refers to stocks listed on the New York Stock Exchange. Red Chips are stocks listed on the Hong Kong Exchange. These companies are usually domiciled outside China but have at least 30% of their stakes held by state-owned organizations or provincial and municipal governments of China. P Chips refer to shares of companies which are majority-owned by entrepreneurs from China and derive the bulk of their revenues in the mainland. These companies are typically incorporated in offshore tax havens and are listed in Hong Kong and other major exchanges outside of China. Since our last update a year ago, the China ETF universe that we've been tracking has continued to evolve, with a few interesting developments. The number of ETFs on our list witnessed the first decline since it was created about 10 years ago. Two new ETFs have been added to the list since our last update, but 16 have been suspended or de-listed (Appendix Below). This means the Chinese ETF boom in recent years has entered a period of "consolidation." It also means that global investors' appetite for Chinese assets has been rather weak. Investors' weak appetite for Chinese assets is also reflected in the constant net withdrawals from these China-related ETFs - a remarkable development considering the sharp rally in Chinese equities, both domestic and investable, since early 2016. Total assets under management (AUM) of these ETFs have increased slightly so far this year compared with a year ago. However, the increases have been entirely due to price increases (Chart 1). Indeed, net capital flows have constantly been negative since 2013, according to our calculations. Investors' lukewarm attitude toward Chinese ETFs stands in stark contrast to other EM bourses. AUMs of EM equity ETFs have been chasing the market rally to new records of late (Chart 2). It appears that investors, especially smaller retail investors, have remained highly uncomfortable with China's macro conditions, despite improving growth figures, and have been left out of the bull market. This could be a contrarian sign that Chinese equities are underweighted and under owned - confirmed by depressed equity multiples. Chart 1Constant Negative Fund Flows To China ETFs Chart 2China ETFs: Out Of Favor Looking forward, the Chinese ETF universe will continue to expand, and the recent market liberalization efforts will likely lead to increasing supplies of ETFs focused on the Chinese onshore bond market. Despite cyclical swings in both economic growth and financial markets, it is almost a sure bet that foreign ownership in Chinese assets will grow over time. Yan Wang, Senior Vice President China Investment Strategy yanw@bcaresearch.com 1 Please see China Investment Strategy Weekly Report, "Embracing Chinese Bonds," dated July 6, 2017, available at cis.bcaresearch.com. Appendix Broad Market By Market Cap - A Share By Market Cap - Investible By Sector - A Share By Sector - Investible Leveraged Plays Currency Fixed Income - Mainland Fixed Income - Offshore Cyclical Investment Stance Equity Sector Recommendations

Highlights The Kingdom of Saudi Arabia (KSA) is taking a well-timed tactical decision to make room for increased Libyan and Nigerian output, by reducing allocations to refiners by more than 500k b/d in September. The bulk of these reductions will be directed at U.S. refiners, which are running their units at close to record output, while reducing their crude imports and boosting product exports. This will keep the year-on-year (yoy) reductions in OECD commercial oil stocks now showing up in the data on track, driven by continued sharp draws in U.S. inventories. Most importantly, these reductions will occur in the highly visible, high-frequency data produced by the U.S. every week. Energy: Overweight. Reports of foreign workers being pulled from Venezuelan oil fields will keep markets on edge. We remain long Dec/17 $50/bbl calls and short $55/bbl calls in Brent and WTI, which are up 127% and 74% since inception on June 22 and June 15, respectively. Base Metals: Neutral. Aluminum rallied on the back of news reports China's Shandong province ordered more than 3.2mm MT/yr of capacity shuttered by end-July. While surprising, such actions are not inconsistent with the stricter enforcement of environmental regulations in China we expect going forward. Precious Metals: Neutral. We remain long gold as a strategic portfolio hedge. Recent geopolitical tensions between the U.S. and North Korea are supporting this position, which is up 2.1% since inception on May 4, 2017. Ags/Softs: Underweight. Grains were treading water ahead of today's WASDE. We remain bearish, but continue to avoid shorting the complex. Feature Chart of the WeekU.S. Refiners Running At Close To Record Rates KSA's decision to reduce crude oil allocations to refiners in September, particularly in the U.S., is a well-timed tactical move.1 U.S. refinery net crude inputs hit record levels in early June at 17.3mm b/d, and remain close to that level (Chart of the Week). U.S. product exports continue at near-record levels, while imports have been trending lower (Chart 2). Crude oil exports from the U.S. are running close to record levels, and imports are trending lower (Chart 3). U.S. exports of crude and products hit a record in January at 5.9mm b/d - 5.24mm b/d of products, and just under 650k b/d for crude exports. At the end of July, total exports of crude and products stood at 5.44mm b/d, or 7.4% below the record set in January. U.S. product exports fell to 4.6mm b/d, while crude exports stood at 845k b/d. It is worthwhile pointing out that, in terms of total oil and products exports, the U.S. ranks among the top exporters in the world: KSA exports ~ 7mm b/d of crude, while Russia exports ~ 5mm b/d of crude. Chart 2U.S. Product Export Remain Strong,##BR##While Imports Continue Trending Lower ... Chart 3... While U.S. Crude Exports Remain High,##BR##And Imports Are Moderating With net U.S. crude and product imports declining (Chart 4), we expect U.S. commercial oil inventories - crude and products - to continue to draw sharply, which, since they account for close to 45% of OECD inventories, will draw down total DM stock levels as well (Chart 5). Indeed, U.S. commercial inventories drew close to 4% yoy in July, based on EIA historical data, the second month in a row the yoy comparisons came in negative in America. For the OECD as a whole, July marked the first month this year that the yoy percent change in stock levels was negative (-1.8%). Thus, as the summer driving season - and peak refiner crude demand - reaches its denouement next month, KSA's well-timed move to reduce shipments to U.S. refiners will push inventories lower and advance OPEC 2.0's agenda to clear out surplus OECD commercial oil inventories over the short term (Chart 6). Chart 4U.S. Net Crude And##BR##Product Imports Are Falling ... Chart 5... Which Will Support Continued Draws In##BR##Commercial Oil Stocks (Crude And Products) Chart 6KSA Will Continue Reducing##BR##Shipments To U.S. Refiners The OPEC 2.0 Agreement Is Holding ... On Average ... KSA is following through on Energy Minister Khalid al-Falih's "whatever it takes" assertion and making room for Libya and Nigeria, which together have added some 750k b/d of production to the market vs. April's level - 470k b/d for Libya and 280k b/d for Nigeria. April happens to be the month during which OPEC's producers recorded their largest production cuts vs. October's levels (1.12mm b/d), based on the EIA's historical tallies. OPEC 2.0 benchmarks to October 2016 production levels. Among OPEC members, neither Libya nor Nigeria were bound by the historic OPEC 2.0 Production Agreement. However, for those states that did obligate themselves to the agreement, compliance has been fairly high on average. OPEC member states that are party to the 2.0 deal have overproduced relative to their agreed production volumes by some 20k b/d over the January - July period on average.2 So, relative to the deal the OPEC members agreed, they've managed to cut 800k b/d of crude production on average versus their October 2016 production levels.3 During this period, Iraq stands out for its overproduction, having pumped 100k b/d on average over its agreed OPEC 2.0 volume of 4.35mm b/d (Chart 7). Among the non-OPEC members of the OPEC 2.0 coalition, Russia's compliance appears to be holding up, at close to 300k b/d below its October levels of crude and liquids production in 2Q17 and July (Chart 8). Oman produced ~ 980k b/d, over the first seven months of the deal vs. 1.02mm b/d in October, while Kazakhstan has faltered, with production averaging 1.88mm b/d in Jan - July, versus 1.79mm b/d in October. Chart 7Iraq Stands Out For Overproduction;##BR##Libya, Nigeria Not Covered In OPEC 2.0 Deal Chart 8Russia And KSA##BR##Continue To Lead OPEC 2.0 ... But Markets Await Articulated Strategy We continue to expect compliance with the OPEC 2.0 deal to remain relatively high to March 2018, which will draw OECD storage down to five-year average levels. We also are maintaining our expectation Brent prices will trade to $60/bbl by year end, with WTI trading ~ $58/bbl. Nonetheless, when we update our balances this month, we will continue to model for "compliance fatigue" among the OPEC 2.0 coalition. The fact that KSA and Russia are able to keep their rapport strong and compliance levels among OPEC and non-OPEC states relatively high, is a necessary condition for keeping OPEC 2.0 a viable coalition. However, the sufficient condition remains articulating a position on managing production via OPEC 2.0 that all these states can buy into, and support with concrete action. If, once the deal expires, the parties to the OPEC 2.0 coalition are left to go their own way and resume a production free-for-all, prices almost surely will fall, as the battle for market share is resumed. The ironic outcome of all this likely would be further destruction of capex budgets, which will set up another violent price surge that kills demand. We have no doubt the principal negotiators in OPEC 2.0 continue to discuss this, and that they are working on guidance. Bottom Line: KSA's tactical move to reduce exports to the U.S. likely will accelerate the commercial oil storage drawdown now apparent in OECD inventories, if current U.S. trends hold up - i.e., refinery runs remain high, exports of crude and products remain strong, and imports continue to fall yoy. Strategically, OPEC 2.0 still needs to convince markets there is a longer-term game plan for managing its output, short of a production free-for-all. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com 1 This tactical move was reported by Reuters earlier this week. Please see "Saudi Arabia cuts crude oil allocations in September by more than its OPEC pledge," which was published by reuters.com August 8, 2017. 2 We are using the production levels specified by the Cartel in its "OPEC Bulletin 11 - 12/16" on p. 35. 3 This likely overstates the actual production available for export by KSA, since the Kingdom typically consumes some 500 - 600k b/d of crude domestically over the June - September period as direct-burn fuel to power generation producing electricity for air conditioners. So the reported data likely are noisy at this time of year. 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

Special Report

Highlights Dear Clients, We are publishing a Special Report prepared by my colleague Jonathan LaBerge who examines the case for allocating capital to EM stocks within a global equity portfolio. I hope you will find this report insightful. Best regards, Garry Evans The relative performance of emerging market equities is challenging the downward trend channel that has been in place for the past seven years. This has led to renewed interest in EM from global investors, and warrants a revisit of the role of emerging market equities within a global equity portfolio. While EM recorded the highest regional equity return last cycle (2002-07), they were surprisingly not the "ideal" regional equity market in an efficient portfolio allocation. Recently, several compositional changes within the EM equity universe give the appearance of much lower commodity exposure than is truly the case. But EM equities will still be correlated with broad commodities prices because the later reflect Chinese growth dynamics. Cyclical indicators for China's economy suggest that the broad trend in commodities prices is likely to be lackluster over the coming year, at best. Consequently, EM stocks offer a poor risk/return profile, justifying an underweight stance within a global equity portfolio. Feature Chart I-1Change In Trend, Or Another Failed Rally? In U.S. dollar terms, the relative performance of emerging market (EM) stocks has been in an uptrend for over 18 months, and now appears to be challenging the downward trend channel that has been in place for the past seven years (Chart I-1). This has led to a renewed interest in EM, particularly among global investors. This report takes the recent outperformance of EM stocks as an opportunity to revisit their past and future contribution to a global equity portfolio, and what this might mean for an allocation to EM equities over the coming year. We conclude that EM's return behavior during the last economic cycle (2002-2007), its continued link to commodities prices, and China's growth dynamics all contribute to a poor risk/return profile for EM over the coming year. Barring compelling signs of a durable commodity bull market, investors should underweight EM stocks within a global equity portfolio. EM Equities In A Global Context: Some Historical Perspective When examining whether emerging markets are attractive from the perspective of global equity allocation, a starting point is to analyze the fundamental drivers of regional earnings. One major driver of global earnings over the past 20 years has been commodities prices; Chart I-2 highlights how 12-month forward EPS for stocks in all major regions have been correlated with commodities since the late-1990s. Chart I-2ACommodities Prices Are Correlated With Earnings... Chart I-2B...Even In Developed Markets This can be largely explained by the fact that commodities tend to be a pro-cyclical asset class. However, the super cycle in commodities prices in the 2000s not only bolstered the earnings of global resource companies, it also powered earnings growth for export-oriented industrials as well as domestic demand plays in commodity-producing countries. Chart I-3Strong Correlation Between ##br##Commodities And EM Emerging markets were among the largest beneficiaries of the commodity boom; net commodity-exporting countries made up roughly 45% of EM market capitalization throughout the last economic cycle, whereas stocks in the resource sector made up between 25-30% of the index by weight. Unsurprisingly, the relative performance of EM stocks closely tracked commodities prices over this period (Chart I-3). But despite this, EM was surprisingly not the "ideal" regional equity market last cycle within an active portfolio, even though it had the highest return. Chart I-4A presents a scatterplot of annualized regional equity volatility and return from 2002 - 2007, measured in US$ terms. The chart also shows the ex-post Modern Portfolio Theory (MPT) efficient frontier, with Chart I-4B presenting the efficient regional allocation at each point along the frontier. Chart I-4AEmerging Market Stocks Had The Highest Return Last Cycle... Chart I-4B...But Were Only The Favored Market For High-Risk Portfolios Chart I-5From 2002-2007, Earnings Drove More ##br##Of The Rally In DCM Than EM While the charts show that the efficient allocation to emerging market stocks did rise to a maximum of 100% during the last economic cycle, it did not become the dominant region until the portfolio became considerably more volatile than the global equity benchmark. Indeed, Chart I-4B shows that developed commodity markets (DCM) were the preferred commodity play for most of the efficient frontier, owing to their superior performance in risk-adjusted terms. This risk-adjusted outperformance may have occurred because DCM returns last cycle were driven more by earnings than by multiple expansion; Chart I-5 highlights that EM stock prices benefitted from multiple expansion last cycle by outpacing forward earnings, versus the opposite in the case of DCM. Since the onset of the U.S. recession in 2008, Chart I-6A and Chart I-6B highlight that the ex-post efficient portfolio has been much more skewed than during the last economic cycle. The charts show that the frontier since 2008 has been extremely short, with efficient allocations only accruing to three countries with typically defensive stock markets: the U.S., Japan, and Switzerland, with a heavy bias towards the former. From the perspective of a global equity portfolio, this historical review leads to two conclusions: 1) investors should not allocate to EM unless they are bullish on commodities prices and, 2) if investors are bullish towards commodities, developed commodity markets have historically been a better risk-adjusted bet than emerging markets as a commodity play. Chart I-6ASince 2008, The Efficient Frontier Has Been Highly Skewed... Chart I-6B...Towards Defensive Markets (Mostly The U.S.) Chart I-7These Trends Give The False Appearance ##br##Of Lower EM Commodity Exposure EM And Commodities Prices: Has The Relationship Really Changed? More recently, a narrative has developed in the market that EM stocks are now far less sensitive to commodities prices than used to be the case. Proponents of this theory point to the following changes in the composition of emerging market equity benchmarks: First, the market capitalization weight of net commodity exporting countries has fallen precipitously since the onset of the collapse in oil prices in 2014 (Chart I-7, panel 1). On average, net commodity exporters made up between 40-45% of EM equity market cap from 2000 to 2013, but their share now stands at 27%. Second, Chart I-7, panel 2, shows that the market cap weight of resource sectors (energy plus materials) in emerging markets has fallen from roughly 30% to 14% over the past five years, a trend that pre-dated the decline in the share of net commodity exporters. Third, the enormous rise in the market capitalization of technology companies as a share of total EM market cap has been specifically cited by many market participants (Chart I-7, panel 3), especially since EM is now heavily overweight the tech sector relative to the global average. Broadly speaking, a fourth compositional change within the EM equity benchmark generally captures all of the shifts noted above, and is the focus of our remaining analysis below: the rise in the weight of emerging Asia as a share of overall EM (Chart I-7, panel 4). Among emerging markets, net commodity exporters tend to be located outside of Asia (with the exception of Indonesia and Malaysia), and emerging Asia accounts for essentially all of EM tech market cap. Consequently, investors who argue that EM equities have largely or fully decoupled from commodities prices are essentially arguing that emerging Asian equities are far less affected by changes in commodity markets than they used to be. This idea is deeply flawed, as shown below: Based on export share, Chart I-8 highlights that emerging Asia is far more economically exposed to China than developed markets and EM ex-Asia. While China is gradually becoming more of a services-oriented economy, Chart I-9 highlights that the sum of primary industry (raw material extraction), secondary industry (manufacturing and construction), and real estate services still account for over half of China's economic activity, well above that of industrialized nations such as the U.S. This underscores that emerging Asia's trade exposure to China is fundamentally rooted in economic activity that is closely linked to commodity demand. Chart I-8Emerging Asia Has High ##br##Trade Exposure To China Chart I-9Chinese Growth Still Largely ##br##Reflects Industrial Activity Within the commodity-linked segment of China's economy, Chart I-10 shows that there is little evidence of a weaker relationship between output and commodities prices. Simple regression analysis underscores that the Li Keqiang index, a growth proxy for China's industrial sector, is strongly linked to the year-over-year % change in spot commodities prices since the beginning of the commodity bull market, and that this relationship has in fact been increasing in strength over time. In addition, Chart I-11 underscores that China remains by far the largest consumer of base metals globally. Demand in the global oil market is considerably more diversified than the market for base metals, but China is the second-largest end market for oil (14% of global oil consumption), and accounted for over a quarter of the growth in total oil demand in 2016.1 Chart I-10Moderating Chinese Growth Will ##br##Be Negative For Commodities Chart I-11China Is By Far The Most Important ##br##End Market For Base Metals Finally, Chart I-12 shows a regression model between forward earnings expectations for emerging Asia and commodities prices, both at the overall index level and even for the financial sector (which, along with real estate, accounts for almost 25% of emerging Asian market capitalization). The fit for both models is extremely strong and, similar to the increasing strength of the Li Keqiang / commodity price relationship, the chart shows that commodities prices have begun to lead the growth in forward earnings, when the relationship used to be much more coincident. Chart I-12Emerging Asian Earnings Are Strongly ##br##Correlated With Commodities Prices The bottom line for investors is that Charts I-8-12 show emerging Asian economies are strongly linked economically to China, and that China remains the dominant driver of aggregate commodity demand. This means that while EM stocks may not have as much direct commodity exposure as they used to, they will continue to experience a high correlation with commodities prices because that the latter will be driven by swings in China's business cycle. In brief, Chinese growth fluctuations are instrumental to emerging Asia's economic and equity market performance. This is the rationale behind the very strong link between earnings expectations for emerging Asia and commodities prices: the latter reflect cyclical variations in the Chinese economy. EM Stocks: A Lackluster Bet Given The Outlook For Commodities Our earlier discussion of EM's historical contribution to a global equity portfolio revived elements of Modern Portfolio Theory (MPT), at least from an ex-post perspective. Ex-ante, investors need to make judgements about the likely risk, return, and cross-correlation of an asset when assessing its likely contribution to a diversified portfolio. Regarding the latter factor, Chart I-13 highlights that EM's correlation with global ex-EM has actually fallen quite substantially over the past year, which is a potential argument in the minds of some investors in favor of an increased allocation to EM. When recalling the lessons from Modern Portfolio Theory, most investors tend to focus on the key insight that lowly-correlated assets are valuable from the perspective of constructing a portfolio with an attractive risk/return profile. While this is true, many investors often forget that this is only valid given an expectation of a positive return. The efficient allocation to an asset that has a strongly negative correlation with other assets but has a negative return expectation is basically zero. This means that global investors eying an increased allocation to emerging markets should be squarely focused on EM equities' absolute performance, which as we have highlighted above are likely to be closely linked to commodity returns. Over the coming 6-12 months, Chart I-14 paints an uninspiring picture for commodities prices based on two measures of China's money supply. In turn, interest rates lead money growth and the rise in the former over the past nine months heralds further deceleration in the latter. This implies that the Chinese economy will likely continue to moderate, which is negative for the broad trend in commodities prices. Chart I-13A Significant Decline, But Focus On Return ##br##Expectations, Not Correlation Chart I-14Interest Rates And Money Growth Paint ##br##A Poor Picture For Commodities As noted above, China's share of the global oil market is much lower than that of base metals, and we do not expect China's oil demand to shrink even if its industrial sector slumps. But from the perspective of allocating to EM equities within a global portfolio, Table I-1 highlights that broad spot commodity price indexes tend to be more relevant predictors of forward earnings growth than energy prices alone. This means that a rise in oil prices (were it to occur for idiosyncratic supply reasons) might be positive for major oil producers such as Russia,2 but is unlikely to provide a broad-based catalyst for EM stocks. Table I-1Explanatory Power Of Commodity Price Indexes In Modeling ##br##12-Month Forward Earnings Per Share Growth (2002-2016) Finally, our analysis above has focused on the fundamental drivers of EM stocks, and has shown how DM investors are likely to have little basis to be bullish about emerging markets earnings over the coming 6-12 months. Chart I-15 highlights how this is also true about the potential for EM multiple expansion relative to their global peers. The chart shows that periods of relative EM multiple expansion have, like relative earnings expectations, tended to be associated with rising commodities prices, implying that a significant re-rating of EM equities is unlikely over the coming year. This is in addition the fact that EM stocks are neither cheap nor expensive in absolute terms,3 meaning that there is less room for multiple expansion in EM than many investors believe. Chart I-15No Relative Multiple Expansion ##br##Without Rising Commodities Prices Investment Conclusions In terms of gauging the contribution of EM equities to a global equity portfolio, this report has highlighted the following points: While EM stocks had the highest return of any regional equity market during the last economic cycle (2002-2007), this return profile was accompanied by an outsized degree of volatility. For all but the riskiest portfolios, developed commodity markets were preferred as a commodity play over emerging markets. Several compositional changes within the EM equity universe give the outward appearance of much lower commodity exposure, but this exposure has merely become indirect. While EM's weight towards net commodity exporters and resource sectors has declined, this has shifted benchmark exposure to emerging Asia which has significant economic exposure to China and its industrial sector (the dominant driver of global commodities prices). As such, share prices in EM overall and emerging Asia in particular will still be strongly correlated with commodities prices even given the region's significant weight towards the technology sector.4 Cyclical indicators for China's economy suggest that broad commodity price gains over the coming year are likely to be lackluster, at best (and may very well be negative). Even if global oil prices were to rise, this is unlikely to provide a broad-based catalyst for EM stocks if industrial metals prices relapse, as we expect. These conclusions underscore that it is highly unlikely emerging market stocks will sustainably decouple from commodities prices over the cyclical investment horizon, and that the uptrend in EM relative performance since early-2016 has likely been driven significantly by expectations of further China's growth acceleration and commodity gains. In our judgement, these circumstances have created a poor risk/return profile for emerging market equities, justifying an underweight stance within a global equity portfolio over the coming year. Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com 1 Source: BP Statistical Review of World Energy, June 2017. 2 Note that we recommend an overweight stance towards Russian equities within an EM equity portfolio. 3 Please refer to the Emerging Markets Strategy Weekly Report titled, "EM Equity Valuations Revisited," dated March 29, 2017, link available on page 15. 4 For a further discussion of the impact of the technology sector on the relative performance of emerging market stocks, please see Emerging Markets Strategy Weekly Report titled, "Can Tech Drive EM Stocks Higher?" dated May 17, 2017, link available on page 15.

Special Report

Highlights Dear Clients, We are publishing a Special Report prepared by my colleague Jonathan LaBerge who examines the case for allocating capital to EM stocks within a global equity portfolio. I hope you will find this report insightful. Best regards, Arthur Budaghyan The relative performance of emerging market equities is challenging the downward trend channel that has been in place for the past seven years. This has led to renewed interest in EM from global investors, and warrants a revisit of the role of emerging market equities within a global equity portfolio. While EM recorded the highest regional equity return last cycle (2002-07), they were surprisingly not the "ideal" regional equity market in an efficient portfolio allocation. Recently, several compositional changes within the EM equity universe give the appearance of much lower commodity exposure than is truly the case. But EM equities will still be correlated with broad commodities prices because the latter reflect Chinese growth dynamics. Cyclical indicators for China's economy suggest that the broad trend in commodities prices is likely to be lackluster over the coming year, at best. Consequently, EM stocks offer a poor risk/return profile, justifying an underweight stance within a global equity portfolio. Feature Chart I-1Change In Trend, Or Another Failed Rally? In U.S. dollar terms, the relative performance of emerging market (EM) stocks has been in an uptrend for over 18 months, and now appears to be challenging the downward trend channel that has been in place for the past seven years (Chart I-1). This has led to a renewed interest in EM, particularly among global investors. This report takes the recent outperformance of EM stocks as an opportunity to revisit their past and future contribution to a global equity portfolio, and what this might mean for an allocation to EM equities over the coming year. We conclude that EM's return behavior during the last economic cycle (2002-2007), its continued link to commodities prices, and China's growth dynamics all contribute to a poor risk/return profile for EM over the coming year. Barring compelling signs of a durable commodity bull market, investors should underweight EM stocks within a global equity portfolio. EM Equities In A Global Context: Some Historical Perspective When examining whether emerging markets are attractive from the perspective of global equity allocation, a starting point is to analyze the fundamental drivers of regional earnings. One major driver of global earnings over the past 20 years has been commodities prices; Chart I-2 highlights how 12-month forward EPS for stocks in all major regions have been correlated with commodities since the late-1990s. Chart I-2ACommodities Prices Are Correlated With Earnings... Chart I-2B...Even In Developed Markets This can be largely explained by the fact that commodities tend to be a pro-cyclical asset class. However, the super cycle in commodities prices in the 2000s not only bolstered the earnings of global resource companies, it also powered earnings growth for export-oriented industrials as well as domestic demand plays in commodity-producing countries. Chart I-3Strong Correlation Between ##br##Commodities And EM Emerging markets were among the largest beneficiaries of the commodity boom; net commodity-exporting countries made up roughly 45% of EM market capitalization throughout the last economic cycle, whereas stocks in the resource sector made up between 25-30% of the index by weight. Unsurprisingly, the relative performance of EM stocks closely tracked commodities prices over this period (Chart I-3). But despite this, EM was surprisingly not the "ideal" regional equity market last cycle within an active portfolio, even though it had the highest return. Chart I-4A presents a scatterplot of annualized regional equity volatility and return from 2002 - 2007, measured in US$ terms. The chart also shows the ex-post Modern Portfolio Theory (MPT) efficient frontier, with Chart I-4B presenting the efficient regional allocation at each point along the frontier. Chart I-4AEmerging Market Stocks Had The Highest Return Last Cycle... Chart I-4B...But Were Only The Favored Market For High-Risk Portfolios Chart I-5From 2002-2007, Earnings Drove More ##br##Of The Rally In DCM Than EM While the charts show that the efficient allocation to emerging market stocks did rise to a maximum of 100% during the last economic cycle, it did not become the dominant region until the portfolio became considerably more volatile than the global equity benchmark. Indeed, Chart I-4B shows that developed commodity markets (DCM) were the preferred commodity play for most of the efficient frontier, owing to their superior performance in risk-adjusted terms. This risk-adjusted outperformance may have occurred because DCM returns last cycle were driven more by earnings than by multiple expansion; Chart I-5 highlights that EM stock prices benefitted from multiple expansion last cycle by outpacing forward earnings, versus the opposite in the case of DCM. Since the onset of the U.S. recession in 2008, Chart I-6A and Chart I-6B highlight that the ex-post efficient portfolio has been much more skewed than during the last economic cycle. The charts show that the frontier since 2008 has been extremely short, with efficient allocations only accruing to three countries with typically defensive stock markets: the U.S., Japan, and Switzerland, with a heavy bias towards the former. From the perspective of a global equity portfolio, this historical review leads to two conclusions: 1) investors should not allocate to EM unless they are bullish on commodities prices and, 2) if investors are bullish towards commodities, developed commodity markets have historically been a better risk-adjusted bet than emerging markets as a commodity play. Chart I-6ASince 2008, The Efficient Frontier Has Been Highly Skewed... Chart I-6B...Towards Defensive Markets (Mostly The U.S.) Chart I-7These Trends Give The False Appearance ##br##Of Lower EM Commodity Exposure EM And Commodities Prices: Has The Relationship Really Changed? More recently, a narrative has developed in the market that EM stocks are now far less sensitive to commodities prices than used to be the case. Proponents of this theory point to the following changes in the composition of emerging market equity benchmarks: First, the market capitalization weight of net commodity exporting countries has fallen precipitously since the onset of the collapse in oil prices in 2014 (Chart I-7, panel 1). On average, net commodity exporters made up between 40-45% of EM equity market cap from 2000 to 2013, but their share now stands at 27%. Second, Chart I-7, panel 2, shows that the market cap weight of resource sectors (energy plus materials) in emerging markets has fallen from roughly 30% to 14% over the past five years, a trend that pre-dated the decline in the share of net commodity exporters. Third, the enormous rise in the market capitalization of technology companies as a share of total EM market cap has been specifically cited by many market participants (Chart I-7, panel 3), especially since EM is now heavily overweight the tech sector relative to the global average. Broadly speaking, a fourth compositional change within the EM equity benchmark generally captures all of the shifts noted above, and is the focus of our remaining analysis below: the rise in the weight of emerging Asia as a share of overall EM (Chart I-7, panel 4). Among emerging markets, net commodity exporters tend to be located outside of Asia (with the exception of Indonesia and Malaysia), and emerging Asia accounts for essentially all of EM tech market cap. Consequently, investors who argue that EM equities have largely or fully decoupled from commodities prices are essentially arguing that emerging Asian equities are far less affected by changes in commodity markets than they used to be. This idea is deeply flawed, as shown below: Based on export share, Chart I-8 highlights that emerging Asia is far more economically exposed to China than developed markets and EM ex-Asia. While China is gradually becoming more of a services-oriented economy, Chart I-9 highlights that the sum of primary industry (raw material extraction), secondary industry (manufacturing and construction), and real estate services still account for over half of China's economic activity, well above that of industrialized nations such as the U.S. This underscores that emerging Asia's trade exposure to China is fundamentally rooted in economic activity that is closely linked to commodity demand. Chart I-8Emerging Asia Has High ##br##Trade Exposure To China Chart I-9Chinese Growth Still Largely ##br##Reflects Industrial Activity Within the commodity-linked segment of China's economy, Chart I-10 shows that there is little evidence of a weaker relationship between output and commodities prices. Simple regression analysis underscores that the Li Keqiang index, a growth proxy for China's industrial sector, is strongly linked to the year-over-year % change in spot commodities prices since the beginning of the commodity bull market, and that this relationship has in fact been increasing in strength over time. In addition, Chart I-11 underscores that China remains by far the largest consumer of base metals globally. Demand in the global oil market is considerably more diversified than the market for base metals, but China is the second-largest end market for oil (14% of global oil consumption), and accounted for over a quarter of the growth in total oil demand in 2016.1 Chart I-10Moderating Chinese Growth Will ##br##Be Negative For Commodities Chart I-11China Is By Far The Most Important ##br##End Market For Base Metals Finally, Chart I-12 shows a regression model between forward earnings expectations for emerging Asia and commodities prices, both at the overall index level and even for the financial sector (which, along with real estate, accounts for almost 25% of emerging Asian market capitalization). The fit for both models is extremely strong and, similar to the increasing strength of the Li Keqiang / commodity price relationship, the chart shows that commodities prices have begun to lead the growth in forward earnings, when the relationship used to be much more coincident. Chart I-12Emerging Asian Earnings Are Strongly ##br##Correlated With Commodities Prices The bottom line for investors is that Charts I-8-12 show emerging Asian economies are strongly linked economically to China, and that China remains the dominant driver of aggregate commodity demand. This means that while EM stocks may not have as much direct commodity exposure as they used to, they will continue to experience a high correlation with commodities prices because that the latter will be driven by swings in China's business cycle. In brief, Chinese growth fluctuations are instrumental to emerging Asia's economic and equity market performance. This is the rationale behind the very strong link between earnings expectations for emerging Asia and commodities prices: the latter reflect cyclical variations in the Chinese economy. EM Stocks: A Lackluster Bet Given The Outlook For Commodities Our earlier discussion of EM's historical contribution to a global equity portfolio revived elements of Modern Portfolio Theory (MPT), at least from an ex-post perspective. Ex-ante, investors need to make judgements about the likely risk, return, and cross-correlation of an asset when assessing its likely contribution to a diversified portfolio. Regarding the latter factor, Chart I-13 highlights that EM's correlation with global ex-EM has actually fallen quite substantially over the past year, which is a potential argument in the minds of some investors in favor of an increased allocation to EM. When recalling the lessons from Modern Portfolio Theory, most investors tend to focus on the key insight that lowly-correlated assets are valuable from the perspective of constructing a portfolio with an attractive risk/return profile. While this is true, many investors often forget that this is only valid given an expectation of a positive return. The efficient allocation to an asset that has a strongly negative correlation with other assets but has a negative return expectation is basically zero. This means that global investors eying an increased allocation to emerging markets should be squarely focused on EM equities' absolute performance, which as we have highlighted above are likely to be closely linked to commodity returns. Over the coming 6-12 months, Chart I-14 paints an uninspiring picture for commodities prices based on two measures of China's money supply. In turn, interest rates lead money growth and the rise in the former over the past nine months heralds further deceleration in the latter. This implies that the Chinese economy will likely continue to moderate, which is negative for the broad trend in commodities prices. Chart I-13A Significant Decline, But Focus On Return ##br##Expectations, Not Correlation Chart I-14Interest Rates And Money Growth Paint ##br##A Poor Picture For Commodities As noted above, China's share of the global oil market is much lower than that of base metals, and we do not expect China's oil demand to shrink even if its industrial sector slumps. But from the perspective of allocating to EM equities within a global portfolio, Table I-1 highlights that broad spot commodity price indexes tend to be more relevant predictors of forward earnings growth than energy prices alone. This means that a rise in oil prices (were it to occur for idiosyncratic supply reasons) might be positive for major oil producers such as Russia,2 but is unlikely to provide a broad-based catalyst for EM stocks. Table I-1Explanatory Power Of Commodity Price Indexes In Modeling ##br##12-Month Forward Earnings Per Share Growth (2002-2016) Finally, our analysis above has focused on the fundamental drivers of EM stocks, and has shown how DM investors are likely to have little basis to be bullish about emerging markets earnings over the coming 6-12 months. Chart I-15 highlights how this is also true about the potential for EM multiple expansion relative to their global peers. The chart shows that periods of relative EM multiple expansion have, like relative earnings expectations, tended to be associated with rising commodities prices, implying that a significant re-rating of EM equities is unlikely over the coming year. This is in addition the fact that EM stocks are neither cheap nor expensive in absolute terms,3 meaning that there is less room for multiple expansion in EM than many investors believe. Chart I-15No Relative Multiple Expansion ##br##Without Rising Commodities Prices Investment Conclusions In terms of gauging the contribution of EM equities to a global equity portfolio, this report has highlighted the following points: While EM stocks had the highest return of any regional equity market during the last economic cycle (2002-2007), this return profile was accompanied by an outsized degree of volatility. For all but the riskiest portfolios, developed commodity markets were preferred as a commodity play over emerging markets. Several compositional changes within the EM equity universe give the outward appearance of much lower commodity exposure, but this exposure has merely become indirect. While EM's weight towards net commodity exporters and resource sectors has declined, this has shifted benchmark exposure to emerging Asia which has significant economic exposure to China and its industrial sector (the dominant driver of global commodities prices). As such, share prices in EM overall and emerging Asia in particular will still be strongly correlated with commodities prices even given the region's significant weight towards the technology sector.4 Cyclical indicators for China's economy suggest that broad commodity price gains over the coming year are likely to be lackluster, at best (and may very well be negative). Even if global oil prices were to rise, this is unlikely to provide a broad-based catalyst for EM stocks if industrial metals prices relapse, as we expect. These conclusions underscore that it is highly unlikely emerging market stocks will sustainably decouple from commodities prices over the cyclical investment horizon, and that the uptrend in EM relative performance since early-2016 has likely been driven significantly by expectations of further China's growth acceleration and commodity gains. In our judgement, these circumstances have created a poor risk/return profile for emerging market equities, justifying an underweight stance within a global equity portfolio over the coming year. Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com 1 Source: BP Statistical Review of World Energy, June 2017. 2 Note that we recommend an overweight stance towards Russian equities within an EM equity portfolio. 3 Please refer to the Emerging Markets Strategy Weekly Report titled, "EM Equity Valuations Revisited," dated March 29, 2017, link available on page 15. 4 For a further discussion of the impact of the technology sector on the relative performance of emerging market stocks, please see Emerging Markets Strategy Weekly Report titled, "Can Tech Drive EM Stocks Higher?" dated May 17, 2017, link available on page 15.

Feature Turkey's banking system has in recent years relied on enormous liquidity provisions by the central bank (Chart I-1) to sustain its ongoing credit boom, and hence economic growth. Since early this year, the authorities have doubled down: they have also begun using fiscal policy to prop up growth. Chart I-1Turkey: Central Bank Large Liquidity Injections On the whole, this combination of colossal credit and fiscal stimulus is indisputably bearish for the currency. Despite strong performance by Turkish stocks this year, we are maintaining our bearish call on the lira. The lira is set to depreciate by 20-25% in the next 12 months or so versus both an equally-weighted basket of the U.S. dollar and the euro. Bringing Fiscal Stimulus Into Play The Turkish authorities have recently begun using fiscal means to stimulate growth: Last summer, a sovereign wealth fund was set up by presidential decree to pool shares in companies owned by the government and use them as collateral to raise debt and initiate spending on various infrastructure projects. The target size of the fund is US$ 200 billion, compared with the government non-interest expenditure of US$ 165 billion in the last 12 months. This would effectively allow the government to issue debt and increase expenditures off-balance sheet. In addition, this past March, the government decided to recapitalize the Credit Guarantee Fund. This initiative allowed it to underwrite US$ 50 billion, or 7% of GDP, worth of credit to Turkish companies. This is considerable as it compares with US$ 93 billion worth of loan origination by commercial banks last year. By assuming credit risk on these loans, the government is effectively encouraging banks to lend, in turn boosting economic growth. In effect, this has lowered lending standards and given a green light to banks to flood the economy with credit. Even though interest rates have risen since last November, credit growth has accelerated as banks have provided loans covered by government guarantees (Chart I-2). On top of this quasi-fiscal stimulus, government expenditures excluding interest payments have accelerated (Chart I-3). Chart I-2Bank Loan Growth Has Accelerated ##br##Despite Higher Interest Rates Chart I-3Turkey: Fiscal Spending Has Surged Such a rise in government spending has been financed by commercial banks whose holdings of government bonds have risen sharply. Essentially, government spending has also been funded by commercial banks' money creation. In short, fiscal and credit stimulus have boosted domestic demand, thereby widening the country's current account deficit once again (Chart I-4A and Chart I-4B). Chart I-4AWidening Twin Deficit Chart I-4BWidening Twin Deficit Given that the starting point of the government's fiscal position is good - public debt stands at only 28% of GDP - the authorities have ample room to rely on fiscal levers to promote growth. However, a widening fiscal deficit will be bearish for the currency. Bottom Line: Widening twin (current account and fiscal) deficits (Chart I-4A and Chart I-4B) are a bad omen for the lira. Monetary Tightening? What Monetary Tightening? Chart I-5Turkey: Money/Credit Growth Is Too Strong Although interbank and lending rates have risen in recent months, money and credit growth have been booming (Chart I-5). This does not support the idea that monetary policy is tight. On the contrary, thriving money and credit growth suggest that the policy stance is very easy. The Central Bank of Turkey (CBT) raised various policy rates and capped the overnight liquidity facility at the beginning of this year. However, commercial banks' usage of the late liquidity window facility - the one facility that has been left uncapped - has literally gone exponential - it has risen from zero to TRY 70 billion in the past 8 months. On the whole, the central bank’s net liquidity injections into the banking system continue to make new highs, even though the price of liquidity has been rising. Adding all the liquidity facilities – the intraday, overnight and late window facilities – the CBT's outstanding funding to banks is 90 billion TRY, or 3% of GDP, more than ever recorded (Chart 1, bottom panel). This entails that monetary policy is loose rather than tight. On the whole, commercial banks are requiring more and more liquidity, and the CBT is continuously supplying it. These injections maintain liquidity in the banking system to a sufficiently high level to allow aggressive money/credit creation among commercial banks. Bottom Line: The CBT is facilitating/accommodating an economy-wide credit binge by providing copious amounts of liquidity to commercial banks. The Victim Is The Lira The lira will inevitably depreciate in the months ahead: Chart I-6Turkey: Central Bank's Foreign ##br##Reserves Have Been Depleted The lira's exchange rate versus an equally-weighted basket of the U.S. dollar and the euro has been mostly flat year-to-date, despite the CBT intervening in the market to support the lira by selling U.S. dollars. Aggressive selling of CBT foreign exchange reserves has so far prevented much steeper lira depreciation in Turkey. However at this stage, the central bank is literally running out of reserves and will soon lose its ability to support the currency (Chart I-6). A developing country with foreign exchange reserves worth less than three months' imports is considered vulnerable. Therefore, at 0.5 months of imports coverage, or US$ 9.7 billion, the CBT has little capacity to continue supporting the currency via interventions. Economic growth has recovered: export volumes are very strong, driven by shipments to Europe, while loan growth is supporting private domestic demand and government expenditures have mushroomed. The ongoing economic recovery will boost inflation, and strong domestic demand will assure the current account deficit widens. This will weigh on the exchange rate. Core inflation measures have subsided from 10% to 7%, but remain well above the central bank's target of 5%. Provided inflation is a lagging variable, the acceleration in money growth and domestic demand this year will lead to higher inflation in the months ahead. Wage growth remains high and our profit margin proxy for both manufacturing and service industries - calculated as core CPI divided by unit labor costs - has relapsed signifying deteriorating corporate profitability (Chart I-7). This in turn will force businesses to raise prices. Provided demand is strong, companies will likely succeed in passing through higher prices to customers. In brief, odds are that inflation will rise significantly soon. Escalating unit labor costs also offsets the benefit of nominal currency depreciation. Chart I-8 illustrates that the real effective exchange rate is not cheap based on consumer prices, or unit labor costs. Chart I-7Companies Profit Margins Are Shrinking Chart I-8The Lira Is Not Cheap At All As inflation rises, residents' desire to convert their deposits from local to foreign currency will increase. In fact, this is already happening - households' foreign currency deposit growth is accelerating. In short, lingering high inflation will continue to weigh on the currency's value. Bottom Line: The authorities have doubled down on fiscal and credit stimulus, warranting a doubling down on bearish bets on the lira. Investment Implications On the whole, the authorities will continue resorting to fiscal and monetary stimulus to sustain economic growth. According to the Impossible Trinity theory, in countries with an open capital account structure, the authorities can control either interest rates or the exchange rate, but not both simultaneously. Chart I-9Bank Stocks Have Rallied Despite ##br##Shrinking Net Interest Margins In Turkey, policymakers will eventually opt to control interest rates, meaning they will not have much control over the exchange rate. We suggest currency traders who are not shorting the lira do so at this time. We remain short the lira versus the U.S. dollar. A weaker lira will undermine U.S. dollar returns on Turkish stocks and domestic bonds. Dedicated EM equity investors as well as those overseeing EM fixed income and credit portfolios should continue to underweight Turkish assets within their respective EM universes. Bank stocks have rallied strongly, and have decoupled from interest rates (Chart I-9). This reflects the recent credit binge, where banks are making profits on loan originations while the government is holding responsibility for bad loans. These dynamics could persist for a while. However, both loan growth and banks' profitability will be hurt if the credit guarantee scheme is not renewed. So far, it is estimated that TRY 200 billion of an announced TRY 250 billion of this credit guarantee scheme has been utilized. Continuous credit guarantee schemes and accumulation of off-balance-sheet liabilities by the government will widen sovereign credit spreads. In many EM countries, including Turkey, bank share prices have historically correlated with sovereign spreads. Hence, rising sovereign risk will weigh on banks stocks too. Finally, as the lira begins to depreciate and inflation rises, local interest rates will have to climb. This will also weigh on bank share prices. In brief, we are reiterating our negative/underweight stance on Turkish banks. Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com Stephan Gabillard, Senior Analyst stephang@bcaresearch.com Equity Recommendations Fixed-Income, Credit And Currency Recommendations