Money/Credit/Debt

Special Report

Dear Client, I am visiting clients this week, and as such there will be no Weekly Report. Instead, we are sending you this Special Report written by my colleague Jonathan LaBerge. Jonathan argues that while the recent acceleration of the Canadian economy is genuine, the rise in Canadian household debt-to-income over the past 16 years has been so large that a credit-driven downturn in spending is now virtually unavoidable over the long run. I hope you will find this report both interesting and informative. Best regards, Peter Berezin, Chief Strategist Global Investment Strategy Highlights The recent economic improvement in Canada is genuine. In isolation, this supports the Bank of Canada's decision to gradually raise interest rates. However, over the long run, the historical experience suggests that the substantial leverage of Canadian households will ultimately cause a serious credit-driven downturn. Several myths about Canada's housing market have obscured the true extent of its credit market imbalances, heightening the risk that policymakers will ultimately overplay their hand when tightening monetary conditions. There are multiple potential triggers that could eventually spark a credit-driven downturn in Canada, but none of them seem likely to have a major impact on the economy over the coming 6-12 months. Favor a pro-cyclical stance over the coming year, but look to shift to a bearish structural view at some point beyond the immediate investment horizon. Feature Several developments over the past few months have altered the outlook for the Canadian economy. However, these events have not had a consistent impact on the narrative for Canadian assets. Whereas a sharp rebound in real GDP growth and a hawkish pivot from the Bank of Canada have been signs of a strengthening economy, the crisis surrounding Home Capital Group (a Canadian non-bank mortgage lender) was an ominous sign for many investors concerned about the deeply imbalanced Canadian housing market.1 In this report we argue that the cyclical improvement in the Canadian economy is legitimate, and that the Bank of Canada is likely to move forward with gradual policy tightening following Wednesday's move. However, the rise in Canadian household debt-to-income over the past 16 years has been so large that a credit-driven downturn in spending is now virtually unavoidable over the long run, rather than a risk. We highlight how, in many ways, the imbalances in the Canadian housing market are even worse than the market narrative would suggest. We also provide a checklist of factors to monitor in order to judge when Canada's day of reckoning will arrive. For now, it does not appear to be imminent. From an investment standpoint, our conclusions imply that investors should pursue a "two-staged" approach when allocating to Canadian assets. Over the coming 6-12 months, a cyclical improvement in the economy means that Canadian risky asset prices and government bond yields are likely to rise, and we believe that this stage is worth playing. But over the secular horizon, the reverse is likely to unfold, meaning that a rally in Canadian assets over the coming year will create excellent "selling conditions" for investors looking to position for a bearish structural view. Economic Momentum Is Spurring Tighter Monetary Policy... The Bank of Canada is now back on a path towards tighter monetary policy, and a close examination of the Canadian economy, as well as our outlook for global oil inventories, supports the BoC's view: Real consumer spending picked up significantly in Q1, rising from 2.7% to 3.1% on a year-over-year basis. Chart 1 highlights that the rise in real spending has been supported by a rebound in employment growth and consumer confidence (the latter is at a 9-year high). On the employment side, Chart 1 also shows that the acceleration in job growth is not limited to provinces that are strongly associated with oil sands production. In fact, the chart shows that employment in Canada excluding Alberta and Saskatchewan has been in an uptrend since mid-2014, when fiscal and monetary policy began to respond to the shock from a collapse in the price of oil. All Canadian employment cylinders are now firing, given the job recovery in oil sands provinces. Real Canadian gross fixed capital formation turned positive in Q1 after a significant decline into negative territory, and a simple model based on business confidence, oil prices, and the Canadian dollar (stripped of its correlation with oil) suggests that it will continue to accelerate modestly over the coming year (Chart 2). Chart 1Genuine Signs Of A Stronger Economy Chart 2Further Gains In Investment Likely Chart 3 shows a model for oil prices, based on global industrial production, oil production, OECD oil inventories, and oil consumption in the major countries and China. If OPEC is successful in reducing inventories to their 5-year moving average, as BCA's commodity strategists expect, the model implies that oil prices will rise materially. This is likely to provide a tailwind for the Canadian economy, at least in nominal terms. While the pace of tightening is likely to be gradual because of the weakness in Canadian core inflation, Chart 4 suggests that the decline in inflation over the past few months may simply represent the correction towards more fundamentally-justified levels. The chart shows a model of core inflation based on lagged real core consumer spending and the Canadian dollar (as a proxy for imported inflation/deflation), and highlights that actual inflation has overshot the model value over the past three years. But the chart also shows that the fundamentally-justified level of core inflation remains in an uptrend, suggesting that recent weakness is likely temporary and is thus not an impediment to higher policy rates over the coming year. Chart 3Falling Inventories Will Be Bullish For Oil Chart 4The Dip In Core Inflation Is Temporary Bottom Line: The recent economic improvement in Canada is genuine and, in isolation, supports the Bank of Canada's decision to gradually raise interest rates. ...But It Will All Likely End In Tears Chart 5Higher Household Leverage Than In The U.S. Pre-Crisis While we agree that the Bank of Canada is on a path to gradually raise interest rates over the coming year and that the economy is currently in good shape, the odds are good that tighter policy (and/or other factors) will eventually inflict considerable damage to the Canadian economy via the housing market and its impact on highly leveraged consumers. In this regard, the pickup in Canadian economic growth likely represents a happy moment in an otherwise sad story. Chart 5 compares Canada's mortgage debt-to-disposable income, total household debt-to-GDP, and the total household debt service ratio to that of the U.S. The chart neatly illustrates the fundamental basis for a bearish secular outlook for the Canadian economy, which is that household debt levels have risen enormously since 2000, to a level that is worse today than in the U.S. in 2007. "So what?" ask some investors. Household debt levels vary significantly across countries, meaning that an elevated level of household debt-to-income does not necessarily spell economic doom. Chart 6 counters this point by showing the relationship between the historical change in household debt-to-GDP (y-axis) versus the starting point for the ratio (x-axis) during episodes of significant household leveraging. The change in debt-to-GDP is shown as a 10-year average of the year-over-year change in the ratio, in order to compare Canada's recent debt binge with other long-term booms in credit. In terms of very significant increases in household credit-to-GDP from an already above-average level, Chart 6 shows that Canada's experience (an average yearly increase of 3.3%) has been among the most severe cases. The chart also shows that while there are a few exceptions, other observations in the neighborhood of Canada's have had a strong tendency to be associated with harsh economic consequences once the credit binge has come to an end. In particular, while the chart shows that the countries at the center of the euro area sovereign debt crisis saw a more rapid rise in household debt-to-GDP than observed in Canada, this occurred from a lower base. When measuring the total change in household debt-to-GDP, Canada has experienced almost the same magnitude rise from 2000 to today as what occurred in Spain and Portugal during the last economic cycle. In terms of a comparison with the U.S., Chart 7 presents a long-term perspective on the inverse relationship between household credit growth and real per capita consumption in the U.S. The chart highlights that 10-year upcycles in household debt-to-GDP correlate well, with a lag, to 10-year downcycles in real per capita spending. Periods where the relationship is less tight have tended to be associated with less severe increases in household debt-to-GDP, suggesting that investors can be more confident that debt growth will eventually negatively impact consumer spending the stronger the credit binge has been. Chart 6The Historical Experience Of Household Leveraging Does Not Paint A Pretty Picture For Canada Chart 7Upcycles In Household Leveraging Correspond To Downcycles In Real Spending As a final point, Chart 7 underscores a sobering fact: The U.S. has only seen two instances of a 3% or greater average annual rise in household debt-to-GDP over the course of a decade: the first was in the 1920s, and the second was from 1998 to 2007. Clearly, in both cases the rise in debt ended very poorly for the U.S. economy. This, along with the prevalence of serious debt crises following credit binges similar in magnitude to Canada's experience, makes it clear that a credit-driven downturn in spending is a highly probable event for the Canadian economy over the long run, rather than a risk. Bottom Line: The available historical evidence suggests that the substantial leveraging of Canadian households that has already occurred will ultimately cause a serious credit-driven downturn. Debunking Some Housing Market Myths: It's Worse Than You Think Chart 816 Years Of Too-Easy Money The risk that the Bank of Canada will eventually "over-tighten" is magnified by the fact that there is still an ongoing debate within Canada about whether any housing market imbalances even exist. Many market participants still employ several arguments about the Canadian housing market that, at first blush, appear to mitigate the risk of serious long-term consequences of Canada's debt boom. But these arguments are flawed, and an in-depth review of these fallacies highlights the economic risk of higher interest rates. Myth #1 - Sustainable Demand And Affordability The first myth about Canada's housing market is that the rise in house prices and household debt is sustainable because of how long the boom has lasted without consequence. However, besides the ominous historical experience highlighted in Charts 6 and 7 above, Chart 8 makes it clear that the substantial build-up in Canadian household debt since 2000 has occurred primarily due to too-easy monetary policy, rather than legitimate housing market fundamentals. The chart presents Canadian household debt-to-GDP versus the Bank of Canada's target for the overnight rate. The dotted line in panel 2 is a Canadian version of the well-known Taylor rule of monetary policy, with panel 3 showing the difference between the actual policy rate and that prescribed by the rule. The chart shows that the rise in household debt-to-GDP began precisely when the policy rate fell below the Taylor rule, and that this gap has persisted for the past 16 years. We acknowledge that the Bank of Canada felt it was necessary to keep interest rates relatively low during the last economic cycle because of the persistent strength in the Canadian dollar (which acts to restrain exports). But whatever drag on growth that occurred from a strong currency was not large enough to prevent low interest rates from sparking an enormous rise in household leverage. Myth #2 - No Foreign Money Effect The second myth about the Canadian housing market is that there is no substantial effect on house prices from foreign money and that, by extension, foreign transaction taxes should be discouraged. To us, the issue is not the specific residency status of a particular buyer, but rather whether the housing market is being supported by an inflow of foreign capital. While data limitations make it difficult to prove with certainty that Canada has been struck with a tidal wave of capital from China (with Hong Kong acting as the conduit), Charts 9 and 10 show that the circumstantial evidence is overwhelming. The story that emerges from the charts is that the peak in Chinese real GDP growth in 2010 marked the beginning of significant capital outflow from the country, which appears to have moved through Hong Kong, and was perhaps accelerated by Xi Jinping's crackdown on cronyism that began in 2013. Panel 2 of Chart 9 shows that the average absolute value of Hong Kong's "net errors and omissions" line from the balance of payments spiked after mid-2010,2 as did Canada's "other investment liabilities" with a lag. Chart 10 shows that this period also saw a sharp rise in visitor arrivals to Canada from China and Hong Kong, a rise in the share of Canadian bank loans to nonresidents, and a meteoric rise in house prices in Vancouver and Toronto. Chart 11 presents data from Global Financial Integrity, a Washington-based think tank that tracks illicit financial flows globally. While the data is only available with a lag, the chart shows that GFI's estimate of illicit financial outflows from China has risen significantly following the global financial crisis, which is consistent with the narrative presented in Charts 9 and 10. Chart 9Very Strong Circumstantial Evidence... Chart 10...Of Foreign Capital Inflows Chart 11Clear Evidence Of Chinese Capital Flight Myth #3 - Tight Supply The third myth concerning Canadian housing is the argument that housing supply is tight, which justifies the exponential move in house prices. First, it should be noted that while residential investment as a share of GDP was indeed low in the late-1990s, it rose back to its long-term average within the first three years of the housing boom, and has recently risen to a 27-year high (Chart 12). A similar trend can be observed in housing starts and the number of unsold housing inventories. As such, it seems difficult to make the case that the extraordinary rise in house prices and household debt that we have observed over the past 16 years is ultimately due to scarce housing supply. Chart 13 makes this point more saliently, by presenting a scatterplot of the median house price-to-income ratio versus the population density of several major global markets. Ultimately, in any true market economy, genuine housing supply constraints must be related to high density or else there would be ample room to build additional housing units. Two points are noteworthy: Chart 12There Is No Supply Problem Chart 13'There's Nowhere To Build!': Yeah, Right! The median house price-to-income ratio for Toronto and Vancouver deviate enormously from the level that would be implied by their density given the relationship across global housing markets. Based purely on this analysis of relative density, Toronto and Vancouver house prices are 80% and 140% overvalued, respectively. Around the globe, the housing markets that appear to be the most overvalued relative to population density appear to be the geographically closest to China (Vancouver, Australia, Hong Kong, and the West Coast of the U.S.), which echoes our conclusions about foreign capital inflow above. Myth #4 - A Healthier Canadian Household Debt Distribution The fourth myth concerning Canadian housing is the idea that the household debt binge that we have observed has been a "healthier" rise than what occurred in the U.S. during the last economic cycle. The argument is that the rise in debt in the U.S. from 2001 - 2007 predominantly occurred among "subprime" borrowers, and that this is not occurring in Canada. Comparing Canada to the U.S. last cycle is difficult due to the lack of data on the distribution of Canadian household debt-to-income ratios by income percentile. However, some inferences can be drawn from the OECD's wealth distribution database, and they suggest that Canadian household debt is, in fact, quite concentrated. Chart 14 presents the relationship between the number of households with debt and the median debt-to-income ratio of indebted households, from 2010 to 2012 (depending on the observation). The chart shows that while only about half of Canadian households are indebted (in line with the average of the countries shown and below that of the U.S.), among those with debt the median debt-to-income ratio is substantially higher than most other countries. This is also reflected in Chart 15, which shows that Canada has a high rank of significantly indebted households as a share of all indebted households,3 more so that the U.S. Investors should note that Canada's rank today is likely to be higher than that shown in Chart 15, given that several other highly indebted countries (such as the Netherlands and Portugal) have actually experienced household deleveraging since 2010. Chart 14High Concentration... Chart 15...Of Household Indebtedness Myth #5 - The "CMHC Backstop" The fifth and final myth concerning Canadian housing is the fact that the economy is not significantly exposed to a housing market downturn because of the Canada Mortgage and Housing Corporation's mortgage insurance coverage protects Canadian banks. It is true that the CMHC can act as a backstop for the economy by helping to mitigate mortgage default losses. But Chart 16 highlights that there have been some substantial changes over the past few years in the CMHC's footprint in the mortgage market that casts significant doubt on whether it would be able to materially blunt the losses that are likely to occur from systemic mortgage defaults. First, the chart shows that while half of mortgages in Canada had CMHC insurance coverage in 2010, this has fallen by 14 percentage points in just six years (to 36%). This means that almost 2/3rds of Canadian mortgages are not CMHC-insured. Second, while the CMHC has been aggressive in building equity over the past several years (perhaps in anticipation of a significant housing bust!), this equity buffer is still small relative to its total loans (9%) and is fractional as a share of total Canadian residential mortgage credit (1.5%). As such, while we agree that the CMHC is an effective backstop against idiosyncratic risk in the mortgage market, it is simply too small to act as a credible buffer against large-scale losses. Bottom Line: Several myths about Canada's housing market have obscured the true extent of its credit market imbalances, heightening the risk that policymakers will ultimately overplay their hand when tightening monetary conditions. When Will The Party Come To An End? From our perspective, the most likely catalysts for a credit-driven downturn in spending are a reversal of the factors that drove the rise in household debt in the first place. Chart 17 presents a three-phase view of the rise in household debt-to-income since 2000, and summarizes the major drivers of rising leverage in each phase given our analysis above: persistently easy monetary policy (phase I), fiscal and monetary easing (phase II), and foreign capital inflow (phase III). Given this, higher interest rates, fiscal drag, and/or a shock to foreign capital would appear to be the most likely triggers for a credit-driven downturn: Chart 16A Substantially Lower CMHC Footprint Chart 17The Major Drivers Of Rising Household Leverage Higher Interest Rates: Tighter monetary policy is an obvious (and most likely) trigger for a major reversal in the Canadian housing market. It is not yet clear how aggressively the Bank of Canada will raise interest rates over the coming 6-12 months, but Chart 18 highlights that the household debt service ratio will quickly rise to a new high even if the Bank of Canada hikes rates by 150 bps over a two-year period, owing to the relatively short maturity of Canadian mortgage contract terms. Still, the chart shows that this does not occur until mid-2019 at the earliest. Fiscal Drag: IMF forecasts for Canada's cyclically-adjusted primary balance suggest that government spending and investment will remain a positive contributor to growth into next year (Chart 19). But beginning in 2019, fiscal policy is forecast to become a persistent drag on growth, and it is even possible that the sharp deceleration in fiscal thrust set to occur next year could act as the proximate cause of serious problems in the Canadian housing market. Chart 18Not An Imminent Threat, But Watch Out Chart 19Fiscal Drag Set To Begin In 2019 Chart 20Macroprudential Measures Didn't Kill The Vancouver Housing Market A Domestically-Driven Shock To Foreign Capital Inflow: Some investors have pointed with concern to dramatic declines in the sales-to-listings ratios in Vancouver and Toronto following foreign taxation announcements in these markets. We agree that the impact of new or existing macroprudential measures may eventually cause a severe fallout in the housing market, but for now the experience of Vancouver suggests that such an event is not imminent. Chart 20 presents the 3- and 12-month rate of change in Vancouver house prices, with the vertical line denoting the announcement of the foreign transaction tax. While it is clear that the tax sharply slowed the rate of appreciation in Vancouver house prices, it did not cause an outright decline (the 3-month rate of change only briefly turned negative before returning to positive territory). Cyclically, we would become more concerned were we to observe a combination of additional restrictions on foreign capital inflow, higher minimum down payment thresholds for houses priced at or below median levels, and a significantly lower allowable gross/total debt service ratio. An Externally-Driven Shock To Foreign Capital Inflow: We noted earlier in the report that there is strong circumstantial evidence showing that Canada's property market is benefiting from large capital inflows from China, with Hong Kong acting as the conduit. Given this, the Canadian housing market could be subject to a shock from exogenous changes in the flow of this capital, perhaps triggered by cyclical changes in China's economy or, more likely, actions by Chinese policymakers to materially slow the pace of capital flight. While it is very difficult on a high frequency basis to track whether the impact of foreign capital on Canada's housing market is growing or weakening, the indicators shown in Charts 9 and 10 on page 9 form the basis of our monitoring effort. The list above has focused on potential triggers that are specific to the factors that led to the build-up in Canadian household debt. Clearly there are additional macro factors that could trigger the onset of a major debt payback period in Canada, and chief among these would be the next U.S. or global recession. For example, we recently noted how continued tightening from the Fed could set the stage for a U.S. recession in 2019, which could easily trigger either a prolonged period of stagnant Canadian growth or an active deleveraging event.4 Bottom Line: There are multiple potential triggers that could eventually spark a credit-driven downturn in Canada, but none of them seem likely to have a major impact on the economy over the coming 6-12 months. Investment Implications Canadian household leverage has risen enormously over the past 16 years, and a detailed analysis of Canada's housing market shows that an eventual credit-driven downturn in spending is a highly probable event for the Canadian economy over the long run (rather than a risk). However, among the most probable triggers for a serious housing market shock, only higher interest rates are set to occur over the coming year. Given that monetary tightening will be gradual in its pace, it does not seem probable that a major downturn in spending is imminent. From an investment standpoint, these conclusions imply the following stance towards Canadian dollar assets over the coming 6-12 months: Overweight the Canadian dollar: The cyclical improvement in the Canadian economy, along with our bullish view on oil prices, suggests that the Canadian dollar is set to appreciate over the coming year. We acknowledge that our constructive view on oil prices is contrarian and that, for now, we are ahead of the market. Continued weakness in oil prices remains the chief risk to a bullish stance on the CAD. But our detailed analysis of the global oil market strongly implies that the current level of oil inventories is too high and is set to draw materially over the coming months, which will be undoubtedly positive for oil prices barring the development of a major global demand shock. Maintain Canadian equities on upgrade watch: Canadian equities have materially underperformed their global peers over the past six years, due to fairly significant de-rating from overvalued levels as well as a downtrend in relative 12-month forward earnings (mostly vs the U.S.; Chart 21). Relative performance in common-currency terms has also been hurt by a declining Canadian dollar. Looking out over the next year, there are at least some tentative signs to be optimistic about Canadian stocks. First, Chart 22 highlights that Canadian stocks are now moderately cheap relative to their global peers based on a composite valuation indicator. Second, our expectation of an uptrend in oil prices would likely bolster relative forward earnings, and could act as a re-rating catalyst for the broad market. Chart 21Multiples And Earnings Have Worked Against Canadian Stocks Chart 22No Longer Expensive Underweight Canadian bonds within a hedged global fixed-income portfolio: Canadian government bonds have recently underperformed their global peers, and this trend is likely to continue in response to tighter monetary policy. Over the longer term, the likelihood of a major credit-driven downturn in spending means that the secular investment implications for Canada are precisely the opposite of that described above. This means that investors should pursue a "two-staged" approach to investing in Canadian assets. The fact that the Canadian economy is currently accelerating and a significant reversal in the Canadian housing market does not seem to be imminent means that there is an opportunity for Canadian assets to potentially outperform (or underperform in the case of government bonds) over the coming 6-12 months. Such a period of cyclical improvement would likely (temporarily) dampen investor concerns about a major housing market correction, creating much better "selling conditions" for Canadian risky assets than from current levels. We acknowledge that the "two-stage" nature of this strategy is nuanced, and we have provided a checklist of potential triggers for the housing market in this report so that investors can gauge the likelihood that a material payback period is about to begin. We will continue to monitor both the cyclical improvement in the Canadian economy and the magnitude of imbalances in the household sector, and will provide investors with regular updates as they develop. Stay tuned! Bottom Line: Investors should pursue a "two-staged" approach when allocating to Canadian assets. Favor a pro-cyclical stance over the coming 6-12 months, but look to shift to a bearish structural view at some point beyond the immediate investment horizon. Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com Appendix A A Quick Recap Of Home Capital: Not A Systemic Issue In April, the share price of Home Capital Group (a Canadian non-bank mortgage lender) collapsed by 75% in response to a major liquidity crisis for the firm. The crisis ultimately stemmed from a set of mortgage loans with falsified income documentation, which to many outside observers was strongly reminiscent of the aberrant practices of U.S. subprime lending institutions during the last cycle that eventually spawned the global financial crisis. However, as highlighted below, Home Capital Group's problems were largely idiosyncratic (i.e., not systemic) in nature: Home Capital's business model involves lending to Canadians who lack a stable credit history, but who are generally otherwise creditworthy (commonly referred to as "near-prime" borrowers). Since these borrowers subsequently build a credit history by staying current on their mortgage loans with Home Capital, they often switch to a big-five bank after the term of the loan is complete. As such, Home Capital faces substantial client retention challenges, which is an idiosyncratic income statement problem rather than a balance sheet problem with systemic implications. To combat the tendency of its loan book to shrink, in 2014 Home Capital increased the size of its sales force by partnering with a set of established mortgage brokers. Some of the loans that had been originated by these brokers had falsified income documentation, which led to an internal investigation. Following the investigation, the company failed to disclose the results to investors during a period where the company's operating performance was impacted by the fraud. This eventually led to enforcement action from the Ontario Securities Commission. The disclosure of enforcement, along with several other events (such as the termination of its CEO in late-March) severely eroded investor confidence in the firm and essentially caused a bank run. From a macro perspective, there are two important takeaways from this series of events. First, it is important to note that Home Capital experienced a liquidity rather than a solvency crisis. While the former can, of course, lead to the latter, the run on Home Capital did not occur because of deteriorating loan performance, unlike what occurred in the U.S. with the subprime market. Indeed, Home Capital's first quarter results show that net impaired loans as a percent of gross loans have continued to trend lower over the past several quarters (Chart A1). Second, the fact that Home Capital's mortgage book tends to shrink underscores the underlying creditworthiness of at least some of its borrowers, because these households would probably not be able to shift their mortgages to the big-five banks if loan qualification was an issue. As a final point, Chart A2 presents some perspective about the apparent prevalence of mortgage fraud in Canada by showing the number of U.S. mortgage loan fraud suspicious activity reports (SARs) in the lead-up to the subprime financial crisis. The chart not only shows the sharp rise in the number of SARs from 2002-2003 to 2007-2008, but it also shows that the volume of reports numbered in the tens of thousands. By contrast, Canadian news stories reporting on a rise in the number of mortgage fraud complaints in Canada quote a trivially small number of cases. For example, a recent article from the Vancouver Sun stated that British Colombia's Financial Institutions Commission statistics "show complaints roughly doubled from 109 in 2013 to about 200 in 2016, and about a third of complaints allege loan application fraud."5 Chart A1No Deterioration In Loan Performance Chart A2No Evidence That This Is Happening In Canada While it is technically correct to state that this is a doubling in the rate of fraud cases, it is from what appears to be an extremely small base. Adjusting by a factor of 10 to account for the difference in population, Canada would need to see 3,000-to-6,000 cases of mortgage fraud per year in order to be comparable to what occurred in the U.S. in the latter half of the housing market bubble. There is simply no evidence that mortgage fraud on this scale of magnitude is occurring. 1 See Appendix A on page 19 for a review of the Home Capital debacle and why concerns of systemic mortgage fraud are quite likely overblown. 2 If Hong Kong has been a conduit for capital flight from China, the flow of capital would only temporarily show up in Hong Kong's balance of payments. For example, one quarter of significant capital inflow might be followed by a quarter of significant capital outflow as the money enters from China and exits towards the rest of the world. As such, we use the absolute value of Hong Kong's net errors and omissions line to see whether the magnitude of the flow has increased. 3 Defined as having a debt-to-income ratio in excess of 3. 4 Please see Global Investment Strategy Weekly Report, "The Timing Of The Next Recession," dated June 16, 2017, available at gis.bcaresearch.com. 5 Sam Cooper, "Regulator Tracks The Rise In Mortgage Fraud Complaints In B.C. As House Prices Jump," Vancouver Sun, June 19, 2017. Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights EM growth is set to falter due to budding weakness in Asia's trade, a decline in commodities prices, and the frailty of EM banking systems. U.S./DM bond yields are heading higher for now and China's money/credit growth is set to decelerate. Together, these will trigger a selloff in EM risk assets. The EM equity outperformance versus DM has been extremely narrow and, hence, it is unsustainable. The EM tech sector is unlikely to support the equity rally much further because these stocks are overbought, and the Asian semiconductor cycle is entering a soft patch. Take profits on the yield curve flattening trade in Mexico. Stay long MXN on crosses versus BRL and ZAR and continue overweighting Mexican bonds. Feature Higher bond yields within the advanced economies and policy tightening in China remain the key threats to EM risk assets in the near term (the next three months). In the medium-term (the next three to 12 months or so), the principle risk is weaker growth in EM/China, and hence contracting corporate profits in EM. While this rally has lasted longer and has gone further than we had anticipated, we find the risk-reward for EM risk assets extremely unattractive. In fact, the huge amount of money that has flown into EM equity and debt markets in the past year amid poor fundamentals suggests to us that the next move will not be a simple correction but rather a major bear market. EM Recovery To Falter Although on the surface global growth appears to be on solid footing, there are early signs of a slowdown in Asian exports. Both Taiwanese exports of electronic parts and the country's overall exports to China have rolled over - the latter two lead global export volumes by a few months, as shown in Chart I-1. The reason why Taiwanese exports of electronic parts lead global trade cycles is because these parts are used in the assembly of final products, and producers order and receive these parts before final products are made and shipped. Similarly, a lot of Taiwanese exports to China serve as inputs into final products assembled in China and shipped worldwide. This is why Taiwan's overall shipments to China lead global trade cycles. On top of this, Korea's overall manufacturing and semiconductor shipments-to-inventory ratios have relapsed. Historically, these ratios have correlated with the KOSPI (Chart I-2). Chart I-1Signs Of Slowdown ##br##In Asian Trade Chart I-2Korea's Manufacturing ##br##Growth Has Peaked Outside the manufacturing-based Asian economies, most other EMs are basically commodities plays, except for India and Turkey. The latter two countries are not only relatively small, but Indian stocks are also expensive and overbought while Turkey is sufferings from its own malaise. In short, if the Asian tech cycle rolls over, China slows down and commodities prices relapse, EM growth will falter. That is why the focus of our analysis has been and remains on China's growth, commodities prices and the Asian trade cycle. Meanwhile, many banking systems in the developing world remain frail following the credit excesses of the preceding years. BCA's Emerging Markets Strategy service remains bearish on commodities, and believes the breakdown in the correlation between commodities prices and EM risk assets since the beginning of this year is temporary and unsustainable. As for the increased importance of the technology sector in the EM equity benchmark, we offer further analysis on page 10. Our negative view on EM growth is not contingent on a relapse in U.S. and euro area growth. In fact, our current baseline scenario is that DM growth will remain solid, and government bond yields in these markets will rise further. Although growth in both the U.S. and euro area is robust, their importance for EM has become small. For example, exports to the U.S. and EU altogether account for 35% of total exports in China, 22% in Korea and 20% in Taiwan. All in all, if commodities prices continue to downshift and Asian trade slows, as we expect, EM growth will decelerate. Bottom Line: EM growth is set to falter notably, despite solid demand growth in DM. Liquidity Backdrop To Deteriorate Investors and market commentators often use the term "liquidity" loosely, and denote numerous things by it. We use the term 'liquidity' to signify the level and/or direction of interest rates as well as the level and/or direction of money/credit growth. Below we review some different perspectives of liquidity: EM narrow money (M1) growth points to both lower share prices and a relapse in EPS growth in the months ahead (Chart I-3). Chart I-3EM: Narrow Money (M1) Points To EPS Downturn And Lower Stock Prices This is an equity market cap-weighted aggregate of narrow money growth. M1 growth in China - the largest market cap in the EM equity benchmark - has been essential in driving aggregate EM M1 cycles in recent years. More importantly, China has been tightening liquidity, yet the majority of investors remain complacent about its impact on growth. In this regard, investors should remind themselves that monetary policy works with time lags, and the considerable rise in China's interbank rates and corporate bond yields will produce a growth slowdown in the real economy later this year. Chart I-4 demonstrates that China's broad money growth (M2) - which has in effect dropped to an all-time low - leads bank and non-bank credit origination. This suggests the odds of a slowdown in bank and non-bank credit flows are considerable. There has been no stable correlation between the size of DM central banks' balance sheets and EM stock prices, bond yields and currencies since 2011. Therefore, the Fed's move to reduce its balance sheet by itself should not matter for EM risk assets from a fundamental perspective. Nevertheless, EM risk assets have been negatively correlated with U.S. TIPS yields (Chart I-5), and the potential further rise in U.S./DM real and nominal yields will hurt EM sentiment, with flows to EM drying up. Chart I-4China: M2 Heralds ##br##Slowdown In Credit Growth Chart I-5EM Currencies To Depreciate ##br##As U.S. Real Yields Drift Higher Importantly, traders' bets on U.S. yield curve flattening have risen, as evidenced by large short positions in 2-year U.S. notes and considerable long positions in 10- and 30-year bonds. The unwinding of these positions will drive bond yields higher. Chart I-6Precious Metals Signal ##br##Higher Real Yields Ahead Notably, precious metal prices have failed to break out amid a weak U.S. dollar and have lately relapsed (Chart I-6). Precious metals prices could be sensing a further rise in U.S. real yields and/or an upleg in the U.S. dollar. Both the rise in U.S. yields and a stronger dollar will be negative for EM. Bottom Line: We maintain that U.S./DM bond yields are heading higher in the months ahead and China's money/credit growth is set to decelerate. Altogether these will trigger a selloff in EM risk assets. Underwhelming EM Technicals It is a well-known fact that flows into EM debt funds have been enormous, making EM fixed-income markets vulnerable to a reversal of these flows at the hands of tightening liquidity and EM growth disappointments, as argued above. This section focuses on a number of bearish technical signals for EM share prices. In particular: The EM equity implied volatility curve - 12-month VOL minus 1-month VOL - is at a record steep level, based on available history (Chart I-7). Periods of VOL curve flattening have historically coincided with a selloff in EM share prices, as evidenced by Chart I-7. Given that the VOL curve is record steep, the odds of flattening are substantial. Consistently, the probability of an EM selloff is considerable. Chart I-7A Sign Of Top In EM Share Prices? In absolute terms, EM equity implied 1-month VOL is at an all-time low and reflects enormous complacency about EM. EM equity breadth has also been poor. The MSCI EM equally weighted stock index (where each stock commands an equal weight) has considerably underperformed the EM market cap-weighted index since May 2016 (Chart I-8). This suggests the EM rally has been very narrowly driven. The same measure for DM stocks has done relatively better (Chart I-8). Remarkably, EM has underperformed DM based on equal-weighted equity indexes since July 2016 (Chart I-9). This confirms that EM outperformance against DM since early this year has been largely driven by a few stocks, namely the five companies accounting for the bulk of the EM tech index. Furthermore, EM ex-tech stocks have also failed to establish a bull market, in that the index remains below its prior low (Chart I-10). Chart I-8EM Equity Breadth ##br##Has Been Poor Chart I-9EM Versus DM: Relative ##br##Equity Performance Chart I-10EM Ex-Technology Stocks: ##br##Rebound But No Bull Market Finally, the magnitude of the EM rally this year is somewhat misleading. Only three out of 11 sectors - technology, real estate and consumer discretionary (mainly, autos) - have outperformed the EM benchmark this year. Table I-1 illustrates that these three sectors have been responsible for about 50% of the EM rally year-to-date while their market cap is only 36% of total. Table I-1EM Rally In 2017: Return Decomposition Bottom Line: The EM equity outperformance versus DM has been extremely narrow: it has been due to five tech companies that are currently very overbought (see Chart I-8 on page 7). Valuations EM equity valuations are not cheap, as most of the rally since the early 2016 bottom has been driven by a multiple expansion rather than a rise in corporate earnings (Chart I-11). We are not suggesting EM stocks are expensive, but they do not offer good value either. In fact, good companies/countries/sectors are expensive, while those, that appear "cheap", command low multiples for a reason. As for currencies, they are not cheap either. The real effective exchange rate of EM ex-China is rather elevated after the rally of the past year or so (Chart I-12). Finally, not only are EM sovereign and corporate spreads close to record lows, but also local government bond yield spreads over U.S. Treasurys are at multi-year lows (Chart I-13). Chart I-11Decomposing EM Equity ##br##Return Into P/E And EPS Chart I-12EM Ex-China Currencies ##br##Are Not Cheap And Vulnerable Chart I-13EM Local Bond Yields Spreads ##br##Over U.S. Treasurys Is Low Bottom Line: Adjusted for fundamentals, EM equity, currency and credit market valuations are rather expensive. The odds are that the reality will underwhelm expectations, and that EM risk assets will sell off. A Word On EM Tech: Is This Time Different? During our recent trip to Europe, many clients argued that the increased weight of technology in the EM equity benchmark will cause EM share prices to decouple from the traditional variables they have historically been correlated with, like commodities prices, commodities stocks and others. In brief, the argument is that EM has entered a new paradigm, and past correlations will not work. The last time we at BCA heard similar arguments was back in early 2000 at the peak of the global tech bubble. At the time, the argument was that this time was truly different - that tech stocks could drive the market higher regardless of the old indicators and the performance of other sectors. Chart I-14 portrays that in 2000 the EM equity index, for several months, decoupled from global mining and energy stocks when tech and telecom stocks went ballistic. Chart I-14EM And Commodities Stocks: Can The Recent Decoupling Persist? Back in 2000, the bubble was in tech and telecom stocks. These two sectors together comprised 33% of the EM benchmark as of January 2000 (Chart I-15). This compares with a 27% weighting of technology stocks alone in the EM benchmark now. The combined weight of energy and materials is currently 14% versus 19% in January 2000, as can been seen in Chart I-15. Chart I-15EM Equities Sector Composition Now And In Late 1990s To be sure, we are not suggesting that tech stocks are in a bubble as they were in 2000, and that a bust in share prices is imminent. However, several observations are noteworthy: Chart I-16EM Equities Sector ##br##Composition Now And In Late 1990s Just because EM tech stocks have skyrocketed in the past six months does not mean they will continue to do so. In fact, EM tech is already extremely overbought and likely over-owned (Chart I-16). As global bond yields rise, high-multiples stocks, especially social media/internet companies, could selloff. We, like all macro strategists, can add little value on how to value internet/social media companies and assess their business models. However, we can shed some light on the business cycle in the semiconductor sector that influences performance of heavyweight companies like TSMC and Samsung. As Chart I-1 and I-2 on pages 1 and 3 demonstrate, there are signs that the semi/electronics cycle in Asia has peaked. We do not mean that this sector is headed toward recession. But this is a very cyclical sector, and some slowdown is to be expected following the growth outburst of the past 18 months. This will be enough to cause a correction in semi stocks from extremely overbought levels. The tight correlation between EM share prices and energy and mining stocks has persisted for the past 20 years (Chart I-14 on page 10), and we believe it will re-establish as technology stocks' shine diminishes. Finally, we have been recommending an overweight position in Taiwanese, Korean, and Chinese stocks primarily because of their large tech exposure. For now we maintain this strategy. Bottom Line: While the technology sector could make a difference for EM economies and equity markets in the long run, it is unlikely to support the current rally and outperformance much further. Indeed, tech stocks are heavily overbought, and the Asian semiconductor cycle is entering a soft patch. In brief, the overall EM equity benchmark is at a major risk of relapse and underperformance versus the DM bourses. Stay underweight. Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com Mexico: Take Profits On Yield Curve Flattener And Stay Long MXN On Crosses Mexico's 10/1-year swap curve has inverted for the first time in history and we are taking a 160 basis points profit on our yield curve flattening trade recommended on June 8, 2016 (Chart II-1). Will the central bank begin cutting interest rates soon? Is it time to get bullish on stocks? We do not think so: Inflation is well above the central bank's target and is broad based (Chart II-2). Notably, wage growth is elevated (Chart II-3). Chart II-1Mexico's Yield Cruve Has Inverted: Take Profits Chart II-2Mexico: Inflation Is Above The Target Chart II-3Mexico: Wage Inflation Is High Provided productivity growth is meager in Mexico, unit labor costs - which are calculated as wage per hour divided by productivity (output per hour) - are rising. This will depress companies' profit margins and make them eager to hike selling prices. This will in turn prevent inflation from falling and, consequently, hamper Banxico's ability to cut rates for now. Meanwhile, the impact of higher interest rates will continue filtering through the economy. High interest rates entail further slowdown in money and credit growth and, hence, in domestic demand. Both consumer spending and capital expenditure by companies are set to weaken a lot (Chart II-4). This will weigh on corporate profits and share prices. Fiscal policy is not going to support growth either because policymakers will opt to consolidate the recent improvement in the fiscal deficit. This is especially true given the latest selloff in oil prices. Notably, oil accounts for about 20% of government revenues. Even though non-oil exports and manufacturing output are accelerating (Chart II-5), non-oil exports - that make about 30% of GDP - are not large enough to offset the deceleration in domestic demand from monetary tightening. Chart II-4Mexico: Domestic Demand To Buckle Chart II-5Mexico: Exports Are Robust Investment Conclusions The outlook for Mexican stocks in absolute terms is poor as domestic demand will slump, further hampering corporate profits. Meanwhile, inflation is still elevated to justify rate cuts by the central bank. Within an EM equity portfolio, we recommend neutral allocation to this bourse mainly due to our expectations of the peso outperforming other EM currencies. The Mexican peso is still cheap (Chart II-6). Therefore, we continue to recommend long positions in MXN versus ZAR and BRL. If EM currencies depreciate and oil prices drop further as we expect, it will be hard for the peso to appreciate versus the U.S. dollar. However, the peso will outperform many other EM currencies. Mexican local currency bonds and sovereign credit offer good value relative to their EM counterparts. (Chart II-7). Fixed income investors should continue to overweight Mexican local currency and sovereign credit within their respective EM benchmarks. Chart II-6Mexico: Peso Is Cheap Chart II-7Continue Overweighting Mexican Bonds Stephan Gabillard, Senior Analyst stephang@bcaresearch.com Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights EM equity breadth has moved into negative territory, DM-based excess liquidity measures are set to roll-over, and China-based liquidity measures are also weak. Individually, each of these factors are not enough to raise alarm bells, but together they point to a period of heightened risks for EM assets and commodity currencies. AUD/CAD and NZD/JPY are set to suffer in this environment. EUR/USD will rise to 1.15-1.16, but unlike in 2015, it should not receive much of a fillip from EM volatility. Feature Chart I-1Technical Risk In EM An interesting development has unfolded in emerging markets. While the pause in the EM rally has hit investors' radar screens, the more puzzling event concerns breadth. Not only has the advanced/decline line rolled over, but more worrisomely, it has recently moved into negative territory. Historically, when more stocks are declining rather than advancing, EM equities tend to experience sharp selloffs (Chart I-1). This development is important when put into a global context. EM stocks and related assets like commodity currencies have been buoyed by plentiful global liquidity conditions. However, global liquidity is set to deteriorate. A rocky second half may emerge in EM assets. Global Liquidity Is Slowing Following in the Federal Reserve's footsteps, DM central banks are moving away from monetary accommodation. Last week, European Central Bank President Mario Draghi made a speech that was interpreted as representing an abandonment of the ECB's dovish bias. With the anticipation that its bond-buying program will be tapered early in 2018 and reports that the ECB is having problems buying its quota of German and Finnish bonds, global bonds suffered, with Bund and T-Note yields moving up 33 and 23 basis points since June 27, respectively. The ECB is not the only central bank to have changed its tack. The Bank of Canada's communications have been crystal clear that it intends to increase rates this summer, or early fall at the latest. Even the perennially dovish Riksbank is moving away from its easy bias, as Sweden's resource utilization points to a continued acceleration in core inflation. But does this even matter? The global economy is strong, and beginning to remove accommodation is not quite the same thing as pushing rates into tight territory. The advanced economies are unlikely to suffer much from this development. However, the picture for EM is more concerning. Some key leading indicators of EM activity have already begun to roll over. For example, Taiwanese IP, a key bellwether of overall EM strength, is now contracting on a year-on-year basis (Chart I-2, top panel). Meanwhile EM PMIs rolled over three months ago and EM narrow money growth, a key forecaster of EM profits, is slowing sharply (Chart I-2, bottom panel). Despite these negative developments, EM stocks have remained resilient. The factor underpinning this impressive performance has been the rise in global liquidity. More technically, the rise in the global Marshallian K - the ratio of money to nominal GDP - over the past six months. Excess money has had to go somewhere. Among the many refuges, EM has been a key pole of attraction, with massive inflows supporting assets prices. The 8% appreciation in EM currencies versus the dollar since their January 2016 trough has been a vivid illustration of this phenomenon. The driver of the rise in excess money has been the ratio's numerator, dollar-based liquidity. The Fed's various QE programs were key determinants of dollar-based liquidity (Chart I-3). However, its tapering in late 2014 was enough to prompt a contraction of the measure. Now that the Fed is intent on decreasing its balance sheet while the ECB tapers and other smaller DM central banks begin increasing rates, the small improvement witnessed in the past three months is likely to end. The recent weakness in gold prices, despite the softness in the dollar, could be a sign that markets are beginning to sniff out the imminent tightening of global liquidity conditions. Chart I-2EM/China Profits Growth To Roll Over (I) EM Growth ##br##Has Deteriorated, Profits Will Suffer Chart I-3The Fed Balance Sheet Runoff ##br##Will Hurt Global Liquidity Additionally, not only are global central banks, led by the Fed, tightening or looking to tighten policy, they are doing so despite an absence of actual inflation. As a result, this means DM real yields are set to rise. As Chart I-4 illustrates, rising real DM yields have historically been a harbinger of poor EM bond performance. In fact, the action in DM real yields since mid-2016 already points to a problematic second half for EM bonds. As a result, EM bond investors are likely to suffer some losses in the coming months. Such losses would not only tighten EM financial conditions, but would also be symptomatic of capital leaving the region. Less money in those markets simply means less liquidity. With EM corporate spreads near historical lows, a repricing of credit risk on the back of softening global and EM liquidity is likely to prompt both a selloff in EM stocks and in EM currencies (Chart I-5). As a result, DM commodity currencies, the NZD and AUD in particular, could suffer. Chart I-4EM Financial Conditions##br## Are Set To Deteriorate Chart I-5If Liquidity Dries, Spreads Widen ##br##And EM Stocks Fall Bottom Line: In November 2016, a new leg of the EM rally began - a move driven by an expansion in global liquidity, even as a key bellwether of EM economic activity rolled over in the interim. Global excess liquidity is set to roll over as DM central banks abandon their dovish biases and the Fed begins to let its balance sheet run off. With EM weaker from a technical perspective, the second half of 2017 could be a tough environment for EM plays. Chinese Liquidity Joins The Fray In May 2015, EM equities in U.S.-dollar terms peaked just before global liquidity began to roll over. Compounding the risks, back then Chinese economic conditions were also problematic. Excess capacity and massive deflationary forces were wearing down on profits and investment. China is thus another key factor to watch. In this optic, beyond DM liquidity, a key driver of the rebound in EM last year was actually Chinese liquidity conditions. In the second half of 2015, China's own Marshallian K - based on M2 relative to nominal GDP growth - was rebounding sharply, as the PBoC was easing policy and the fiscal authorities were pressing on the gas pedal, expanding both public expenditures and pushing credit growth through the economy. However, that was then. Today, China has joined the tightening party. The quarterly moving average of Chinese interbank rates has increased by 100 basis points over the past year. Crackdowns on real estate and excess leverage have also resumed. Most importantly, the issuance of bonds by small and medium banks - a key source of grease to total social financing - has also massively decelerated, which points to a sharp slowdown and even a contraction in the Chinese credit impulse (Chart I-6). Thanks to this development, the Chinese Marshallian K is now in negative territory. The global impact of tighter Chinese monetary conditions is also flashing a red flag. Our indicator is based on the relative performance of Chinese bank stocks and USD/HKD. Underperformance of Chinese banks tends to send warning signs that tightening policy is beginning to negatively affect the outlook for Chinese credit growth. Additionally, USD/HKD is at an 18-month high because Hong Kong interest rates have not been able to follow U.S. ones, as loan demand by mainland-China entities has been poor. Most of the time, this indicator tends to move with EM stock prices, providing very little information. However, as Chart I-7 illustrates, this gauge is at its most useful when it diverges from EM equity prices. In each case, such as in 2007, 2011, and 2014, the divergences between the falling price-based Chinese liquidity indicator and rising EM stock prices was resolved by a correction in the latter. Today, the indicator points to a large amount of downside risk for EM stocks. Chart I-6Chinese Credit Impulse Will Slow Chart I-7A Worrying Divergence Again, it is important to reiterate that in and of itself, such a divergence is not enough to prompt investors to run for the hills and ditch EM stocks and related plays. However, when this happens as DM liquidity is also set to deteriorate, and most crucially, when EM breadth turns negative, decreasing EM exposure makes sense. Bottom Line: Chinese liquidity conditions are also deteriorating. The People's Bank of China may not want to push the economy into another slowdown cycle, which will most likely limit how far the Chinese central bank will tighten policy. However, this tightening has not been priced in by EM equities, and is happening as DM central banks are also reducing accommodation and as EM breadth has greatly deteriorated. A sizeable correction in EM plays is becoming increasingly likely. Investment Implications Chart I-8Global Liquidity Leads EM ##br##By More Than A Year A tightening of dollar-based liquidity and Chinese-based liquidity is a big problem for non-China EM economies. EM economies outside of China and OPEC nations still run an annual current account deficit of more than US$200 billion. They need liquidity. Moreover, they still have at least US$3.6 trillion in foreign-currency debt. With liquidity conditions deteriorating, we should expect a widening of EM spreads, falling EM stock prices and falling commodity currencies. In fact, we are today in the window of maximum risk. Chart I-8 shows the combined G7 and Chinese Marshallian K, standardized. This indicator tends to have long leads over EM equity prices. It turned negative in the summer of 2006, though EM stock prices did not peak until the fourth quarter of 2007. It turned negative again in the early days of 2010, but EM equity prices did not peak until April 2011. The indicator moved below zero in mid-2014, yet EM equities only sold off in the second quarter of 2015. This time around, the combined liquidity indicator became negative in early 2016, suggesting great risks for EM assets and related plays in the second half of 2017. High carry EM currencies like the BRL or the TRY are at risk. The ZAR looks especially poorly positioned as well but the RUB seems better cushioned against these risks. The MXN could suffer too as Mexico has a lot of U.S. dollar-denominated debt. Nonetheless, MXN remains much cheaper than the BRL and could still outperform its Brazilian brethren. The SGD is very sensitive to global liquidity conditions, as Singapore is a key banking center for EM, and could also suffer substantially against the USD. In terms of timing for the G10 currency markets, the deterioration of EM breadth has historically been a dangerous sign for commodity currencies (Chart I-9). This combination of deteriorating liquidity and breadth is often associated with a sharp selloff in NZD/JPY (Chart I-10). Investors should short this cross, and we are re-opening this trade this week. Chart I-9Commodity Currencies##br## Prefer A Fresh Breadth... Chart I-10...So Does ##br##NZD/JPY The dynamics highlighted above also explain why despite our positive stance on Canada and the CAD, we are not willing to chase the selloff in USD/CAD further, and prefer to play the CAD's strength through its crosses. The risk-reward ratio seems better this way, as we are not as negatively exposed to an EM selloff as we would be buying the CAD against the USD. Indeed, a cleaner way to play the BoC's change of tone while gaining exposure to an EM-risk off theme, is to short AUD/CAD, a trade that is already on our book. On the domestic front, this week the Reserve Bank of Australia disappointed markets and did not try to indicate a change in stance away from its dovish bias. Markets have taken notice, with the AUD incapable of rallying against a weak USD, despite very strong trade data yesterday. Meanwhile, the BoC is telegraphing a rate hike in the very near future. Additionally, an abnormal gap has emerged between AUD/CAD and AUD/USD. As Chart I-11 shows, historically, AUD/CAD and AUD/USD have tracked one another. This makes sense. The Australian economy is very levered to Asian growth and liquidity dynamics, while Canada is a crucial link in the North American supply chain. With the U.S. and Canadian business cycles so tightly integrated, the CAD tends to mimic the greenback when compared to non-USD currencies. Chart I-11AUD/CAD Is A Short The points in time when AUD/CAD has been much stronger than the AUD/USD deserve closer attention. They are periods of booms in EM Asia, such as the middle of the 1990s, or 2004 to 2005. Today, AUD/CAD is again out of line with AUD/USD, reflecting the boom in EM assets prices in 2016 and in the first half of 2017. However, if our view is correct that EM is entering a dangerous zone, AUD/CAD should weaken further. Chart I-12When Investors Are Short, ##br##EUR/USD Likes EM Selloffs Last but certainly not least the euro. EUR/USD has much momentum and could continue to rally into the 1.15-1.16 zone. In fact, historically, EM shocks have been able to lift the euro, albeit temporarily. This definitely was the case in 2015 when EM sold off: in April 2015, when EM began to weaken, in August 2015, when a temporary selling climax emerged after the Chinese floated the CNY, and in December 2015, after the Fed hiked. The euro spiked in all three instances. However, investors were very short EUR/USD entering each of these periods, and the ensuing rallies were short-covering rallies (Chart I-12). This time around, investors are very long the euro, suggesting that the euro has not been used as a funding vehicle to the same extent as it was in 2015. Additionally, in all these previous episodes, EUR/USD traded at a small discount to the fair value implied by real rate differentials, today it is trading at a premium. Thus, the same kind of short-covering rally is unlikely. As a result, we do not anticipate EUR/USD to break out of its range on the back of an EM risk-off event. That being said, EUR could outperform GBP in this type of environment. The pound remains very dependent on global liquidity conditions to finance its current account deficit of more than 4% of GDP. With big financial institutions announcing more divesture from the U.K., these hot-money flows could prove even more crucial. As a result, we are removing our call to short EUR/GBP if it moves above 0.88, and expect a move in EUR/GBP toward 0.92-0.93 in the second half of 2017. 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 greenback slipped on weak as the ADP employment, the ISM-non manufacturing employment component, and continuing as well as initial jobless claims all underperformed expectations. While the dollar reacted negatively to this news, the Fed's hawkish stance should ultimately help the USD. Supplementing the increases in interest rates, are plans to reverse the multi-year quantitative easing program.The FOMC is also increasingly worried about the "quite high" stock valuations which, could lead to financial instability. U.S. 10-year yields have gone up 4 basis points following the release of the minutes, after the 20 bps spike following initial Fed comments on June 27. Report Links: Who Hikes Next? - June 30, 2017 Look Ahead, Not Back - June 9, 2017 Capacity Explosion = Inflation Implosion - June 2, 2017 The Euro Chart II-3EUR Technicals 1 Chart II-4EUR Technicals 2 The euro's strength extends as the union experienced strong services and composite PMI measures this Wednesday. While it is true that the ECB may be looking to draw back its excessively easy monetary policy, Draghi and Praet have highlighted that accommodative policy is still needed as inflationary pressures are not yet entrenched. The euro's recent appreciation and weak producer price numbers could vindicate this view. The euro's strength has also weighed on manufacturing activity, as PMIs underperformed expectations. This is likely to weigh on EUR/USD going forward, especially as European stocks have been underperofming U.S. ones in recent weeks. EUR/SEK can face considerable pressure ahead due to the Riksbank's change in rhetoric. Report Links: Who Hikes Next? - June 30, 2017 A Market Update: June 23, 2017 Europe's Divine Comedy Part II: Italy In Purgatorio - June 21, 2017 The Yen Chart II-5JPY Technicals 1 Chart II-6JPY Technicals 2 Recent data in Japan has been mixed: National inflation came in at 0.4%, while Tokyo ex fresh food and energy inflation contracted by 0.2%. Both of these measures underperformed expectations. On the other hand, Japan's job-to-applicant ratio continues to climb, coming in at 1.49, and outperforming expectations. This last data point is key, as it highlights that the Japanese labor market is very tight, and that the stage is set for inflation to come back to Japan. However, as evidenced by the recent disappointments in data, the currency holds the key to unleash inflation in Japan. Thus, not only is a selloff in the yen needed for inflation to remerge, but this selloff would feed on itself, as a falling currency and a tight labor market would raise inflation (and thus lower real rates, as Japanese 10-year rates are anchored at 0), which would push the yen down further. Report Links: Who Hikes Next? - June 30, 2017 A Market Update: June 23, 2017 Updating Our Intermediate Timing Models - April 28, 2017 British Pound Chart II-7GBP Technicals 1 Chart II-8GBP Technicals 2 Recent data in the U.K. has been mixed: Markit manufacturing PMI declined from last month's reading and also came in below expectations at 54.3. Construction PMI also declined and came in below expectations at 53.4 However credit had a strong showing as net lending to individuals, consumer credit and mortgage approvals all came in above expectations at 5.3 billion pounds, 1.73 billion pounds and 65 thousand respectively. Various BoE members have stated that rising interest rates might be necessary to keep a lid on the island's high inflation. Although there are still some voices within the BoE who are more cautious, given the uncertainty that Brexit poses, overall the BoE has shown a much more hawkish tone in recent weeks. Report Links: Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 The Last Innings Of The Dollar Correction - April 21, 2017 Australian Dollar Chart II-9AUD Technicals 1 Chart II-10AUD Technicals 2 The AUD has experienced considerable weakness this week, following a drawback in inflation estimates for June by the TD Securities measure, of 2.3% from 2.8% and a less hawkish than anticipated RBA. While retail sales beat expectations of 0.2% - coming in at 0.6% - the pace of appreciation in the RBA Commodity Index in SDR terms continues to slow Nevertheless, these factors were not the only contributors to the recent AUD weakness. Australia remains highly levered to emerging markets, and the Fed tightening remains a major risk for the AUD. Report Links: Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 U.S. Households Remain In The Driver's Seat - March 31, 2017 New Zealand Dollar Chart II-11NZD Technicals 1 Chart II-12NZD Technicals 2 Recent data in New Zealand has been mixed: The annual trade balance underperformed expectations, coming in at a deficit of 3.75 billion U.S. dollars. However the ANZ business confidence index continued climbing, and now stands at the highest level in 8 months Overall the New Zealand economy continues to be one of the best performing in the G10. If one were to be guided merely by domestic factors, the RBNZ should be the next central bank to hike after the Fed. However the picture is slightly more nuanced, as the RBNZ is still worried about foreign developments, particularly EM weakness. This justifies why they continue to state that "monetary policy will remain accommodative for a considerable period". Thus, we continue to be bullish on the NZD against the AUD, while we are shorting it against the JPY, as a mean to benefit from a potential EM dislocation. Report Links: Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 U.S. Households Remain In The Driver's Seat - March 31, 2017 Canadian Dollar Chart II-13CAD Technicals 1 Chart II-14CAD Technicals 2 USD/CAD has broken down from a crucial technical level following Poloz's statements about the Canadian economy. He states that the "cuts have done their job". When asked about oil, the reply was reassuring, declaring that the expected level of WTI is at USD 40-50 bbl, which implies that fluctuations within that band should not influence movements the BoC path, helping the CAD in the process. He also suggested that "the adjustment we've been talking about... is largely complete now". While inflation is weak, the BoC governor highlighted that forward looking indicators for inflation should be monitored instead of current inflation. These variables are pointing to stronger growth, and are in line with the bank's expectations of a closing output gap in the first half of 2018. While this may be true, a strengthening CAD will remain a risk for inflation. Report Links: Who Hikes Next? - June 30, 2017 Exploring Risks To Our DXY View - May 26, 2017 Bloody Potomac - May 19, 2017 Swiss Franc Chart II-15CHF Technicals 1 Chart II-16CHF Technicals 2 Recent data in Switzerland has been mixed: Although real retail sales yearly growth came in negative at -0.3%, it outperformed expectations and was better than last month. Additionally, the SVME PMI came also blew away expectations, increasing from last month's 55.6 reading to 60.1. However Consumer price inflation came in at -0.1%, underperforming expectations. The Swiss economy continues to be haunted by the ghost of deflation. Nonetheless, some economic indicators appear to be ticking up, most likely as a result of the sharp rally in EUR/CHF. We continue to believe that a rally of EUR/CHF beyond 1.1 is unlikely, as most of the good news in the euro area are already priced into the euro. Furthermore, any disappointments, particularly in EM could trigger a selloff in this cross. Report Links: Who Hikes Next? - June 30, 2017 Updating Our Intermediate Timing Models - April 28, 2017 The Fed And The Dollar: A Gordian Knot - April 14, 2017 Norwegian Krone Chart II-17NOK Technicals 1 Chart II-18NOK Technicals 2 The Labor Force survey, which measures the number of unemployed people as a percentage of the total civilian labor force came in at 4.6%, increased since last month. This measure shows that despite the increase in oil prices the Norwegian labour market continues to be tepid. The Norges Bank agrees with our assessment, as it lowered its projected near term policy rate path. Furthermore, they projected that rates in Norway will not rise until the beginning of 2019. The reasons for this are two fold: first, inflation should continue to remain weak, as the pass through from the collapse in the currency has faded. Additionally, bubbly real estate prices, which were the only factor, which could incite the Norges Bank to become more hawkish, have gone down, following reform in lending standards. Thus, despite its good value, the NOK will continue to underperform amongst commodity currencies. Report Links: Who Hikes Next? - June 30, 2017 A Market Update: June 23, 2017 Exploring Risks To Our DXY View - May 26, 2017 Swedish Krona Chart II-19SEK Technicals 1 Chart II-20SEK Technicals 2 It is true that the Riksbank recently hinted towards a more neutral stance, acknowledging "that inflation has recently been slightly higher than expected", which has made it "less likely than before that the Riksbank will cut the repo rate in the near term". However, the Riksbank also highlighted the fact that the bank is "prepared to implement further monetary policy easing if necessary to stabilize inflation". A very nuanced statement referred to the exchange rate, which "is important that [it] does not appreciate too rapidly", further stating that "this could happen if, for example, the Riksbank's monetary policy deviates clearly from that of other countries." This conclusively highlights that the bank is wary of diverging rates lifting undesirably on the krona, which is a limiting factor for substantial krona strength in the near term. However, the change of guard at the helm of this central bank in early 2018 could change all this caution. Report Links: Who Hikes Next? - June 30, 2017 Bloody Potomac - May 19, 2017 Updating Our Intermediate Timing Models - April 28, 2017 Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Closed Trades

Highlights While the yield curve is a critical indicator for developed economies, its significance in China should be put in proper perspective, as the country's market-based financial intermediation is much less important compared with the West. The inverted Chinese yield curve indicates tighter interbank liquidity in recent months, but the impact on the economy should be limited. The PBoC will at minimum pause its liquidity tightening campaign, which will provide a window for bonds to rally. Go long Chinese onshore corporate bonds. The near term impact of MSCI's A Share inclusion should be negligible for the broader market. Valuation indicators of the select 222 large-cap names are much more attractive compared with their domestic peers, which may well provide a catalyst for some catch-up rally. Feature Chart 1China's Inverted Yield Curve The Chinese authorities' tightening measures on the financial sector have significantly pushed up interest rates across the curve, particularly in the short end, leading to rapid yield-curve flattening. By some measures, long-dated interest rates are currently lower than short rates, generating an inverted yield curve (Chart 1). Some have viewed an inverted Chinese yield curve as a harbinger of an impending material growth slowdown. While the yield curve is undoubtedly a critical indicator for developed economies, its significance in China should be put in proper perspective. In short, bank loans still play a dominant role in financial intermediation, the interest rates on which are still largely determined by the policy lending rate. Therefore, a simple comparison of the Chinese yield curve to its counterparts in the West misreads the situation and is overly alarmist. Moreover, we suspect that the phase of maximum strength of policy tightening is over, at least in the near term. Therefore, Chinese interest rates are likely to fall in the coming three to six months. This week we recommend a long position in Chinese onshore corporate bonds. Why The Yield Curve Matters Less For China To be sure, the yield curve is among the most relevant and watched indicators in some developed economies. In the U.S., for example, an inverted yield curve, defined as U.S. 10-year Treasury yields resting below three-month Treasury yields, has historically been a reliable indicator in predicting economic recessions (Chart 2). Evidence from other developed economies such as Japan and Europe is less compelling, but a flat/inverted yield curve is still generally regarded as a market signal for growth problems. Chart 2U.S. Yield Curve Inversion Predicts Economic Recession The reasons for the linkage between yield curve inversion and economic recessions have been the subject of lengthy debates among academia, policymakers and investors. From a financial market perspective, it is generally accepted that an inverted yield curve occurs when the bond market anticipates a significant slowdown in growth and/or decline in inflation, which bids down long-term yields, while policymakers fail to respond in a timely manner, which holds short-term rates at elevated levels. Yield curve inversion is typically followed by aggressive monetary easing as central banks wake up to the economic reality predicted by the bond market. Economically, the costs of funding in most developed countries are tightly linked with interest rates in the bond markets. One of banks' key functions as financial intermediaries is to transform maturity - i.e. to "borrow short and lend long," and therefore interest rates of bank loans are tied to government bond yields at the longer end, while their costs of funding are linked to the shorter end. Therefore, an inverted yield curve typically compresses banks' interest margins, which tends to hinder credit origination and slow down business activity. For example, Chart 3 shows that U.S. mortgage interest rates historically have been tightly linked with 10-year Treasury yields, while interest rates of banks' deposit base and interbank rates for "wholesale" funding are both determined by short-term Treasury yields, which is in turn determined by the fed funds rate. In China, the yield curve plays a much smaller role than in the developed world, simply because the country's market-based financial intermediation is much less important. Traditionally both lending rates and deposit rates of commercial banks were rigidly set by the People's Bank of China, and there was little lending/borrowing activity outside the formal commercial banking system. The situation has been gradually changing in recent years as a result of financial reforms. Banks are given flexibility to set their own interest rates, and non-bank lending, or shadow banking activity that is more driven by market interest rates, has expanded. However, commercial banks still play a dominant role. Chart 3U.S. Bank Loan Rates Follow Treasury Yields Closely Chart 4China: Bank Loans Still Dominate Bank loans currently account for over 70% of China's total non-equity social financing, both in terms of flow and total outstanding stock (Chart 4). Commercial banks' average lending rate still closely tracks the PBoC policy benchmark. Banks' prime lending rate moves in lock step with PBoC interest rate adjustments, and average interest rates on new mortgages are also primarily determined by the policy rate (Chart 5). Banks' cost of funding is also primarily determined by retail deposit interest rates, which are in turn set by the PBoC. Retail deposits account for about 80% of total loanable funds for large banks, or 70% for smaller banks (Chart 6). Repo and interbank transactions, which are subject to the central bank's liquidity tightening, only account for 14% of smaller lenders' source of funds, or a mere 2% for large lenders. Chart 5Chinese Bank Loan Rates ##br##Still Track PBoC Benchmarks Chart 6Retail Deposits Are Still The Dominant Funding Source ##br##For Commercial Banks The important point is that market signals from China's juvenile and volatile financial markets should be taken with a healthy dose of skepticism, and a simple comparison with the West is often misleading. For example, a significant decline in stock prices in developed economies may well herald a growth recession in their respective economies. In China, however, domestic stock prices have routinely gone through massive boom and bust cycles without any tangible impact on the broader economy, as the equity markets play a marginal role for both the corporate sector in terms of raising capital and for households in managing their wealth. In recent years, China's financial sector reforms have been gradually introducing market forces in setting interest rates, but the process is far from advanced enough to have a meaningful and direct impact on the cost of funding for both the corporate sector and banks. Overall, the inverted Chinese yield curve indicates tighter interbank liquidity in recent months, but the impact on the economy should be limited. PBoC Tightening: Passing The Phase Of Maximum Strength Moreover, it is noteworthy that yield-curve flattening has been a global phenomenon rather than a China-specific development (Chart 7). What's different is that in other countries the flatter yield curve has been mostly due to falling yields of longer-dated bonds, while in China it has been entirely driven by a sharp increase in short-term yields due to the PBoC's liquidity tightening.1 Looking forward, the PBoC will maintain close scrutiny on the financial sector to keep financial excesses in check. However, we believe the phase of maximum strength of liquidity tightening is likely over, at least in the near term. There is no case for genuine monetary tightening, as inflation is extremely low and growth momentum is already softening. It is very unlikely that the PBoC will tighten monetary conditions further, amplifying deflationary pressures in the process.2 The PBoC's tightening measures have already significantly reduced the pace of leverage buildup and excesses in the financial system. Banks' exposure to non-bank financial institutions has tumbled, net issuance of commercial banks' negotiable certificates of deposits has turned negative of late, and overall off-balance-sheet lending by financial institutions, or shadow banking activity, has slowed sharply in recent months (Chart 8). In other words, the tightening campaign has achieved the intended consequences, diminishing the odds of further escalation. Chart 7Synchronized Yield Curve Flattening Chart 8Financial Excesses Are Being Reined In Global developments are also conducive for some loosening by the PBoC. Last week's rate hike by the Federal Reserve has further pushed down both U.S. interest rates and the dollar. The spread between Chinese 10-year government bond yields and U.S. Treasurys has widened sharply of late, which is helping stabilize the RMB (Chart 9). All of this has reduced pressure on the PBoC to follow the Fed with additional domestic tightening. Already, the PBoC has stepped in to ease liquidity pressure in the interbank system in recent weeks. After massive liquidity withdrawals early this year, the PBoC has been injecting liquidity into the interbank market through various open market operations in the past two months, according to our calculations - likely a key reason why interbank rates have stopped rising of late (Chart 10). Chart 9China - U.S. Interest Rate Spread Versus##br## Exchange Rate Chart 10The PBoC Is Stepping In ##br##To Ease Interbank Liquidity Pressure Chart 11Onshore Corporate Bonds ##br##Are Attractive Chinese corporate bonds will benefit the most, should the authorities stop further tightening (Chart 11). Onshore corporate spreads have widened sharply since late last year amid the PBoC crackdown, and are now substantially higher than in other countries. Chinese corporate spreads should recover without further escalation in liquidity tightening, and will also benefit from the ongoing profit recovery in the corporate sector. We expect both quality spreads and government bond yields to drop in the next three to six months, lifting corporate bond prices. Bottom Line: The PBoC will at minimum pause its liquidity tightening campaign, which will provide a window for bonds to rally. Go long Chinese onshore corporate bonds. A Word On The MSCI A-Share Inclusion MSCI Inc. announced this week its decision to include Chinese A shares in its widely followed emerging market and world equities indexes. The company will add 222 China A large-cap stocks to its EM benchmark at a 5% partial inclusion factor, which will account for about 0.73% of EM market cap. This marks a major milestone in China's capital market development and financial sector liberalization. Increasing participation of foreign institutional investors will also over the long run help improve China's corporate governance and regulatory practices - all of which are instrumental for improving the efficiency of domestic capital market as well as the efficiency of capital allocation. Table 1Valuation Of China A-Share Universe The near-term market impact, however, should be negligible. After all, the inclusion will take effect June next year. In addition, foreign investors already have access to these A share companies through the existing Stock Connect channels between Chinese domestic exchanges and Hong Kong. Moreover, potential capital inflows from global managed assets benchmarked to MSCI indexes in the initial step will be marginal. It is estimated that a total of US$18 billion, or RMB 125 billion, foreign capital may follow the MSCI decision into the A share market, a tiny fraction of A-shares' almost RMB 40 trillion market cap. That said, the valuation indicators of the select 222 large-cap names look attractive compared with their domestic peers, with median trailing P/E and P/B ratios at 23 and 2 times, substantially lower than other major domestic indexes (Table 1). MSCI inclusion may well provide a catalyst for some catch-up rally. We will follow up on this issue in the following weeks. Yan Wang, Senior Vice President China Investment Strategy yanw@bcaresearch.com 1 Please see China Investment Strategy Weekly Report, "China: Financial Crackdown And Market Implications," dated May 18, 2017, available at cis.bcaresearch.com. 2 Please see China Investment Strategy Weekly Reports, "A Chinese Slowdown: How Much Downside?," dated June 8, 2017, and "Chinese Growth: Testing Time Ahead," dated April 6, 2017, available at cis.bcaresearch.com. Cyclical Investment Stance Equity Sector Recommendations

Highlights The FOMC statement reaffirmed that the Fed remains in hiking mode. If the Fed keeps raising rates in line with the "dots," monetary policy will move into restrictive territory by early 2019. By then, the unemployment rate will have fallen to a level where it has nowhere to go but up. Unfortunately, history suggests that once unemployment starts rising, it keeps rising. The good news is that today's economic imbalances are not as formidable as those that existed in the lead-up to the past few recessions. The bad news is that cracks are starting to form. We are especially worried about the health of the U.S. commercial real estate sector. Remain overweight global equities for now, but look to significantly pare back exposure next summer. Feature The U.S. Expansion Is Getting Long In The Tooth Chart 1How Low Can It Go? The current U.S. expansion has now reached eight years, making it the third longest in the post-war era. History teaches that expansions do not die of old age. Rather, they are usually murdered by some combination of Fed tightening and the unwinding of the imbalances that were built up during the boom years. Thinking about the present, there is good and bad news on both fronts. Let's start with the Fed. This week's FOMC statement reaffirmed that the Fed remains in hiking mode. The good news is that real rates are still very low by historic standards, suggesting that the economy is unlikely to stall out this year. The bad news is that the Fed has less scope to raise rates than in the past. Chart 1 shows estimates of the real neutral rate developed by Fed researchers Thomas Laubach and Kathryn Holston, along with John Williams, President of the San Francisco Fed and Janet Yellen's close confidante. Their calculations suggest that the real neutral rate has plummeted over the past decade in the U.S. and the euro area, with lesser declines recorded in Canada and the U.K. In the U.S., the real neutral rate currently stands at 0.4%. Assuming the Fed raises interest rates in line with the "dots," rates will move into restrictive territory in early 2019. Given that monetary policy affects the real economy with a lag of 12-to-18 months, the Fed may not realize that it has raised rates too much until it is too late. The Downside Of A Low Unemployment Rate One might argue that this justifies a "go-slow" approach to tightening monetary policy. There is certainly validity to this view, but it is not without its drawbacks. The unemployment rate has now fallen to 4.3%, 0.4 points below the Fed's estimate of NAIRU. As Chart 2 illustrates, the odds of a recession rise when the unemployment rate reaches such low levels. Some commentators have argued that the headline unemployment rate understates the amount of economic slack. We are skeptical that this is the case. Table 1 compares a wide variety of measures of labor market slack with where they stood at the height of the business cycle in 2000 and 2007. The main message of the table is that the unemployment rate today is broadly where one would expect it to be based on these collaborating indicators. Taken together, these indicators suggest that slack is comparable to what it was in 2007, albeit still above the levels seen in 2000. Table 1Comparing Current Labor Market Slack With Past Cycles As we noted last week, the easing in U.S. financial conditions over the past six months is likely to boost growth in the second half of this year (Chart 3). If growth does accelerate, the unemployment rate - which is already 0.2 points below where the Fed thought it would be at the end of this year when it made its December 2016 projections - will fall below 4%. There is a high probability that this will fuel inflation, reversing the largely technically-driven decline in most core inflation measures over the past few months. Chart 3U.S.: Easy Financial Conditions Will Support Growth In H2 2017 The market is not pricing this in at all. In fact, 2-year breakeven inflation rates have tumbled by 87 basis points since March. A bit more inflation would be a welcome development. Not only have market-based projections of inflation fallen since the Great Recession, but long-term survey-based measures have dipped as well (Chart 4). Of course, one can have too much of a good thing. The experience of the 1960s is illustrative in that regard. Chart 5 shows that much like today, inflation in the first half of that decade was well anchored at just below 2%. However, once the unemployment rate fell below 4%, inflation soared. Core inflation rose from 1.5% in early 1966 to nearly 4% in early 1967, ultimately making its way to 6% by 1970. Chart 4Inflation Could Use A Boost Chart 5Inflation In The 1960s Took Off ##br##Once The Unemployment Rate Fell Below 4% If the Fed today wants to avoid the same fate, it will have to take steps to lift the unemployment rate back up to NAIRU. Unfortunately, history suggests that it is difficult to raise the unemployment rate a little bit without inadvertently raising it by a lot. Once unemployment starts to rise, a vicious circle tends to erupt where increasing joblessness leads to slower income growth, falling confidence, and ultimately, less spending and higher unemployment. In fact, there has never been a case in the post-war era where the three-month moving average of the unemployment rate has risen by more than one-third of a percentage point without a recession ensuing (Chart 6). Chart 6Even A Small Uptick In The Unemployment Rate Is Bad News For The Business Cycle Imbalances Are Growing The vicious circle described above tends to be amplified when there are large imbalances in the economy. The good news is that today's imbalances are not as formidable as those that existed in the lead-up to the past few recessions. The bad news is that cracks are starting to form. The ratio of household debt-to-disposable income is still close to post-recession lows, but this is largely because mortgage debt continues to be weighed down by a depressed homeownership rate (Chart 7). In contrast, consumer credit is rebounding: Student debt is going through the roof and auto loans are nearly back to pre-recession levels as a share of disposable income (Chart 8). Not surprisingly, this is starting to translate into higher default rates (Chart 9). The fact that this is happening at a time when the unemployment rate is at the lowest level in 16 years is a cause for concern. Chart 7Low Homeownership Rate Keeping A Lid On Mortgage Debt Chart 8Consumer Credit: Making A Comeback... Chart 9...With Defaults Starting To Rise In Some Categories Meanwhile, the ratio of corporate debt-to-GDP has risen above 2000 levels and is closing in on its 2007 peak (Chart 10). Contrary to the widespread notion that "wages aren't rising," real wages are increasing more quickly than corporate productivity (Chart 11). As the labor market continues to tighten, corporate profitability could suffer, setting the stage for rising defaults and increasing layoffs. Chart 10U.S. Corporate Sector Has Been Feasting On Credit Chart 11Real Wages Now Increasing Faster Than Productivity Worries About Commercial Real Estate We are particularly worried about the health of the commercial real estate (CRE) market. CRE prices currently stand 7% above pre-recession levels in real terms, having risen by a staggering 82% since the start of 2010 (Chart 12). Financial institutions hold $3.8 trillion in CRE loans, $2 trillion of which are held by banks. As a share of GDP, the outstanding stock of CRE bank loans in most categories is near pre-recession levels (Chart 13). Chart 12Commercial Real Estate Prices Have ##br##Surpassed Pre-Recession Levels Chart 13CRE Debt Is Rising Going forward, the fundamental underpinnings for the CRE market are likely to soften. The retail sector is already under intense pressure due to the shift in buying habits towards eCommerce. CMBX spreads in this space are rising. Vacancy rates in the apartment sector have started to tick higher and rent growth has slowed (Chart 14 and Chart 15). The number of apartment units under construction stands at a four-decade high according to Census data, despite a structurally subdued pace of household formation (Chart 16). Most of these units are likely to hit the market in 2018, which will result in a further increase in vacancy rates. Chart 14Vacancy Rates Are Bottoming Outside The Industrial Sector... Chart 15...While Rent Growth Is Losing Steam Chart 16Apartment Supply Is Surging, But Will There Be Enough Demand? There are fewer signs of overbuilding in the office sector. Nevertheless, vacancy rates are likely to rise, given the recent increase in the number of new projects in the pipeline. On the flipside, demand growth for new office space is set to weaken, as a tighter labor market leads to slower payroll gains. The Fed estimates that the U.S. needs to add only 80,000 workers to payrolls every month to keep up with a growing labor force, down from about 150,000 in the two decades preceding the Great Recession.1 The secular shift towards increased office density and teleworking will only further depress office demand over time. Chart 17Tighter Lending Standards Could Lead To Lower CRE Prices The one bright spot is industrial real estate. Thanks to a revival in U.S. manufacturing, vacancy rates remain low and rent growth is rising. However, if the U.S. economy does accelerate over the remainder of the year, the dollar is likely to strengthen, putting a dent in the profitability of U.S. manufacturing companies. Standing back, how worried should investors be about the CRE sector? For now, there is limited cause for concern. U.S. financial institutions have been tightening lending standards on CRE loans for seven straight quarters. Consequently, the average loan-to-value ratio for newly securitized loans has fallen about four points to 60% since 2015, and is now down eight points compared to 2007. However, if vacancy rates keep rising, real estate prices will fall, leading to a decline in the value of the collateral backing CRE loans. This could prompt lenders to pull back credit, causing prices to fall further (Chart 17). Seasoned real estate investors are no strangers to such vicious cycles, and if the next one begins at a time when growth is slowing because the economy is running out of spare capacity and financial conditions are tightening, it could easily trigger a recession. Fiscal Policy To The Rescue? Could looser fiscal policy delay the day of reckoning? The answer is yes, but much will depend on when the stimulus arrives and what form it takes. The best-case scenario is that fiscal policy is eased just as the economy is beginning to slow of its own accord. A burst of stimulus that arrives on the scene too early would be less desirable, although not necessarily counterproductive, since it would allow the Fed to step up the pace of rate hikes, thereby giving it more scope to cut rates later in response to slower growth. In practice, however, calibrating the amount of monetary tightening that is necessary to offset a given amount of fiscal loosening is difficult to achieve. This is especially the case in today's environment where another fight over the debt ceiling looms large, a new health care bill is making its way through the Senate, and Trump's tax agenda remains heavy on promises but short on specifics. Our expectation is that Congress will pass a "balanced" budget which equates revenues with expenditures over the 10-year budget horizon. How this affects growth is hard to predict with any certainty. On the one hand, spending cuts tend to depress aggregate demand more than tax cuts raise demand. In economic parlance, the fiscal multiplier for government spending is larger than for taxes. On the other hand, the tax cuts are likely to be front-loaded, while the spending cuts will be back-dated. If history is any guide, this means that the latter will never see the light of day. In addition, some of the budgetary impact from cutting statutory tax rates will be paid for through dynamic scoring, the questionable practice of assuming that lower personal and corporate tax rates will significantly spur growth. On balance, we expect fiscal policy to turn modestly stimulative over the next few years. However, given the uncertainty involved, there is a risk that the Fed either raises rates too much - thereby choking off growth - or by not enough, causing the unemployment rate to fall to a level where it has nowhere to go but up. Both outcomes could trigger a recession. Investment Conclusions Right now, our recession timing model, as well as the models maintained by various regional Fed banks, assign a low probability of a severe slowdown in the coming months (See Box 1 for details). These models, however, tend to send reliable signals only over a fairly short horizon. Looking further ahead, we see a heightened probability of weaker growth in the second half of 2018, which could set the stage for a recession in 2019. The good news is that today's economic imbalances are not as daunting as they were in the late innings of many past economic expansions. Thus, the 2019 recession is not likely to be especially severe. The bad news is that valuations across most markets are quite stretched. Thus, like the 2001 recession, the financial market impact could be disproportionally large compared to the economic impact. We are still overweight global equities, but will be looking to significantly reduce exposure by next summer. Once the equity bear market begins - most likely late next year - a 20%-to-30% retracement in U.S. stocks is probable. Given that correlations across stock markets tend to rise when risk sentiment is deteriorating, it is likely that other global bourses will also suffer if U.S. stocks weaken. Indeed, considering that most stock markets have a beta to the S&P 500 that exceeds one, other regions could suffer even more than the U.S. As the U.S. economy falls into recession, the Fed will stop raising rates. This will cause the dollar to weaken, although not before it has appreciated by about 10% in trade-weighted terms from current levels. Thus, while we remain bullish on the dollar over the next 12 months, we are much less sanguine about the greenback over the long haul. As the dollar weakens, the yen and euro will strengthen, imparting deflationary pressures on those economies. If our timing for the next recession proves correct, neither the ECB nor the BoJ will hike rates for the remainder of the decade. The Bank of England is a tougher call. The neutral rate of interest is higher in the U.K. than in continental Europe. Last week's election results represented a clear rejection of fiscal austerity. A more expansionary fiscal stance would give the BoE some scope to raise rates. A weaker pound has also given the economy a much needed competitive boost. With inflation picking up, it is not surprising that the BoE struck a more hawkish tone this week. Nevertheless, Brexit negotiations are liable to drag on for some time, which will constrain the ability of the BoE to tighten monetary policy. Stay long GBP/EUR and GBP/JPY over the next 12 months, but remain short GBP/USD. Housekeeping Note: Closing Our Tactical S&P 500 Short Hedge As noted above, we remain cyclically overweight global equities over a 12-month horizon. However, on occasion, we have put on a tactical hedge whenever equities appeared to be technically overbought. Such a situation arose six weeks ago. While the stock market did dip briefly shortly after we initiated the trade, it subsequently rallied back. At the time of initiation, we indicated that the trade would have a lifespan of six weeks. The clock has now run out, and we are closing the trade for a loss of 2%. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 1 Please see Rhys Bidder, Tim Mahedy, and Rob Valletta, "Trend Job Growth: Where's Normal?" FRBSF Economic Letter, 2016-32, Federal Reserve Bank Of San Francisco (October 24,2016), and Daniel Aaronson, "Estimating The Trend In Employment Growth," Chicago Fed Letter, No. 312, Federal Reserve Bank Of Chicago (July 2013). BOX 1 The Message From Our Recession Timing Model Chart Box 18Near-Term Recession Risk Remains Low Our recession timing model is based on eight variables: The Conference Board's Leading Economic Indicator, the Coincident Economic Indicator, the fed funds rate, inflation expectations, the unemployment rate, oil prices, credit spreads, and the yield curve. We use a logistic regression framework to model the probability of a recession. Currently, our model shows that the odds of a recession are low (Chart Box 18, panel 1). Only one of the components, namely, a rising fed funds rate, is signaling a risk of a recession. The various models developed by regional Federal Reserve banks also show very low near-term odds of a recession (panels 2 and 3). Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights The latest reading from all the indicators confirms that the growth improvement in the manufacturing sector since early last year has moderated, while the sharp recovery in producer prices has stalled. However, it is premature to be overly alarmed by a pending Chinese growth relapse. Betting on a material Chinese slowdown solely based on some sort of credit "impulse" estimate misses the big picture. The dramatic decline in real interest rates rather than an increase in new lending is what played a pivotal role in Chinese reflation since last year. Strategically we lean against being overly bearish. The Chinese economy will likely continue to moderate, but the downside risk appears low at the moment and overall business activity will remain buoyant. Feature Investors have become less sanguine on China's growth outlook in recent weeks, as the latest macro numbers are no longer unanimously positive. Concerns about a significant relapse in the Chinese economy are re-emerging, and the authorities' recent policy tightening has further heightened investors' anxiety levels. Judging from our recent conversations with clients, "China risk" is now clearly back on the radar. China's growth recovery since early last year played a major role in boosting some global risk assets such as commodities prices and emerging market equities. By the same token, will a China slowdown end the global reflation trade? The Divergence In Manufacturing PMIs Chart 1The Divergences In PMIs Investors' anxiety over China's cyclical trend has been amplified by the recent divergence between the official manufacturing Purchasing Manager Index (PMI) and the one compiled by Caixin Media group, a private source. The official survey for May still showed expansion, while the private PMI dropped slightly below the critical 50 threshold (Chart 1, top panel). Historically such divergences are not uncommon, and the private PMI appears to show sharper swings than its official counterpart, probably due to its smaller sample size and its focus on smaller private firms. Meanwhile, there were some commonalities: the sub-indices of output and new orders for both surveys remained above the expansionary threshold, while input costs and output prices for both dropped into contractionary territory. Taken together, the latest reading from all the indicators confirms that the growth improvement in the manufacturing sector since early last year has moderated, while the sharp recovery in producer prices has stalled - consistent with other recent macro variables. Meanwhile, the service industry is still showing solid expansion, according to both surveys, underpinning overall business activity (Chart 1, bottom panel). In short, it is premature to be overly alarmed by a pending Chinese growth relapse. Credit "Price" Versus "Volume": What Matters More? A common narrative to describe the reason behind China's ongoing growth moderation is policy tightening on both the monetary and fiscal fronts. As the argument goes, last year's growth recovery was driven by a massive increase in credit and fiscal spending, which has since been scaled back. As this credit and fiscal "impulse" fades away, the Chinese economy will tumble, sending shockwaves across the world. In our view, betting on a material Chinese slowdown solely based on some sort of credit "impulse" estimate misses the big picture, and is dangerously misguided. At BCA, we have long paid close attention to credit cycles and their impact on the growth outlook. However, there is no evidence that China's growth recovery since early last year was due to a massive increase in credit expansion and fiscal spending. In fact, total new credit provided by commercial banks and the "shadow banking sector" has been largely stable in recent years, and last year's credit "impulse," measured as the annual change in credit flows, was fairly modest - especially compared with previous bouts of sharp spikes (Chart 2). Similarly, Chinese fiscal spending actually decelerated sharply throughout last year, and dropped by over 10% in December, compared with a year earlier. Even if last year's fiscal retrenchment impacts the economy with a time lag, it is important to note that fiscal spending has already rebounded in recent months, which will become a tailwind for growth down the road. In our view, China's growth recovery since last year has a lot more to do with the "price" of credit rather than "volume." (Chart 3) Real interest rates dropped from double-digit levels that prevailed between 2012 and early 2016 to negative, thanks to a sharp increase in producer prices, while credit growth remained in a broad downtrend. In other words, the dramatic decline in real interest rates rather than an increase in new lending is what played a pivotal role in Chinese reflation. Chart 2Not Much 'Impulse' Chart 3Credit: 'Price' Matters More Than 'Volume' China's PPI has rolled over, which together with the authorities' attempts to tighten has begun to lift real interest rates. This will likely continue to generate some growth headwinds - a risk that clearly warrants close attention. However, monetary conditions currently are still very accommodative, and there is no reason to expect an overkill to choke off the economy. Why Growth Will Not Falter? Moreover, the bearish argument on China's cyclical outlook is fundamentally rooted in the assumption that the country's economy is dangerously imbalanced1 - a shaky house of cards propped up by policy stimulus that will immediately fall down once the policy pump-priming stops. While the structural profile of the Chinese economy will remain a major global macro issue subject to heated debates going forward, the bearish argument underestimates the economy's resilience, and therefore exaggerates the downside risks. First, it is important to note that China's growth challenges in previous years were to a large extent due to excessively tight monetary conditions, a costly policy mistake that amplified deflationary pressures. Real interest rates were kept at double digits for 5 consecutive years between 2012 and early 2016 while other major central banks were all trying desperately to lower borrowing costs within their respective economies. Furthermore, the trade-weighted RMB appreciated by 20% between 2012 and 2015. In fact, the RMB was the only major currency that appreciated in trade-weighted terms during this period (Chart 4), essentially shouldered deflationary stress for the rest of the world. In addition, Chinese regulators tried hard to block credit flows in an ill-conceived attempt to de-lever - which only prolonged credit intermediation channels and pushed loan demand to even costlier "shadow" institutions.2 All of these factors inflicted dramatic deflationary pain on Chinese manufacturers. Indeed, that the Chinese economy did not implode under the double-whammy of weak global demand and draconian domestic policy tightening - and staged a quick turnaround when monetary conditions eased - underscores the surprising resilience of the Chinese corporate sector. Second, the growth recovery since early last year has significantly improved financial conditions within the corporate sector and eased its balance sheet stress. Overall, companies have increased earnings, reduced inventories and beefed up cash positions (Chart 5). The situation can certainly deteriorate, but the sector is also better prepared for deflationary shocks than in previous years. Chart 4The RMB Shift Chart 5Inventory Is Still Very Low Third, even if China's corporate sector, especially industrial enterprises, are indeed as fragile as some bearish analysts claim, Chinese households and the service sector have much healthier fundamentals and therefore are less vulnerable. Consumer confidence has improved significantly in recent months following the growth acceleration, which should further help household consumption. The service sector now accounts for 52% of Chinese GDP, 30% larger than manufacturing. Household consumption and the service sector will provide an important anchor for business activity and prevent a major relapse in economic growth, even if the industrial sector slows more than we currently expect. Finally, the global growth environment is also largely supportive for the Chinese economy. The European economy has been showing some remarkable strength of late, and U.S. growth is likely to pick up after the recent soft patch, as per our U.S. specialists - both of which should bode well for Chinese exports. It is worth noting the recent weaker macro numbers out of China have followed growth disappointments in the U.S. (Chart 6). In fact, the ebbs and flows of "growth surprises" in the world's two largest economies in recent years have been largely in sync, albeit with China experiencing more pronounced volatility. In addition, the risk of an immediate escalation of protectionist backlash between the U.S. and China has also been lowered following President Xi's state visit to the U.S. in April.3 Overall, the Chinese economy is unlikely to slow materially, if the U.S. economy does reasonably well. Chart 6U.S. And China: Synchronized 'Surprises' All in all, we expect the Chinese economy will likely continue to moderate, but the downside risk appears low at the moment. In a reported titled "Chinese Growth: Testing Time Ahead," dated April 6th, we warned that "growth figures coming out of China in the coming months may be viewed as less market friendly."4 Recent Chinese data and investor reactions confirm this judgment. Nonetheless, we maintain the view that the Chinese economy's growth improvement remains largely intact, which will reinforce the upturn in the global business cycle and support global risk assets. Strategically we lean against being overly bearish, and we remain cyclically positive on Chinese equities, particularly H shares. Yan Wang, Senior Vice President China Investment Strategy yanw@bcaresearch.com 1 Please see China Investment Strategy Special Reports, "The Great Debate: Does China Have Too Much Debt Or Too Much Savings?" dated March 23, 2017, and "More On The Chinese Debt Debate," dated April 20, 2017 available at cis.bcaresearch.com. 2 Please see China Investment Strategy Weekly Report, "China: Financial Crackdown And Market Implications," dated May 18, 2017 available at cis.bcaresearch.com. 3 Please see China Investment Strategy Special Reports, "Reflecting On The Trump-Xi Summit," dated April 13, 2017 available at cis.bcaresearch.com. 4 Please see China Investment Strategy Weekly Report, "Chinese Growth: Testing Time Ahead," dated April 6, 2017 available at cis.bcaresearch.com. Cyclical Investment Stance Equity Sector Recommendations

Highlights EM EPS has recovered, supporting the current rally. However, forward-looking indicators portend a reversal and potential renewed contraction in EM EPS. BCA's Emerging Markets Strategy team has a more pessimistic outlook than the BCA house view, which is upbeat on the prospects for China's capex growth and commodity prices. The ongoing liquidity tightening in China amid lingering credit excesses is bound to produce major negative growth surprises. The authorities will reverse the ongoing monetary tightening only if the pain on the ground becomes visible or the economic data deteriorates significantly. Financial markets will sell off considerably in advance. In Chile, take profits on the receiving 3-year swap rate trade; stay neutral on this bourse within an EM equity portfolio. Feature EM Profit Recovery: How Enduring? EM equities have not only advanced in absolute terms but have also outperformed developed market (DM) share prices considerably since early this year. This outperformance has been rationalized by a recovery in EM earnings per share (EPS). Indeed, EM EPS has revived briskly in recent months (Chart I-1A). Chart I-1AEM/China Profits Growth To Roll Over (I) Chart I-1BEM/China Profits Growth To Roll Over (II) For this rally to continue, EM EPS would need to continue to expand further. We do not expect this. On the contrary, our bet is that EM EPS growth will slow considerably later this year and most likely contract in early 2018. Our basis is that the growth (first derivative) and impulse (second derivative) of EM & Chinese narrow money (M1) has in the past led their respective profit cycle (Chart I-1A and Chart I-1B). If these relationships hold and EM EPS growth dwindles later this year, EM share prices should begin to sense it now, and start falling back very soon. Interestingly, EM EPS net revisions have failed to rise above the zero line despite the recent rebound in profits (Chart I-2, top panel). This is in contrast to DM EPS net revisions, which have surged well above zero (Chart I-2, middle panel). As a result, recent EM relative outperformance against their DM peers has occurred despite the drop in relative net EPS revisions (Chart I-2, bottom panel). This presages EM equity analysts are not revising upward their forward estimates for EM EPS, despite the ongoing rally in share prices. This is extremely puzzling (and rare) and may be a reflection of recent weakness in commodities prices - or the fact that expectations for EM EPS growth were already elevated. We do not place much emphasis on analysts' EPS revisions because the latter swing with stock prices - they have zero forecasting power for share prices. We highlight this fact simply to counter the common market narrative that EM corporate earnings growth expectations are improving, driving EM bourses higher. Bottom Line: EM EPS has recovered, supporting the current rally. However, forward-looking indicators portend a reversal and potential renewed contraction in EM EPS nine months ahead. Importantly, EM equity prices relative to DM shares are at a major technical juncture (Chart I-3). A decisive breakout would be a very bullish technical signal, whereas a failure to break out would be an important warning sign. We continue to bet on the latter. Chart I-2EPS Net Revisions: EM And DM Chart I-3Relative Equity Performance: EM Versus DM China's Credit Cycle And Commodities Redux Our overarching theme has been and remains that China is tightening liquidity amid a lingering credit bubble. This cannot end well for financial markets that are exposed China's growth. Here we revisit our rationale for a credit slowdown in China and its impact on EM. Chinese interest rates have risen dramatically since last November across the entire yield curve. The 3-month interbank rate and AA- on-shore corporate bond yields both have risen by about 200 basis points since November 1, 2016. Monetary policy works with a time lag, and higher interest rates warrant a slowdown in credit growth (Chart I-4). In turn, it takes only a deceleration in credit growth for the credit impulse - the second derivative of outstanding credit - to turn negative. The falling credit and fiscal impulse will consequently lead to a relapse in Chinese import volumes and EM EPS (Chart 5), in turn weighing on commodity prices and non-commodity producing countries like Korea and partially Taiwan. Mainland import volumes contracted mildly in the second half of 2015, as demonstrated in Chart I-5. De facto, from the perspective of the rest of the world, China was in mild recession in late 2015. Not surprisingly, global risk assets in general, and particularly those exposed to China, tumbled. Chart I-4China: Higher Rates Point To##br## Negative Credit Impulse Chart I-5China's Credit Impulse Heralds ##br##Slowdown In Its Imports We expect China import volumes to shrink again by the end of this year or early next. Some sort of replay of 2015 is a real possibility. The broad-based yet mild selloff in commodities since early this year (Chart I-6) amid weakness in the U.S. dollar exchange rate gives us confidence in our view. Chart I-6ABroad-Based Selloff In Commodities (I) Chart I-6BBroad-Based Selloff In Commodities (II) Our colleagues at BCA have attributed the selloff in commodities this year to deleveraging in China's shadow banking system, and to traders worldwide closing their long positions. They expect an improving commodities supply-demand balance to support prices going forward. It makes sense to us to explain the selloff in commodities as having been caused by deleveraging in China's shadow banking system. Yet to be consistent, we should also acknowledge that the rally in commodities last year was to a large extent driven by the same forces in reverse: non-commercial buyers (investors) buying commodities both in China and elsewhere. In short, this signifies there was little improvement in worldwide commodities demand last year. In 2016, rising commodities prices provided a significant boost to commodity-producing countries and underlying corporate profits - and ultimately EM risk assets. The drop in commodities prices this year, if sustained, should lead to the opposite dynamic: income/profits among commodities countries/companies will drop. As such, falling commodities prices amid diminishing investor demand for commodities is bearish for EM risk assets. Where we differ from the majority of our colleagues at BCA is that we expect Chinese credit growth to decelerate, thereby weighing on its capital spending and depressing demand for commodities (please refer to Chart I-5). We have written extensively1 on this topic and will not fully rehash our view that China's annual credit growth will decelerate from the current 12% to somewhere around 8% in the next 12-18 months. In short, China's corporate and household credit-to-GDP ratio cannot rise indefinitely from an already high level of 225% of GDP. Credit growth will likely downshift to a level of sustainable nominal GDP growth, which is probably around 8%. Our main disagreement with our colleagues on structural issues is as follows: we believe China's credit excesses are not a natural outcome of the nation's high savings rate but rather the outcome of a speculative credit boom driven by high-risk behavior among creditors and debtors.2 Tightening liquidity amid such speculative excesses creates a very bearish backdrop for risk assets exposed to China's credit cycle. The bullish camp on China has recently pointed to a strong recovery in mainland nominal GDP growth, which in their view suggests that double-digit credit growth in China is not excessive (Chart I-7). However, such a surge in nominal GDP growth has been due to the GDP deflator rising from zero in the fourth quarter of 2015 to 5% in the first quarter of this year. Importantly, the swings in the GDP deflator almost perfectly correlate with the fluctuation in commodities prices (Chart I-8). This proves how much China's economy is exposed to commodities cycles and how much of nominal GDP swings are stipulated by resource price swings. Chart I-7China: Credit And ##br##Nominal GDP Growth Chart I-8China's GDP Deflator Is Very Sensitive##br## To Commodities Prices As commodities prices decline, China's GDP deflator, producer prices and nominal GDP growth will all dwindle. Thereby, China's underlying steady state nominal GDP growth is probably around 8% at best (5.5-6% real growth), with inflation of 2-2.5% (assuming flat commodities prices). If this is indeed the case, corporate and household credit growth of 12% entails a further build-up of leverage and an escalating non-public credit-to-GDP ratio, which already stands at 225% of GDP: corporate debt is 180% and household debt is at 45% of GDP. Bank loans account for 70%, while shadow (non-bank) funding channels (corporate bonds, trust products, entrusted loans, and banker's acceptance) constitute 30% of outstanding non-public credit or 65% of GDP. Both are growing at an annual rate of 11-12.5% (Chart I-9). On the whole, the share of shadow banking is non-trivial and its current growth pace is unsustainable amid ongoing regulatory tightening and rising interest rates. Furthermore, banks are themselves exposed to shadow banking as their claims on non-depository financial institutions have risen exponentially from RMB 3 trillion to RMB 27 trillion over the past five years. In regard to non-standard credit assets,3 our estimates are that banks' off-balance-sheet exposure is RMB 10 trillion compared with RMB 18.3 trillion of their balance-sheet non-standard credit assets. The off-balance-sheet credit exposure to non-standard credit assets is much larger for medium and small banks than the largest five (Table I-1). We discussed these issues in greater detail in our June 15, 2016 Special Report titled "Chinese Banks' Ominous Shadow". Chart I-9Bank Loans And Non-Bank (Shadow) Credit Growth With banks being forced by regulators to bring off-balance-sheet assets onto their balance sheets, their capital adequacy ratios will drop and their ability to sustain double-digit credit growth will be curtailed. Chart I-10Stay With Short Small / Long Large ##br##Banks Equity Trade The risks to medium and small banks is greater than to the large five banks. That is why we reiterate our recommendation from October 26, 2016 to short small banks versus large ones (Chart I-10). As a final note, we are often asked whether the government will provide a bail out if things deteriorate. Yes, we concur that policymakers will step in and backstop a financial system to preclude a systemic crisis. However, they are tightening now, and like the rest of us have little visibility. The authorities will meaningfully reverse the ongoing monetary tightening only if the pain on ground becomes visible or economic data deteriorate considerably. Financial markets will sell off materially in advance. Bottom Line: Investors should not be long China-plays, commodities and EM risk assets when mainland policy tightening is occurring amid lingering speculative credit excesses. Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com Strategy For Chilean Markets We recommended receiving 3-year swap rates on November 2, 2016 and this position has panned out with rates dropping by 30 basis points. We now recommend booking profits. The following has led us to conclude that the risk-reward profile of this position is no longer attractive: The improvement in narrow money (M1) growth points in a bottom in the economic activity indicator (Chart II-1). Mining production plunged amid the strikes in the world's largest copper producer Codelco (Chart II-2, top panel) and manufacturing production has also been contracting (Chart II-2, bottom panel). A period of improvement in mining and manufacturing output from a very low base is likely. Chart II-1Book Profits On Receiving ##br##3-Year Swap Rate Position Chart II-2Chile: Money And Economic##br## Activity Are Bottoming Out This will ameliorate overall business conditions and cause the central bank, at least for the time being, to halt the easing cycle. The pace of expansion in employment, wage growth, and consumer credit remains decent (Chart II-3). This will put a floor under household spending growth for now. Odds are that copper prices will decline meaningfully in the next nine months or so, which will cause the Chilean peso to depreciate. Although a depreciating currency will not to lead to materially higher interest rates in Chile, it will limit downside in local rate expectations. Finally, local 3-year swap rates and their spread over U.S. 3-year bond yields are extremely low from a historical perspective (Chart II-4). At this point, there is little value left in Chilean local rates. Chart II-3Chile's Mining And Manufacturing ##br##A Period Of Stabilization Ahead Chart II-4Chile: Consumer Spending##br## Is Holding Up Investment Conclusions Chart II-5Chilean Local Rates Spreads Over ##br##U.S. Treasurys: Not Much Value Left We do not expect the central bank to hike but the downside in local rates is limited for the time being. Take profits on the receiving 3-year swap rate trade. As to equities, the outlook for relative performance is balanced; we continue recommending a benchmark weight in Chile for dedicated EM equity portfolios. For absolute return investors, the risk-reward profile is not attractive because our profit margin proxy points to a relapse in corporate earnings (Chart II-5). Unit labor costs are rising faster than the core inflation rate, producing a profit margin squeeze (Chart II-5, bottom panel). Finally, we continue shorting the peso versus the U.S. dollar as a bet on lower copper prices. 1 Please refer to the Emerging Markets Strategy Special Reports titled, "Do Credit Bubbles Originate From High National Savings?", dated January 18, 2017, Misconceptions About China's Credit Excesses", dated October 26, 2016 and "China's Money Creation Redux And The RMB", dated November 23, 2016, available at ems.bcaresearch.com 2 Please refer to the Emerging Markets Strategy Special Reports titled, "The Great Debate: Does China Have Too Much Debt Or Too Much Savings?", dated March 23, 2017, "Do Credit Bubbles Originate From High National Savings?", dated January 18, 2017, "Misconceptions About China's Credit Excesses", dated October 26, 2016 and "China's Money Creation Redux And The RMB", dated November 23, 2016, available at ems.bcaresearch.com 3 Non-standard credit assets are banks' claims on corporates that are not classified as loans. For more details please refer to the Emerging Markets Strategy Special Report titled, "Chinese Banks' Ominous Shadow", dated June 15, 2016, available at ems.bcaresearch.com Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights Reflation Trade: The backdrop for global growth and monetary liquidity remains positive, and suggests that risk assets will outperform government debt for the balance of 2017. However, there are some early signs of fading momentum which raises risks for financial markets in 2018. New Zealand: The more dovish tone taken by the RBNZ reflects the more uncertain outlook for New Zealand growth and inflation. Go long 5-year New Zealand government bonds versus 5-year U.S. Treasuries (currency-hedged) and also versus 5-year German government debt (currency-unhedged). South Korea: Large expected increases in fiscal spending from the new government in Seoul will drive up the longer end of the South Korean government bond curve, while the Bank of Korea's easing stance and weak domestic economy will anchor the short-end of the curve. Position for this by entering a 2-year/10-year steepening trade in the South Korean government bond market. Feature "I know it makes no difference to what you're going through; but I see the tip of the iceberg, and I worry about you." - Rush Is The Liquidity Party Starting To Wind Down? Global financial markets continue to enjoy the "sweet spot" of a solidly expanding global economy, but without enough inflation pressure to force central banks to slam on the monetary brakes. That backdrop is starting to change, though. Odds are rising that the European Central Bank (ECB) will begin tapering its bond buying next year, with some hints of that possibly being announced as soon as next week's monetary policy meeting. At the same time, the Bank of Japan (BoJ) - faced with the operational constraints of buying an ever-increasing share of Japanese financial assets - is focused on targeting long-term interest rates rather than increasing liquidity. Even the Federal Reserve is now talking about reducing its massive balance sheet later this year. The liquidity tailwind to global growth and risk assets is now at risk of becoming a headwind. Already, the growth rate of the major central bank balance sheets has rolled over and is on course to decelerate further over the next year (Chart of the Week). Importantly, this downshift in global liquidity momentum is happening as signs of slowing growth have appeared in some major economies like China and the U.S. (Chart 2). Chart of the WeekLiquidity Tailwind To Risk##BR##Assets Is Fading Chart 2Growth Momentum##BR##Already Starting To Cool Off We remain concerned that the Chinese economy will see a policy-induced deceleration in the 2nd half of the year. However, we still expect the U.S. to rebound after the soft patch of growth in the first quarter, and we see nothing in the Euro Area data to suggest that the current solid expansion is at risk of fading quickly. This should allow inflation expectations to drift upward toward the central bank targets given the apparent lack of spare capacity on both sides of the Atlantic (Chart 3). Chart 3Fed & ECB Facing##BR##Economic Capacity Constraints We still expect the Fed to deliver another two rate hikes before year-end and the ECB to begin its exit strategy from the current extraordinary monetary policies by slowing the pace of asset purchases starting early next year. For now, the backdrop will remain supportive for the outperformance of growth-sensitive assets like corporate credit and equities over government bonds in the U.S. and Europe over the balance of 2017. However, the early signals sent by "leading leading" indicators such as our Global Leading Economic Indicator diffusion index (Chart 2, top panel) suggests that liquidity and growth trends will become far more challenging for the markets in 2018. Bottom Line: The backdrop for global growth and monetary liquidity remains positive, and suggests that risk assets will outperform government debt for the balance of 2017. However, there are some early signs of fading momentum which raises risks for financial markets in 2018. Maintain a below-benchmark duration exposure and an overweight allocation to corporate debt in global fixed income portfolios. New Zealand: Safety From A Global Bond Apocalypse? A growing number of the world's most wealthiest (and, arguably, most paranoid) people are reportedly buying real estate in New Zealand as a safe haven place to live if modern civilization collapses.1 While the immediate need for taking such precautions can be debated, there is sound logic in treating New Zealand as a location far removed from the current geopolitical and socio-economic problems of the world. We now see a case for treating New Zealand bonds as a potential "safe haven" market for global fixed income investors. The Economic Backdrop Has Become More Muddled We have been running a SHORT position in New Zealand (paying 12-month OIS rates) in our Tactical Overlay portfolio since last November. Our view then was that the New Zealand economy would surprise to the upside in 2017 and inflation was likely to start drifting upward. This would pressure the Reserve Bank of New Zealand (RBNZ) to raise the Official Cash Rate (OCR) from the highly accommodative level of 1.75%. So far, that expectation has not panned out as the RBNZ has held rates steady amid a more uncertain outlook for the New Zealand economy. Growth indicators have been a bit mixed over the past few months, but the current uptick in the manufacturing purchasing managers' index (PMI) is pointing to real GDP expanding around 3% on a year-over-year basis (Chart 4). If maintained for the full year, this would be slightly above the RBNZ's estimate of potential growth at 2.8%. There are some downside risks, however, given that consumer and business confidence are both below previous cyclical peaks and fiscal policy is expected to be mildly restrictive in 2017 (bottom three panels). The housing market remains a key cyclical wild card. Residential construction has been a significant source of growth over the past few years, driven by a surge in net immigration into New Zealand and declining interest rates (Chart 5). However, the RBNZ is projecting immigration inflows to slow from the current high level, largely due to improving labor market conditions in the developed economies (most notably, Australia, which is the largest source of New Zealand immigrants). Chart 4Stable NZ Growth...For Now Chart 5NZ Housing Activity Starting To Peak Out Slower immigration would reduce the demand for New Zealand housing at a time when mortgage rates have already been rising off the record lows seen in 2016 (bottom panel). This has occurred without any rate hikes from the RBNZ, as rising global bond yields have put upward pressure on New Zealand bank funding costs, which have been passed through to higher mortgage rates. The RBNZ is currently projecting growth in house prices to slow sharply from last year's robust 15% pace to just 5% in 2017. The main drivers are higher borrowing costs and the ongoing impact of macro-prudential regulations against high loan-to-value ratio mortgage lending. Importantly, slower housing activity will not only have a direct impact on GDP growth through softer construction, but will also indirectly dampen consumer spending growth via wealth effects. Yet even with this expected drag on growth from housing, the New Zealand economy is still expected to face capacity constraints over the rest of the year. Higher Uncertainty Over Price Pressures Both the RBNZ and the International Monetary Fund estimate that the output gap has fully closed and is projected to move into positive territory this year (Chart 6). At the same time, the current unemployment rate of 4.9% is below the OECD's estimate of the full employment level and the RBNZ projects a further decline in joblessness in 2017 (third panel). Despite this evidence of the economy reaching capacity constraints, both wage growth and price inflation remain subdued and inflation expectations remain well-anchored around 2% - the midpoint of the RBNZ's 1-3% target range. Wage costs are particularly depressed, growing only 1% on a year-over-year basis in Q1. This may be related to the rise in the labor force participation rate - up to an all-time high of 70.6% in Q1 from a cyclical low of 68.2% at the end of 2015 - that has increased the available supply of labor. The most recent headline inflation print for Q1 was quite strong, taking the year-over-year growth rate up to 2.2%. Yet in the RBNZ's April Monetary Policy Statement (MPS), the central bank took a surprisingly dovish tone, citing uncertainty over the true degree of slack in the economy and downside risks to growth that would prevent a further acceleration of inflation.2 The RBNZ now forecasts inflation to not rise above 2.2% this year and to fall back to 1.1% in both 2018, led by a sharp decline in growth for tradeables, mostly energy and food inflation (Chart 7). Importantly, this forecast includes the recent decline in the trade-weighted New Zealand Dollar (NZD). Non-tradeables inflation is also expected to stabilize on the back of slower housing-related items in the consumer price index. Chart 6RBNZ Not Expecting A Big Rise In Inflation... Chart 7...As Growth In Tradeables Prices Cools A Weaker Case For Tighter Monetary Policy The official RBNZ projection is that the OCR will stay unchanged at 1.75% until September 2019. The market expectation priced into the NZD OIS curve calls for 27bps of hikes over the next twelve months (Chart 8). Our New Zealand Central Bank Monitor has been suggesting the need for tighter monetary policy since mid-2016, but appears to be rolling over (2nd panel). The diminished rate hike expectations have coincided with a decline in the NZD and a sharp underperformance of New Zealand equities. The markets are giving a consistent signal on softening growth prospects in New Zealand, confirming the central bank's more recent dovish turn. Chart 8Market Expectations Of##BR##RBNZ Hikes Are Fading Given the newfound uncertainties over the New Zealand growth and inflation outlook, the case for owning New Zealand interest rate exposure has grown a little bit stronger. Admittedly, we do not envision a major pullback in growth, and inflation may not fall by as much as the RBNZ is expecting given how little spare capacity there appears to be in the economy. Yet there is now just enough uncertainty to keep the central bank on hold for longer than expected, as was noted in the "scenario analysis" section of the April MPS.3 The RBNZ noted that if the level of spare capacity is smaller than currently assumed, then the latest growth forecast will result in inflation eventually moving to 2.0% in 2018 and 2.3% in 2019, resulting in the OCR needing to rise to 2.25% in two years. Alternatively, if housing demand slows even faster than current projections, inflation would be below the 2% target during the next two years and the OCR would need to fall to 1.25% by the end of 2018. Our takeaway from this is that, even in the more positive scenario, interest rates are not expected to rise by much more than the markets are currently discounting. Position For Tighter New Zealand Spreads Versus Treasuries & Bunds The economic risks in New Zealand now appear evenly balanced. This argues for stable monetary policy and diminished bond volatility. Current market forwards for both government bonds and NZD swaps shows that very little movement in interest rates is expected over the next year (Chart 9). We generally agree with this pricing, although the uncertainty over the degree of spare capacity, and underlying inflation pressures, make a directional view on interest rates or the shape of the yield curve an unattractive risk proposition. A more interesting opportunity presents itself in looking at spread trades between New Zealand government bonds versus other developed market sovereign debt. The yield betas for New Zealand versus the U.S. and Germany have fallen steadily over the past year (Chart 10), indicating that New Zealand bonds can be more insulated from the rise in yields that we expect for U.S. Treasuries and German Bunds over the latter half of 2017. Given the competitively high yields on offer in New Zealand, even on a currency-hedged basis (bottom panel), we see a case for going long New Zealand interest rate exposure versus U.S. and Germany. Chart 9Higher NZ Bond Yields##BR##Priced Into Forwards Chart 10NZ Bonds: Now Lower Beta##BR##With Higher Hedged Yields At current yield levels, going long New Zealand versus Germany looks more compelling relative to spread compression trades versus U.S. Treasuries. We see strong potential for New Zealand-Germany spreads to tighten faster than the forwards over the next six months (Chart 11), largely through rising German yields as the ECB signals that a tapering of bond purchases is set to begin next year. The downside potential for New Zealand-U.S. spread compression looks less likely from current tight levels, although if Treasury yields rise by as much as we expect in the coming months, some spread tightening should occur here, as well. Chart 11Go Long 5Yr NZ Bonds Vs##BR##USTs and German OBLs Based on our analysis, we are closing our current NZD rates trade in our Tactical Overlay portfolio with a tiny profit of +3bps , and entering two new trades: long 5-year NZD government bonds versus 5-year U.S. Treasuries, on a currency-hedged basis; and long 5yr NZD government bonds versus 5-year German government debt, on a currency-unhedged basis.4 We are choosing to hedge the currency exposure back into USD for the former given the view of BCA's currency strategists that the EUR/USD exchange rate is now stretched too far to the upside and is at risk of declining as the Fed delivers on additional rate hikes in the coming months.5 In other words, we see a greater potential for a decline in NZD/USD than NZD/EUR in the next 3-6 months. Bottom Line: The more dovish tone taken by the RBNZ reflects the more uncertain outlook for New Zealand growth and inflation, in contrast to the strong likelihood of additional Fed rate hikes and an ECB taper announcement in the next few months. Go long 5-year New Zealand government bonds versus 5-year U.S. Treasuries (currency-hedged) and also versus 5-year German government debt (currency-unhedged). South Korea: A Bad Moon Rising For Bond Yields Chart 12Markets Not Worried##BR##About The New President The new South Korean president, Moon Jae-In was elected on May 9th, ending a year of political turmoil after the previous president's scandal and impeachment. Our colleagues at BCA Geopolitical Strategy view Moon and his Democratic Party as a major shift to the political left.6 The new president's policy agenda is aimed at economic stimulus for the working class alongside reforms of the country's chaebol industrial giants. Korean financial markets have greeted the election result positively, with the benchmark KOSPI equity index up 2.7%, and the Korean won up 1% versus the U.S. dollar, from the pre-election levels on May 8th. (Chart 12). This is consistent with past market behavior, as the won tends to be less reactive toward domestic events (i.e. after the previous president's impeachment, the won actually strengthened) and more sensitive to international uncertainties (i.e. North Korea-U.S. military tensions, as occurred in mid-March). Korean interest rates, however, have shown little response to the change in leadership in Seoul, with bond yields unchanged since the election. We see this as presenting an opportunity for fixed income investors. Clearly, the new regime in Seoul represents a real change for the Korean people, but it also represents a potential shift in the economic backdrop - namely, through an expected large fiscal stimulus from the new government - that will impart a steepening bias to the Korean interest rate curve. A Sluggish Economy Greets The New President While the steady, if unspectacular, pace of global growth in the past few years has been enough to absorb spare capacity in many countries, South Korea's sub-par economic performance has left the country with a widening output gap (Chart 13). Policymakers are well aware that consumer spending, which contributes about 60% of GDP, has been steadily weakening alongside slowing credit growth. Chart 13Sluggish Growth In South Korea The new government will attempt to boost domestic consumption, and thus overall growth, by increasing social welfare spending. Moon's economic agenda calls for raising the minimum wage by 55% by 2020, increasing subsidies for education costs and parental leave, and doubling the basic pension payment for the elderly regardless of their income level. It might prove to be very effective in the short term at boosting consumer spending, but this may not prove to be a sustainable driver of growth in South Korea, where the marginal swings in the economy have historically been driven more by exports. Youth joblessness is another problem that Moon will attempt to tackle with his ambitious economic program. While the labor market may appear healthy, with an overall unemployment rate of only 3.7%, the situation is far more challenging for young adults in South Korea - the jobless rate for those aged 20-29 is 11.3%. One of the reasons for such a high unemployment rate among young South Koreans is that university graduates, of which there are many in this highly-educated nation, expect (and look for) high-paying jobs, but cannot find enough of them.7 The labor market has become more competitive in recent years as weak economic growth has limited the ability of private sector, especially large corporations, to hire as much. To solve this problem, the new government has promised to create 810,000 jobs in the public sector. Creating public sector jobs may temporarily solve the high unemployment rate, but in the long run, this will also cause larger fiscal burdens for taxpayers. Position For A Steeper South Korean Yield Curve Headline CPI inflation in South Korea is currently hovering around the 2% target of the Bank of Korea (BoK), while core CPI growth is lower at 1.3%. The BoK has maintain the policy rate at 1.25% since June 2016, with a bias towards additional easing given the lack of sustained inflationary pressure amid weak domestic demand. The BoK did sound a slightly more upbeat tone on the economy at last week's monetary policy meeting, led by the spillover effects from improving global growth rather than a more bullish expectation on the Korean consumer. Importantly, the central bank still expects inflation pressures to remain subdued - no surprise given the large output gap. The BoK did note that it is monitoring several factors in judging future policy decisions: the pace of rate hikes by the Fed, trends in global trade, geopolitical tensions, the pace of household debt accumulation and "the directions of the new government's fiscal policies." The latter may end up being the most important factor, as President Moon is proposing an increase in government spending equal to 0.7% of GDP - an amount equal to ½ of the estimated output gap coming after a 2016 budget surplus of 1% of GDP. This increase in fiscal spending could directly drive up the longer-end of Korean yield curve, as this would result in a narrower budget surpluses and greater KGB issuance. At the same time, the lack of domestic inflation pressures, even with the fiscal stimulus, will keep the BoK on an easing bias that will keep short dated yields well anchored. Therefore, we see the potential for the Korean yield curve to eventually steepen and break the downward-sloping trendline in place since 2014 (Chart 14). We recommend positioning for this move by entering a 2-year/10-year steepening trade in the Korean yield curve. Admittedly, this trade is more structural than tactical in nature, as the Moon stimulus policies will take time to unfold. Importantly, a flattening of the 2-year/10-year KGB curve is currently priced into the forwards, meaning that positioning now for a steepener does not incur negative carry (Chart 15). Chart 14More Fiscal Stimulus =##BR##Steeper Korea Curve Chart 15Enter A 2Yr/10Yr##BR##Korean Bond Curve Steepener Also, Korean 10-year bond yields are currently exhibiting a strong correlation to similar maturity U.S. Treasuries with a yield beta around 1.0 (bottom panel). Given our view that longer-dated U.S. yields have upside risk from both additional Fed rate increases and higher U.S. inflation expectations, that high yield beta suggests that the Korean yield curve could suffer some of the same cyclical bear-steepening pressures that we expect for U.S. Treasuries in the next 3-6 months. Bottom Line: Large expected increases in fiscal spending from the new government in Seoul will drive up the longer end curve of the South Korean government bond curve, while the Bank of Korea's easing stance and weak domestic economy will anchor the short-end of the curve. Position for this by entering a 2-year/10-year steepening trade in the South Korean bond curve. Robert Robis, Senior Vice President Global Fixed Income Strategy rrobis@bcaresearch.com Ray Park, Research Analyst ray@bcaresearch.com 1 https://www.theguardian.com/technology/2017/jan/29/silicon-valley-new-zealand-apocalypse-escape 2 The central bank noted that its "suite" of output gap estimates, using varying methodologies, have an unusually wide range at the moment between -1.5% and +2%. 3 http://www.rbnz.govt.nz/monetary-policy/monetary-policy-statement 4 These trades can be done using interest rate swaps as well (receiving NZD rates vs paying USD & EUR rates), as swap spreads are expected to remain broadly stable in all three regions. 5 Please see BCA Foreign Exchange Strategy Weekly Report, "Bloody Potomac", dated May 19 2017, available at fes.bcaresearch.com. 6 Please see BCA Geopolitical Strategy Weekly Report, "Northeast Asia: Moonshine, Militarism, And Markets" dated May 24 2017, available at gps.bcaresearch.com. 7 According to the OECD, Korea's college enrollment rate was a whopping 87% as recently as 2014. The GFIS Recommended Portfolio Vs. The Custom Benchmark Index Recommendations Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

Feature Chart 1 Senior officials at the Federal Reserve have begun preparing the market for the eventual run down of the central bank's balance sheet. After several rounds of quantitative easing (QE), total assets held by the Fed currently stand at US$4.5 trillion - a dramatic increase from US$900 billion before the global financial crisis. Indeed, efforts to shrink the Fed's balance sheet are essentially reverse QE. As the 2013 'Taper Tantrum" suggests, such a profound change in U.S. monetary policy can have a significant impact on interest rates and broader financial assets, and Fed officials are working hard to properly anchor market expectations. In comparison, how the People's Bank of China manages its balance sheet is much less transparent and less understood by market participants, even though the PBoC has the biggest balance sheet among the world's major central banks (Chart 1). Currently, the PBoC's total assets amount to US$4.9 trillion, compared with about US$4.5 trillion for both the Fed and the European Central Bank (ECB). Moreover, its balance sheet has stopped growing since 2015 in local currency terms and has been shrinking in dollar terms, but the impact on the economy and financial markets has so far not been material. Generally speaking, a central bank uses its balance sheet to aid monetary policy. It controls the size and composition of its assets to affect interest rates, and in turn the economy. Through "operation twist" and QE, the Fed significantly increased its holdings of longer-dated Treasury securities and mortgage backed securities (MBS), which currently account for 95% of its assets (Table 1). Therefore, shrinkage of the Fed balance sheet means that the Fed's holdings of long-term securities will gradually be reduced - likely by allowing them to run off at maturity rather than selling them in the open market. This should nonetheless put some upward pressure on long-term risk-free rates going forward. Table 1The Fed's Balance Sheet In a Special Report we published six years ago, we pointed out the explosion in the PBoC's balance sheet and its unique features compared with other central banks.1 In a nutshell, the PBoC's biggest holdings on its asset side were U.S. Treasurys rather than domestic risk-free assets. The Chinese central bank was essentially engaging in a massive "currency swap" in which it accumulated U.S. Treasurys while dramatically increasing the country's monetary base. Meanwhile, it was also working hard to "sterilize" by forcing commercial banks to maintain an increasingly massive sum of required reserves with the central bank. These policy tools, however, were inherently crude and clumsy, with huge volatility in monetary market rates and overall financial volatility being a key after-effect. This week we are revisiting the PBoC's balance sheet to highlight some major shifts in recent years. Some developments are worth highlighting. Dynamics have completely reversed since 2015, when Chinese official reserves began to fall, leading to a shrinking in the PBoC's balance sheet by about US$500 billion since the all-time peak. The "sterilization" process has also been reversed, as the PBoC has been releasing liquidity back into the domestic financial system. The overall liquidity situation has been largely stable. Normally a decline in the PBoC's foreign asset holdings would lead to a decline in the reserve requirement ratio (RRR) to offset the liquidity outflows, leading to a simultaneous decline in both sides of the central bank's balance sheet. The PBoC, however, has been resisting shrinking its balance sheet. As its foreign asset holdings (U.S. Treasurys) have been declining, the PBoC has significantly ramped up domestic asset holdings by increasing direct claims on commercial banks through repos and other lending facilities. The central bank appears to be concerned that a lowered RRR will stoke more domestic capital outflows, which risks creating a vicious circle. How the PBoC manages domestic liquidity has seen major shifts in recent history, and will likely continue to evolve going forward. The RRR, as a monetary policy tool, will likely be gradually phased out.2 Over the long run, this will lead to important changes in the PBoC's balance sheet and the way it conducts monetary policy. In the short term, commercial banks' excess reserves are at close to record low levels. The odds are rising that the RRR will be lowered in the coming months, especially if the RMB stabilizes against the dollar, as we expect.3 Finally, it is worth noting that the most aggressive phase of the Fed's QE efforts coincided with the most rapid phase of the PBoC's balance sheet expansion. This means that both central banks were aggressive buyers of U.S. Treasurys and risk-free assets in previous years. Looking forward, if a shrinking Fed balance sheet leads to a sharp increase in U.S. interest rates and a dollar rally, it could force the PBoC to also liquidate its holdings of U.S. Treasurys to stabilize the RMB exchange rate. This means both the Fed and the PBoC could become marginal sellers of Treasurys, which would have a much more profound impact on U.S. interest rates and the growth outlook. Monitoring the PBoC's balance sheet will become increasingly important for Fed watchers. Yan Wang, Senior Vice President China Investment Strategy yanw@bcaresearch.com 1 Please see China Investment Strategy Special Report, "Delving Into the PBoC'S Balance Sheet," dated July 27, 2011, available at cis.bcaresearch.com. 2 Please see China Investment Strategy Special Report, "More On The Chinese Debt Debate," dated April 20, 2017, available at cis.bcaresearch.com. 3 Please see China Investment Strategy Weekly Report, "Can The RMB Appreciate Against The Dollar, Again?" dated May 11, 2017, available at cis.bcaresearch.com. Table 2 offers a simplified balance sheet of the People's Bank of China. Foreign assets still account for 65.6% of its total assets, down from a peak of 83% in 2014. In comparison, most other major central banks' assets are predominantly domestic government bonds. The explosive growth of the PBoC's holding of foreign assets had been the only source of its balance sheet expansion before 2015. In the past two years the PBoC's domestic assets have increased sharply. Overall the PBoC's balance sheet has stayed flat in the RMB terms. PBoC's holding of foreign and domestic assets has been matched by expansion of reserve money (monetary base) on the liability side of the PBoC's balance sheet, including currency issuances (M0 and cash in the vaults of depository institutions) and deposits of commercial banks in the central bank. Commercial banks' reserve deposits at the PBoC have continued to grow even though the PBoC balance sheet expansion has stalled. (Chart 2) Table 2The PBoC's Balance Sheet Chart 2 PBoC holdings of foreign assets include foreign exchange reserves and gold. Foreign reserves currently account for 63% of PBoC total assets, compared with a peak of 84% in 2014. Official record shows that gold is still a negligible share of its total assets. Other major items on the asset side of the PBoC's balance sheet include claims on the government, commercial banks and other financial corporations. The PBoC's claims on the government (entirely on the central government) account for 4.5% of its total assets. In 2007 the government set up a sovereign wealth management fund to manage part of the country's reserves. The government issued bonds to the PBoC in exchange for foreign exchange reserves, which was used as capital of the investment firm. Legally the PBoC is forbidden to directly hold government bonds. The PBoC's claims on other depository corporations (commercial banks) include loans and rediscounts to commercial banks and the net amount of repurchase agreements, which has increased sharply since 2016. The PBoC claims on other commercial banks were a major policy tool to control liquidity in the early 2000s. The central bank's claims on other financial corporations mainly include loans to the asset management firms that the government set up in the late 1990s to deal with bad loans spun off from commercial banks. There has been no change in this item in recent years. (Chart 3 and Chart 4) Chart 3 Chart 4 On the liability side of the PBoC's balance sheet, the dominant item is reserve money, which includes currency issuances and deposits of depository corporations. Taken together these items account for almost 90% of banks' total liabilities. However, currency issuances (M0 and cash in vault) have been hovering around 20% of the PBoC balance sheet in recent years. Deposits of depository corporations account for about 66%. Deposits of commercial banks in the central bank include required and free reserves. Currency issuance and free reserves make up China's "high power money" that can result in a much larger increase in money supply through the money multiplier. Therefore, adjusting the reserve requirement ratio (RRR) on banks has been a key policy tool for the PBoC to control "loanable" funds and liquidity. The central bank, however, been reluctant to adjust RRR since 2016 despite continued liquidity outflow. Commercial banks used to hold large amounts of free reserves with the central bank, which however have declined sharply in recent years. The massive reserves of commercial banks in the PBoC offer a critical liquidity buffer for banks at times of crisis. As banks' free reserves have been running thin, there is a building case for an RRR reduction in coming months. (Chart 5 and Chart 6) Chart 5 Chart 6 Other major items on the liability side of the PBoC's balance sheet include bond issues, government deposits and foreign liabilities. The central bank started to issue bonds (notes) in 2002 as a way to sterilize foreign capital inflows, a tool that has essentially been phased out. Currently, total outstanding bonds amount to RMB 50 billion, a mere 0.1% of the PBoC total liability, compared with almost 30% in 2007. The PBoC's foreign liabilities are deposits of international financial institutions, which account for a negligible share of its total assets. Government deposits account for 8.4% of the central bank's total liabilities, or RMB 2.88 trillion at the end of April 2017. The PBoC regularly auctions off fiscal deposits to commercial banks as a way to adjust interbank liquidity. (Chart 7 and Chart 8) Chart 7 Chart 8 There are four main items on the PBoC's balance sheet that the central bank uses at its discretion to manage domestic liquidity: claims on depository corporations (banks), deposits of depository corporations, liabilities to the government (fiscal deposits) and bond issues. Claims on depository corporations are on the asset side, and include loans and rediscounts to commercial banks and the net amount of repurchase agreements. The PBoC has significantly expanded some new liquidity tools, such as various lending facilities and open market operations. These assets are mostly short term, allowing the central bank flexibility to adjust the quantity quickly. Reserve deposits of commercial banks, central bank bond issues and fiscal deposits are on the liability side of the PBoC's balance sheet, but reserve deposits play by far the largest role in the central bank's sterilization efforts. Commercial banks reserve deposits are still hovering around record high levels. (Chart 9 and Chart 10) Chart 9 Chart10 Taken together, the ebbs and flows of the PBoC's sterilization operations coincide with the pace of country's foreign reserve accumulation. The PBoC was able to "sterilize" about 80% of foreign capital inflow before 2015, and it has been quickly adjusting its balance sheet to offset domestic capital outflows in the past two years. All these items on the PBoC's balance sheet should be cross-checked to assess its liquidity operations, rather than focusing on one item. Looking forward, the PBoC's liquidity operations will remain contingent on the situation of cross-border capital flows in the near term, and its monetary independence will remain compromised. Over the long run, a free-floating RMB exchange rate will diminish the purpose of PBoC's precautionary holdings of foreign reserves, which will in turn impact how the central bank manages its balance sheet for domestic considerations. (Chart 11 and Chart 12) Chart 11 Chart 12 Cyclical Investment Stance Equity Sector Recommendations

Feature EM risk assets refuse to sell off - regardless of new information and developments that historically would have caused these markets to tumble. Indeed, political turmoil and changes in Brazil and South Africa - two high-beta EM markets - have so far had limited impact on flows and market dynamics. Moreover, while our Reflation Confirming Indicator has rolled over, EM share prices have not reacted at all (Chart I-1). EM stocks have also decoupled with the equal-weighted average of global mining and energy equity indexes (Chart I-2). Chart 1Reflation Confirming Indicator And ##br##EM Stocks Chart 2Commodities Share Prices And ##br##EM Equities: Unsustainable Divergence We do not subscribe to the thesis that EM assets have permanently decoupled from both commodities and their domestic credit cycles, and that tried and tested indicators no longer work. Technology and social media share prices have been instrumental to this latest decoupling, as we wrote in last week's report.1 This group of stocks is in a full-blown mania phase, and it is hard to know when this will end. Yet, exponential price moves always occur at the end of a bull market, and are typically followed by bear markets. As we elaborated in last week's report, the investment call on social media and internet stocks is a bottom-up - not macro - call. Top-down analysis can add some value on the semiconductor cycle, and we suggested last week that it is likely topping out. This week new data releases support the thesis that Asian/global trade in general and the semiconductor cycle in particular are already decelerating. Korean exports data for the first 20 days of May, Japanese preliminary manufacturing PMI for May, and Taiwanese manufacturing output volume growth for April have all decelerated (Chart I-3). Finally, one technical piece of evidence that this rally is late is relative weakness in the equal-weighted MSCI equity indexes. In the EM space, the equally-weighted individual stock index has fared poorly against the EM market cap-weighted index since May 2016 (Chart I-4, top panel). In the U.S., the same measure of market breadth has deteriorated since December 2016 (Chart I-4, bottom panel). Chart 3Asia's Manufacturing Growth ##br##Is Already Decelerating Chart 4The EM And U.S. Equity Rally ##br##Has Been Driven By Large-Cap Stocks Bottom Line: EM financial markets are in the midst of irrational exuberance. The rally is late, but it is impossible to time the top. The forthcoming selloff will be large and protracted. Beware Of China's Budding Growth Slump Interest rates have risen in China sufficiently enough to cause a major growth slowdown in the mainland economy (Chart I-5). Liquidity tightening amid a lingering credit bubble could not be a more dangerous combination. In this context, financial markets are extremely complacent on EM/China plays. China's liquidity tightening continues, and is bound to create a decisive growth relapse in the months ahead, as well as dampen exports in countries that sell to China (Chart I-6). Chart 5China Growth To Slump Chart 6Exports To China To Slump Not only is the People's Bank of China (PBoC) guiding interest rates higher, but there is an ongoing regulatory crackdown on the financial system. Regulators are forcing banks to bring Wealth Management Products (WMPs) and other off-balance-sheet items onto their balance sheets. As a result, banks' capital adequacy and risk matrixes will deteriorate, and they will be forced to slow down credit creation. Chart 7EM Share Prices Ex. Tech Have Not Broken Out Remarkably, policymakers are determined to get things under control. According to The Wall Street Journal,2 key policymakers have issued strongly worded statements. "Strong medicine must be prescribed," said Guo Shuqing, chairman of the China Banking Regulatory Commission (CRBC), according to people familiar with the matter. "If the banking industry gets into a mess," he added, "I will resign." He was appointed as the head of the CRBC last October, and likely has a mandate from the President to tackle speculative excesses in the financial system. In its first quarter Monetary Policy Implementation Report,3 the PBoC repeatedly used the phrase "preventing bubbles." Besides, in his statements, the chairman of the PBoC has frequently emphasized the need to normalize credit growth and curb speculative activities. The former head of the insurance regulator, who has been "accommodating" and "tolerant" of risky activities by insurance companies, was jailed last fall for corruption. These are strong indications confirming that policymakers are determined to curb speculative financial activities. Provided how entrenched and large various speculative financial schemes and the credit bubble have become in China, it will be impossible to tackle speculative excesses without a slowdown in overall credit growth and associated harm to the real economy. This is not to say that policymakers are tightening with intentions to cause a growth collapse. Policymakers in all countries always tighten to cap inflation or credit excesses or normalize interest rates - i.e., they never tighten to cause a major shock to the real economy. This applies to Chinese policymakers at the moment, especially ahead of the party Congress later this year. That said, when the existing imbalances in the economy or financial system are sufficiently large, even minor tightening can cause a financial accident or growth relapse. It is not within policymakers' powers to predict or prevent it. They may alter their policy after the fact, but markets will sell off considerably beforehand. We do not know exactly how financial dynamics in China will evolve in the months ahead, but we are certain that the market consensus is too complacent and that EM asset prices are at major risk. Bottom Line: It is impossible to predict financial accidents (stress among specific institutions) but we are certain that China's credit growth and, consequently, capital spending are bound to slow considerably in the coming months. This bodes ill for producers of commodities and industrial goods both within and outside China. Accordingly, EM risk assets will suffer the most. As a final note, EM ex-technology share prices have not yet broken out and we do expect them to relapse from the current levels (Chart I-7). Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com 1 Please see Emerging Markets Strategy Weekly Report titled, "Can Tech Drive EM Stocks Higher?," dated May 17, 2017, link available at ems.bcaresearch.com. 2 Lingling Wei and Chao Deng, "China's War on Debt Causes Stocks to Drop, Bond Yields to Shoot Up and Defaults to Rise," May 5, 2017, The Wall Street Journal. 3 Please refer to http://www.pbc.gov.cn/zhengcehuobisi/125207/125227/125957/3307990/3307409/index.html (In Chinese only). Equity Recommendations Fixed-Income, Credit And Currency Recommendations