Emerging Markets

Special Report

Highlights Slower nominal GDP growth explains virtually all of the increase in China's debt-to-GDP ratio over the past ten years. The authorities were unwilling to restrain debt growth as it became obvious that nominal income was decelerating because this would have only exacerbated the economic downturn. Excess private-sector savings forced the Chinese government to rely on debt-financed investment by state-owned companies (SOE) and local governments in order to keep aggregate demand elevated. Financial deregulation also encouraged debt accumulation. Debt growth linked to speculative activity can be curbed without endangering the economy, but a lasting solution to the surplus savings problem will require consumers to spend more. This will take a while. At some point over the next few years, the central government will transfer a large fraction of SOE and local government debt onto its own balance sheet. The risk to investors is that this "debt nationalization" happens reactively rather than proactively. Feature If there are too many pro-cyclical factors in the economy, cyclical fluctuations are magnified and there is excessive optimism during the period, accumulating contradictions that could lead to the so-called Minsky Moment. - Zhou Xiaochuan, Former Governor of the People's Bank of China, October 19, 2017 The Calm Before The Storm? Stability begets instability. That is the nature of business cycles, Hyman Minsky famously argued. Rising confidence leads to excessive risk-taking, higher asset prices, and mounting economic imbalances. Eventually the mood sours. Like Wile E. Coyote running off a cliff, investors look down and see that there is nothing but thin air between them and the ground below. Panic ensues. Is China on the verge of its own Minsky Moment? A glance at the evolution of its debt-to-GDP ratio would certainly say so. But before running towards the exit door, consider the following: People have been fretting about spiraling Japanese government debt levels for over twenty years now. And yet, interest rates remain at rock-bottom levels in Japan. China's Savings Glut In many respects, China finds itself facing similar problems to those that have haunted Japan. The simultaneous bust in equity and real estate prices in 1990 sent Japan's private sector into a prolonged deleveraging cycle (Chart 1). In order to prop up demand, the Japanese government was forced to run large budget deficits. In effect, the government had to absorb the excess savings of the private sector with its own dissavings. The abundance of domestic private-sector savings forestalled a financial crisis, but it also led to today's gross government debt-to-GDP ratio of 240%. Like Japan, China suffers from a dearth of spending, or equivalently, an abundance of savings. The IMF estimates that Chinese gross national savings reached 46% of GDP in 2017. While this is down from a peak of 52% of GDP in 2008, it is still abnormally high for any major economy, even by emerging market standards (Chart 2). Chart 1 Japan Relied On Large Fiscal Deficits And Current Account Surpluses To Offset The Rise In Private-Sector Savings Chart 2China's Savings Rate Stands Out Even By EM Standards By definition, whatever a country saves must either be invested domestically or channeled abroad via a current account surplus. China's savings rate has edged lower over the past ten years, but its current account surplus has dropped even more, falling from nearly 10% of GDP in 2007 to 1.4% of GDP at present. As a result, investment as a share of GDP has actually risen to 44%, a three-point increase since 2007 (Chart 3). The decline in China's current account surplus was inevitable (Chart 4). In 2007, China accounted for 6% of global GDP in dollar terms. Today it accounts for 15%. Having a massively undervalued currency, as China had in 2007, is just not politically tenable anymore, especially with Donald Trump in the White House. Simply put, China has become too big to continue exporting its way out of its problems. Chart 3Since The Great Financial Crisis, Chinese Savings Have Been Channeled Into Domestic Investment, Not Funneled Abroad Chart 4Undervalued Currency And Massive Current Account Surplus: Modus Operandi Of The Past Debt As The Conduit Between Savings And Investment How does a country transform savings into investment? In an economy like China where the stock market at times appears to be little more than a casino, the answer is that credit markets must play the dominant role. Households or firms with surplus savings park their funds in banks or other financial institutions. These institutions channel the savings to willing borrowers. Debt ends up being the natural byproduct of surplus savings. China is still a relatively poor country with a lot of catch-up potential. Capital-per-worker is a fraction of what it is among advanced economies (Chart 5). Even with its bleak demographics, China would need to grow by around 6% per year over the next few years just to converge with South Korea in output-per-worker by 2050 (Chart 6). All this means that China needs to invest more than most other economies, which is only possible if it saves more than other economies. Chart 5China Has More Catching Up To Do (1) Chart 6China Has More Catching Up To Do (2) Unfortunately, one can have too much of a good thing. The fact that China's capital stock-to-output ratio has risen dramatically in recent years means that the economy is already investing too much. And the optimal amount of investment will only fall over time as potential GDP growth continues to decelerate. Unless savings come down, China will find itself increasingly awash in excess capacity. Chart 7If Only GDP Growth Did Not ##br## Decelerate Over The Past Ten Years Slower trend growth will also make deleveraging more difficult to achieve. The overall stock of nonfinancial debt grew at an annualized rate of 18.8% between 2008 and 2017. Notably, this growth rate was not much higher than the one of 16.5% between 2003 and 2007 - a period when the debt-to-GDP ratio was broadly stable. The main difference between the two periods lies in the denominator of the debt-to-GDP ratio, not in the numerator: Nominal GDP expanded at an annualized rate of 11.2% between 2008 and 2017, a sizable retreat from the pace of 18.4% between 2003 and 2007. Chart 7 shows that the debt-to-GDP ratio today would be virtually identical to its end-2007 level had nominal GDP continued to grow at its 2003-2007 pace over the past ten years. Financial Deregulation Has Exacerbated The Debt Problem The Chinese government's reluctance to crack down on credit growth was motivated by the desire to support aggregate demand. However, in turning a blind eye to what was happening in credit markets, a lot of debt was generated that was not directly tied to the intermediation of savings into investment. Chart 8Debt And Capital Accumulation Went Hand In Hand Debt can be created when someone borrows money to finance the purchase of goods or services. Debt can also be created when someone borrows money to finance the purchase of pre-existing assets. Crucially, while the former typically requires additional "savings" (i.e., someone needs to reduce their spending relative to their income), the latter does not.1 Granted, savings can still play an indirect role in facilitating debt-financed asset purchases. Financial assets are typically backed by something of value. A mortgage is backed by a piece of property. A corporate bond is backed by both the tangible and intangible capital that a firm possesses. The more a country has been able to save over time, the larger its capital stock will be. China, of course, has been saving like crazy for years. It is thus no surprise that its debt-to-GDP ratio has soared as its capital stock has expanded (Chart 8). Financial deregulation in China has allowed a large share of its capital stock to repeatedly shift hands. Debt has often been created in the process. The problem is that debt-financed asset purchases drive up asset prices, sometimes to unsustainable levels. And the higher the price of the asset, the greater the risk that it will not yield enough income to cover the borrowing costs. When asset prices are rising, borrowers and lenders are apt to disregard this risk, figuring that they can always sell the asset at a high enough price to pay back the loan. But once prices start falling, reality sets in very quickly. Stability begets instability. Consumers Need To Step Up The authorities are keenly aware of the risks discussed above. This is the key reason why they are clamping down on the shadow banking system, which has increasingly become the main source of speculative lending in China. We expect the pressure on shadow banks to persist in 2018. This will continue to weigh on credit growth. The more vexing challenge is how to reduce excessive household savings. The government's current strategy of cramming down the capital stock by taking out excess capacity from sectors such as steel, coal, and solar may be better than nothing, but it still pales in comparison to a strategy of encouraging consumer spending. Higher consumer spending would obviate the need for state-owned companies and local governments to keep people employed in make-work projects. The good news is that there are plenty of ways that China can boost household consumption. Government spending on education, health care, and pensions as a share of GDP is close to half of the OECD average (Chart 9). Increasing social transfer payments would give households the wherewithal to spend more. Unlike in most countries, the poor in China are net savers (Chart 10). Expanding the social safety net would discourage precautionary savings. Chart 9Chinese Social Welfare Spending ##br##Is Lagging The OECD Average Chart 10Low Income Households Are Net ##br##Savers In China The Chinese income tax structure is fairly regressive. Poor households face an effective income tax rate exceeding 40%. This is well above OECD norms (Chart 11).2 A more progressive tax system would boost spending among poorer households. It would also curb inequality, which has increased sharply over the past few decades (Chart 12). The saving rate among the richest 10% of Chinese earners is close to 50%. Policies that shift income from the rich to the poor would reduce overall household savings. Chart 11High Tax Burden For ##br##Low Income Households In China Chart 12Shifting Income To Poorer Households Would Reduce ##br##China's Household Savings Rate Debt Nationalization Is Inevitable Chart 13Ratio Of Workers-To-Consumers Is Peaking,##br## And China Is No Exception Realistically, reforms aimed at encouraging consumption will take a while to implement. In the meantime, debt levels are likely to keep rising. Much of China's debt burden remains on the books of state-owned companies and local governments. At some point over the next few years, the central government will transfer a large fraction of this debt onto its own balance sheet. This would ease concerns about a mass wave of defaults. The key question for investors is whether this de facto "debt nationalization" is done proactively or reactively in response to a crisis. If the latter occurs, investors should steer clear of Chinese assets, as well as China-related plays such as commodities and commodity currencies. If the former pans out, global risk assets could rally. While the truth will fall somewhere between those two extremes, our bet is that the proactive view will prove closer to the mark, at least relative to market expectations (keep in mind that Chinese banks are trading below book value, so a lot of bad news has already been priced in). The Chinese authorities talk a lot about the importance of reducing moral hazard, but in practice, they have shown very little tolerance for defaults. Just as they did in the early 2000s, government leaders could commission state-owned asset management companies to purchase distressed debt from banks and other lenders at inflated prices. Chinese financials, which are nearly 70% of the H-share index, will benefit. Will investors balk at the prospect of the Chinese government blowing out the budget deficit in order to rescue insolvent borrowers? There might be some short-term panic, but as has been the case with Japan, as long as there are plenty of excess domestic savings to go around, the risk of a debt crisis will remain minimal. Indeed, the issuance of more government debt would help alleviate what has become a critical problem for Chinese savers: The lack of safe, liquid domestic assets available for purchase. What is true, from a longer-term perspective, is that the combination of higher debt and slower growth will eventually create a strong incentive for the Chinese government to inflate away debt. As in many other countries, China's "support ratio" -- broadly defined as the ratio of workers-to-consumers -- has peaked (Chart 13). As the growth of output and income falls behind consumption growth, China's savings glut will become a thing of the past. Rather than raising rates, the PBOC will just let the economy overheat. Such a day of reckoning is probably still at least five years away, but eventually inflation will return to China. Concluding Thoughts On The Current Market Environment A true "Minsky moment" in China - one where the financial sector seizes up due to spiraling fears of bankruptcies and defaults - is not in the cards. Nevertheless, China's economy is slowing, and growth is likely to decelerate further over the next few quarters as the authorities restrain credit growth and the property market continues to cool. The slowdown in Chinese growth is occurring at the same time as the economic data has been deteriorating around the world. The equity component of our MacroQuant model - which is highly sensitive to changes in the direction of growth - has been in bearish territory for two straight months (Chart 14). Our base case remains that global growth will stabilize over the next few months at an above-trend pace. Global bond yields are still near record-low levels and fiscal policy is moving in a more stimulative direction (Chart 15). It would be odd for the global economy to deteriorate sharply in such an environment. Chart 14MacroQuant Model Suggests Caution Is Warranted Trade protectionism is an obvious risk to this sanguine cyclical view. BCA has long argued that globalization is under threat from the combination of rising populism and the end of America's role as the world's sole superpower. However, the retreat from globalization will occur in fits and starts. Just as investors were overly complacent about protectionism a few months ago, they have become overly alarmist now. Both the U.S. and China have a strong incentive to reach a mutually-satisfying agreement over trade. President Trump has been able to shrug off the decline in equities because his approval rating has actually risen during the selloff (Chart 16). However, if the problems on Wall Street begin to show up on Main Street - as is likely to happen if stocks continue to fall - Trump will change his tune. Chart 15Global Economy Buttressed By ##br##Accommodative Fiscal And Monetary Policy Chart 16Trump's Approval Rating Has ##br##Actually Risen During Equity Selloff For its part, the Chinese government is also looking to strike a deal. The U.S. exported only $131 billion in goods to China last year. This is already less than the $150 billion in Chinese goods that Trump has targeted for tariffs. China simply cannot win a tit-for-tat trade war with the United States. Bottom Line: The near-term picture for global equities and other risk assets is murky, but the 12-month cyclical outlook is still reasonably upbeat. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 1 For instance, if someone buys stock on margin or takes out a second mortgage on their house, new debt is created without anyone having to cut back on spending. In the context of China, imagine a financial institution which funds the purchase of a building by issuing a certificate of deposit or by selling a "wealth management" product. Both the asset and liability side of the financial institution's balance sheet go up (i.e., new debt is created). Suppose further that the company that sold the building puts the proceeds into a certificate of deposit or wealth management product. The entire transaction is self-financing. The example above illustrates that debt can go up in some situations even if everyone's spending habits remain the same. The need to intermediate savings is one source of debt growth, but it does not have to be the only one. 2 Please see "People's Republic Of China: Selected Issues," IMF Country Report, dated August 15, 2017.

Special Report

Highlights Slower nominal GDP growth explains virtually all of the increase in China's debt-to-GDP ratio over the past ten years. The authorities were unwilling to restrain debt growth as it became obvious that nominal income was decelerating because this would have only exacerbated the economic downturn. Excess private-sector savings forced the Chinese government to rely on debt-financed investment by state-owned companies (SOE) and local governments in order to keep aggregate demand elevated. Financial deregulation also encouraged debt accumulation. Debt growth linked to speculative activity can be curbed without endangering the economy, but a lasting solution to the surplus savings problem will require consumers to spend more. This will take a while. At some point over the next few years, the central government will transfer a large fraction of SOE and local government debt onto its own balance sheet. The risk to investors is that this "debt nationalization" happens reactively rather than proactively. Feature If there are too many pro-cyclical factors in the economy, cyclical fluctuations are magnified and there is excessive optimism during the period, accumulating contradictions that could lead to the so-called Minsky Moment. - Zhou Xiaochuan, Former Governor of the People's Bank of China, October 19, 2017 The Calm Before The Storm? Stability begets instability. That is the nature of business cycles, Hyman Minsky famously argued. Rising confidence leads to excessive risk-taking, higher asset prices, and mounting economic imbalances. Eventually the mood sours. Like Wile E. Coyote running off a cliff, investors look down and see that there is nothing but thin air between them and the ground below. Panic ensues. Is China on the verge of its own Minsky Moment? A glance at the evolution of its debt-to-GDP ratio would certainly say so. But before running towards the exit door, consider the following: People have been fretting about spiraling Japanese government debt levels for over twenty years now. And yet, interest rates remain at rock-bottom levels in Japan. China's Savings Glut In many respects, China finds itself facing similar problems to those that have haunted Japan. The simultaneous bust in equity and real estate prices in 1990 sent Japan's private sector into a prolonged deleveraging cycle (Chart 1). In order to prop up demand, the Japanese government was forced to run large budget deficits. In effect, the government had to absorb the excess savings of the private sector with its own dissavings. The abundance of domestic private-sector savings forestalled a financial crisis, but it also led to today's gross government debt-to-GDP ratio of 240%. Like Japan, China suffers from a dearth of spending, or equivalently, an abundance of savings. The IMF estimates that Chinese gross national savings reached 46% of GDP in 2017. While this is down from a peak of 52% of GDP in 2008, it is still abnormally high for any major economy, even by emerging market standards (Chart 2). Chart 1 Japan Relied On Large Fiscal Deficits And Current Account Surpluses To Offset The Rise In Private-Sector Savings Chart 2China's Savings Rate Stands Out Even By EM Standards By definition, whatever a country saves must either be invested domestically or channeled abroad via a current account surplus. China's savings rate has edged lower over the past ten years, but its current account surplus has dropped even more, falling from nearly 10% of GDP in 2007 to 1.4% of GDP at present. As a result, investment as a share of GDP has actually risen to 44%, a three-point increase since 2007 (Chart 3). The decline in China's current account surplus was inevitable (Chart 4). In 2007, China accounted for 6% of global GDP in dollar terms. Today it accounts for 15%. Having a massively undervalued currency, as China had in 2007, is just not politically tenable anymore, especially with Donald Trump in the White House. Simply put, China has become too big to continue exporting its way out of its problems. Chart 3Since The Great Financial Crisis, Chinese Savings Have Been Channeled Into Domestic Investment, Not Funneled Abroad Chart 4Undervalued Currency And Massive Current Account Surplus: Modus Operandi Of The Past Debt As The Conduit Between Savings And Investment How does a country transform savings into investment? In an economy like China where the stock market at times appears to be little more than a casino, the answer is that credit markets must play the dominant role. Households or firms with surplus savings park their funds in banks or other financial institutions. These institutions channel the savings to willing borrowers. Debt ends up being the natural byproduct of surplus savings. China is still a relatively poor country with a lot of catch-up potential. Capital-per-worker is a fraction of what it is among advanced economies (Chart 5). Even with its bleak demographics, China would need to grow by around 6% per year over the next few years just to converge with South Korea in output-per-worker by 2050 (Chart 6). All this means that China needs to invest more than most other economies, which is only possible if it saves more than other economies. Chart 5China Has More Catching Up To Do (1) Chart 6China Has More Catching Up To Do (2) Unfortunately, one can have too much of a good thing. The fact that China's capital stock-to-output ratio has risen dramatically in recent years means that the economy is already investing too much. And the optimal amount of investment will only fall over time as potential GDP growth continues to decelerate. Unless savings come down, China will find itself increasingly awash in excess capacity. Chart 7If Only GDP Growth Did Not ##br## Decelerate Over The Past Ten Years Slower trend growth will also make deleveraging more difficult to achieve. The overall stock of nonfinancial debt grew at an annualized rate of 18.8% between 2008 and 2017. Notably, this growth rate was not much higher than the one of 16.5% between 2003 and 2007 - a period when the debt-to-GDP ratio was broadly stable. The main difference between the two periods lies in the denominator of the debt-to-GDP ratio, not in the numerator: Nominal GDP expanded at an annualized rate of 11.2% between 2008 and 2017, a sizable retreat from the pace of 18.4% between 2003 and 2007. Chart 7 shows that the debt-to-GDP ratio today would be virtually identical to its end-2007 level had nominal GDP continued to grow at its 2003-2007 pace over the past ten years. Financial Deregulation Has Exacerbated The Debt Problem The Chinese government's reluctance to crack down on credit growth was motivated by the desire to support aggregate demand. However, in turning a blind eye to what was happening in credit markets, a lot of debt was generated that was not directly tied to the intermediation of savings into investment. Chart 8Debt And Capital Accumulation Went Hand In Hand Debt can be created when someone borrows money to finance the purchase of goods or services. Debt can also be created when someone borrows money to finance the purchase of pre-existing assets. Crucially, while the former typically requires additional "savings" (i.e., someone needs to reduce their spending relative to their income), the latter does not.1 Granted, savings can still play an indirect role in facilitating debt-financed asset purchases. Financial assets are typically backed by something of value. A mortgage is backed by a piece of property. A corporate bond is backed by both the tangible and intangible capital that a firm possesses. The more a country has been able to save over time, the larger its capital stock will be. China, of course, has been saving like crazy for years. It is thus no surprise that its debt-to-GDP ratio has soared as its capital stock has expanded (Chart 8). Financial deregulation in China has allowed a large share of its capital stock to repeatedly shift hands. Debt has often been created in the process. The problem is that debt-financed asset purchases drive up asset prices, sometimes to unsustainable levels. And the higher the price of the asset, the greater the risk that it will not yield enough income to cover the borrowing costs. When asset prices are rising, borrowers and lenders are apt to disregard this risk, figuring that they can always sell the asset at a high enough price to pay back the loan. But once prices start falling, reality sets in very quickly. Stability begets instability. Consumers Need To Step Up The authorities are keenly aware of the risks discussed above. This is the key reason why they are clamping down on the shadow banking system, which has increasingly become the main source of speculative lending in China. We expect the pressure on shadow banks to persist in 2018. This will continue to weigh on credit growth. The more vexing challenge is how to reduce excessive household savings. The government's current strategy of cramming down the capital stock by taking out excess capacity from sectors such as steel, coal, and solar may be better than nothing, but it still pales in comparison to a strategy of encouraging consumer spending. Higher consumer spending would obviate the need for state-owned companies and local governments to keep people employed in make-work projects. The good news is that there are plenty of ways that China can boost household consumption. Government spending on education, health care, and pensions as a share of GDP is close to half of the OECD average (Chart 9). Increasing social transfer payments would give households the wherewithal to spend more. Unlike in most countries, the poor in China are net savers (Chart 10). Expanding the social safety net would discourage precautionary savings. Chart 9Chinese Social Welfare Spending ##br##Is Lagging The OECD Average Chart 10Low Income Households Are Net ##br##Savers In China The Chinese income tax structure is fairly regressive. Poor households face an effective income tax rate exceeding 40%. This is well above OECD norms (Chart 11).2 A more progressive tax system would boost spending among poorer households. It would also curb inequality, which has increased sharply over the past few decades (Chart 12). The saving rate among the richest 10% of Chinese earners is close to 50%. Policies that shift income from the rich to the poor would reduce overall household savings. Chart 11High Tax Burden For ##br##Low Income Households In China Chart 12Shifting Income To Poorer Households Would Reduce ##br##China's Household Savings Rate Debt Nationalization Is Inevitable Chart 13Ratio Of Workers-To-Consumers Is Peaking,##br## And China Is No Exception Realistically, reforms aimed at encouraging consumption will take a while to implement. In the meantime, debt levels are likely to keep rising. Much of China's debt burden remains on the books of state-owned companies and local governments. At some point over the next few years, the central government will transfer a large fraction of this debt onto its own balance sheet. This would ease concerns about a mass wave of defaults. The key question for investors is whether this de facto "debt nationalization" is done proactively or reactively in response to a crisis. If the latter occurs, investors should steer clear of Chinese assets, as well as China-related plays such as commodities and commodity currencies. If the former pans out, global risk assets could rally. While the truth will fall somewhere between those two extremes, our bet is that the proactive view will prove closer to the mark, at least relative to market expectations (keep in mind that Chinese banks are trading below book value, so a lot of bad news has already been priced in). The Chinese authorities talk a lot about the importance of reducing moral hazard, but in practice, they have shown very little tolerance for defaults. Just as they did in the early 2000s, government leaders could commission state-owned asset management companies to purchase distressed debt from banks and other lenders at inflated prices. Chinese financials, which are nearly 70% of the H-share index, will benefit. Will investors balk at the prospect of the Chinese government blowing out the budget deficit in order to rescue insolvent borrowers? There might be some short-term panic, but as has been the case with Japan, as long as there are plenty of excess domestic savings to go around, the risk of a debt crisis will remain minimal. Indeed, the issuance of more government debt would help alleviate what has become a critical problem for Chinese savers: The lack of safe, liquid domestic assets available for purchase. What is true, from a longer-term perspective, is that the combination of higher debt and slower growth will eventually create a strong incentive for the Chinese government to inflate away debt. As in many other countries, China's "support ratio" -- broadly defined as the ratio of workers-to-consumers -- has peaked (Chart 13). As the growth of output and income falls behind consumption growth, China's savings glut will become a thing of the past. Rather than raising rates, the PBOC will just let the economy overheat. Such a day of reckoning is probably still at least five years away, but eventually inflation will return to China. Concluding Thoughts On The Current Market Environment A true "Minsky moment" in China - one where the financial sector seizes up due to spiraling fears of bankruptcies and defaults - is not in the cards. Nevertheless, China's economy is slowing, and growth is likely to decelerate further over the next few quarters as the authorities restrain credit growth and the property market continues to cool. The slowdown in Chinese growth is occurring at the same time as the economic data has been deteriorating around the world. The equity component of our MacroQuant model - which is highly sensitive to changes in the direction of growth - has been in bearish territory for two straight months (Chart 14). Our base case remains that global growth will stabilize over the next few months at an above-trend pace. Global bond yields are still near record-low levels and fiscal policy is moving in a more stimulative direction (Chart 15). It would be odd for the global economy to deteriorate sharply in such an environment. Chart 14MacroQuant Model Suggests Caution Is Warranted Trade protectionism is an obvious risk to this sanguine cyclical view. BCA has long argued that globalization is under threat from the combination of rising populism and the end of America's role as the world's sole superpower. However, the retreat from globalization will occur in fits and starts. Just as investors were overly complacent about protectionism a few months ago, they have become overly alarmist now. Both the U.S. and China have a strong incentive to reach a mutually-satisfying agreement over trade. President Trump has been able to shrug off the decline in equities because his approval rating has actually risen during the selloff (Chart 16). However, if the problems on Wall Street begin to show up on Main Street - as is likely to happen if stocks continue to fall - Trump will change his tune. Chart 15Global Economy Buttressed By ##br##Accommodative Fiscal And Monetary Policy Chart 16Trump's Approval Rating Has ##br##Actually Risen During Equity Selloff For its part, the Chinese government is also looking to strike a deal. The U.S. exported only $131 billion in goods to China last year. This is already less than the $150 billion in Chinese goods that Trump has targeted for tariffs. China simply cannot win a tit-for-tat trade war with the United States. Bottom Line: The near-term picture for global equities and other risk assets is murky, but the 12-month cyclical outlook is still reasonably upbeat. Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 1 For instance, if someone buys stock on margin or takes out a second mortgage on their house, new debt is created without anyone having to cut back on spending. In the context of China, imagine a financial institution which funds the purchase of a building by issuing a certificate of deposit or by selling a "wealth management" product. Both the asset and liability side of the financial institution's balance sheet go up (i.e., new debt is created). Suppose further that the company that sold the building puts the proceeds into a certificate of deposit or wealth management product. The entire transaction is self-financing. The example above illustrates that debt can go up in some situations even if everyone's spending habits remain the same. The need to intermediate savings is one source of debt growth, but it does not have to be the only one. 2 Please see "People's Republic Of China: Selected Issues," IMF Country Report, dated August 15, 2017. Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights Apart from rising geopolitical tensions, our main macro themes remain a growth slowdown in China and a rise in U.S. core inflation. This combination bodes ill for EM financial markets. Continue underweighting EM stocks, credit and currencies versus their DM peers. Subsiding NAFTA risks argue for overweighting Mexican stocks within an EM equity portfolio. This is in line with our recent upgrade of Mexican local and U.S. dollar sovereign bonds as well as the peso's outlook versus their EM peers. A new trade: Fixed-income trades should bet on yield curve steepening in Mexico by paying 10-year swap rates and receiving 2-year rates. Close overweight Russian markets positions in the wake of escalating U.S. sanctions. Feature Before discussing Mexico and Russia, we offer an update on our thoughts on the overall market outlook. EM: Looking Under The Hood Investor sentiment remains buoyant on global risk assets, and the buy-on-dips mentality remains well entrenched. On the surface, investors are not finding enough reasons to turn negative on global or EM risk markets. Nevertheless, when looking under the EM hood, we see several leading and coincident indicators that are beginning to flash red. Not only do geopolitics and the U.S.-China trade confrontation pose downside risks, there are also several macro developments that are turning from tailwinds to headwinds for EM risk assets. Specifically: EM manufacturing and Asian trade cycles have probably topped out. The relative total return (carry included) of three equally weighted EM1 (ZAR, BRL and CLP) and three DM (AUD, NZD and CAD) commodities currencies versus an equally weighted average of two safe-haven currencies - the Japanese yen and Swiss franc - has relapsed since early this year, coinciding with the rollover in the EM manufacturing PMI index (Chart I-1). This currency ratio is herein referred to as the risk-on/safe-haven currency ratio. Chart I-1Risk On / Safe-Haven Currency Ratio And EM Manufacturing PMI The risk-on/safe-haven currency ratio also correlates with the average of new and backlog orders components of China's manufacturing PMI (Chart I-2). The latter does not herald an upturn in this currency ratio at the moment. Share prices of global machinery, chemicals and mining companies have so far underperformed the overall global equity index in this selloff, as exhibited in Chart I-3. Chart I-2China's Industrial Cycle Has Rolled Over Chart I-3Global Cyclicals Have Underperformed, Though Not Tech Potential trade wars, the setback in technology stocks and a resurgence of volatility in global equity markets have recently dominated news headlines. Yet, the underperformance of China-exposed global sectors and sub-sectors signifies that beneath the surface Chinese growth is weakening. Meanwhile, global tech stocks have not yet underperformed much (Chart I-3, bottom panel), implying the selloff has not been driven by this high-flying sector. The combination of weakening global trade amid still-robust U.S. domestic demand bodes well for the U.S. dollar, at least against EM and commodities currencies. U.S. and EU imports account for only 13% and 11% of global trade, respectively (Chart I-4). Meanwhile, aggregate EM including Chinese imports account for 30% of world imports. Hence, global trade can slow even with U.S. and EU domestic demand remaining robust. We addressed the twin deficit issue in the U.S. in our February 21 report,2 and will add the following: If U.S. fiscal stimulus coincides with abundant global growth, the greenback will weaken. If on the contrary, the U.S. fiscal expansion overlaps with weakening global trade, U.S. growth will be priced at a premium and the U.S. dollar will appreciate especially against the currencies of economies where growth will fall short. The majority of EM exchange rates will likely be in the latter group. The relative performance of EM versus DM stocks correlates with the relative volume of imports between China and the DM (Chart I-5). The rationale is that EM countries and their publically listed companies are much more leveraged to China's business cycle than DM. The opposite is true for DM-listed companies. Our view is that China's industrial recovery and growth outperformance versus DM since early 2016 is about to end. This, if realized, should undermine EM equities and currencies versus their DM counterparts. Last week, we published a Special Report on the Chinese real estate market.3 We documented that despite a drawdown in housing inventories over the past two years, both residential and non-residential inventories remain very elevated. This, along with poor affordability and the implementation housing purchase restrictions for investors, will dampen housing sales, which in turn will lead to a contraction in property development and construction activity. Chart I-4Global Trade Is More Leveraged To EM Not DM Chart I-5EM Underperforms When Chinese Imports Lag DM Ones Combined with a slowdown in infrastructure investment due to tighter controls on local government finances, this poses downside risks to China's demand for commodities, materials and industrial goods. This is the main risk to EM stocks and currencies, and the primary reason we continue to maintain our negative stance on EM risk assets. Last but not least, it is widely believed that Chinese households are not indebted and that there is a lot of pent-up demand for household credit. Chart I-6 reveals that this conjecture is simply not true - the household debt-to-disposable income ratio has surged to 110% of disposable income in China. The same ratio is currently 107% in the U.S. Given borrowing costs in general and mortgage rates in particular are higher in China than in the U.S. (the mortgage rate is 5.2% in China versus 4.4% in the U.S.), interest payments on debt account for a larger share of households' disposable income in China than in America right now. In the U.S., the surprise on the macro front in the coming months will likely be both rising wage growth and core inflation. Chart I-7 highlights that average hourly earnings in manufacturing and construction have been accelerating. This underscores that wages are rising fast in these cyclical sectors. This will spread to other sectors sooner rather than later. Core inflation in America is rising and has already moved above 2% (Chart I-8). The rise is broad-based as all different core consumer price measures are rising and heading toward 2%. Chart I-6Chinese Households Are As Leveraged As Americans Chart I-7U.S. Wages Are Accelerating Chart I-8U.S. Core Inflation Is Above 2% While this does not entail that the U.S. is heading into runaway inflation, rising core inflation and wage growth will likely lead many investors to believe that the Federal Reserve cannot back off too fast from rate hikes, particularly when the U.S. fiscal thrust remains so positive, even if the drawdown in share prices persist. This may especially weigh on EM risk assets, where growth will be subsiding due to their links with Chinese imports. Bottom Line: Our main macro themes remain a slowdown in China and a rise in U.S. core inflation. This combination bodes ill for EM financial markets. Continue underweighting EM stocks, credit and currencies versus their DM peers. Upgrade Mexican Equities To Overweight In our March 29 report,4 we upgraded our stance on the Mexican peso, local currency bonds and U.S. dollar sovereign credit from neutral to overweight. The main rationale was receding odds of NAFTA abrogation and the country's healthy macro fundamentals. In addition, we instituted a new currency trade: long MXN / short BRL and ZAR. Continuing with this theme, we today recommend upgrading Mexican stocks to overweight within an EM equity portfolio: The odds of NAFTA retraction are rapidly subsiding as the U.S. is shifting its focus to China. Hence, chances are that NAFTA negotiations will be completed this summer, and a deal will be signed off before Mexico's presidential elections on July 1st. A more benign outcome together with an early end to NAFTA negotiations will reduce uncertainty and the risk premium priced into Mexican financial markets. This will help the latter outperform their EM peers. A final note on Mexican politics: The leftist presidential candidate Andres Manuel Lopez Obrador has high chances of winning the presidential elections in July. Yet Our colleagues at BCA's Geopolitical Strategy service believe political risks are overstated.5 The basis is that Obrador will balance the left-leaning preferences of his electorate with the prudent policies needed to produce robust growth. While political uncertainty in Mexico is subsiding, it is rising in many other EM countries such as Russia, China and Brazil. In brief, geopolitical dynamics favor Mexico versus the rest of EM. We expect dedicated EM managers across various asset classes to rotate into Mexico from other EM countries. We outlined two weeks ago that a stable exchange rate will bring down inflation, opening a door for the central bank to cut interest rates no later than this summer. As local interest rate expectations in Mexico continue to subside both in absolute terms as well as relative to EM, Mexican share prices will outpace their EM peers (Chart I-9). Consistently, tightening Mexican sovereign credit spreads versus EM overall should also foster this nation's equity outperformance (Chart I-10). Chart I-9Relative Equity Performance Tracks Relative ##br##Local Bond Yields Chart I-10Relative Equity Performance Tracks Relative ##br##Sovereign Spreads Domestic demand growth has plunged following monetary and fiscal tightening in the past two years (Chart I-11). As both fiscal and monetary policy begin to ease, domestic demand will recover later this year. Chances are that share prices will sniff this out and begin their advance/outperformance sooner than later. Consumer staples and telecom stocks together account for 50% of the MSCI Mexico market cap, while the same sectors make up only 11% of overall EM market cap. Hence, Mexico's relative equity performance is somewhat hinged on the outlook for these two sectors in general and consumer staples in particular. EM consumer staple stocks have massively underperformed the EM benchmark since early 2016 (Chart I-12, top panel), and odds are this sector will outperform in the next six to 12 months as defensive sectors outperform cyclicals. This in turn heralds Mexico's relative outperformance versus the EM benchmark, which seems to be forming a major bottom (Chart I-12, bottom panel). Chart I-11Mexico: Economic Downturn Is Well Advanced Chart I-12Mexican Bourse Is A Play On Consumer Staples Unlike many EM countries, the Mexican economy is much more leveraged to the U.S. than to China. One of our major themes remains favoring U.S. growth plays versus Chinese ones. Finally, Mexican equity valuations have improved quite a bit both in absolute terms and relative to EM. Chart I-13 shows our in-house CAPE ratios for Mexican stocks in absolute terms and relative to the EM overall benchmark: Mexican equity valuations are not cheap but they are no longer expensive. Consistent with upgrading our economic outlook on Mexico, fixed-income investors should bet on yield curve steepening in local rates. We initiated this strategy on January 31 but hedged the NAFTA risk by complementing it with a yield curve flattening leg in Canada. Now, we are closing that trade and initiating a new one: fixed-income traders should consider paying 10-year swap rates and receiving 2-year swap rates. The yield curve is as flat as it typically gets (Chart I-14, top panel). Moreover, 2-year swap rates are not yet pricing enough rate cuts (Chart I-14, bottom panel) but will soon begin gapping down pricing in a large (potentially close to 200 basis points) rate cut cycle. Chart I-13Mexican Equities Are No Longer Expensive Chart I-14Bet On Yield Curve Steepening In Mexico Bottom Line: In line with our recent upgrade of Mexican local and U.S. dollar bonds as well as the currency outlook versus their EM peers, this week we recommend EM dedicated equity portfolios shift to an overweight position in Mexican stocks. Fixed-income trades should bet on yield curve steepening by paying 10-year swap rates and receiving 2-year rates. Investors who are positive on global risk assets should consider buying Mexican local bonds outright. Russia: Geopolitics Trumps Economics Chart I-15Russian Assets Relative To EM Benchmarks:##br## Various Asset Classes The sudden crash in Russian financial markets this week following the imposition of new U.S. sanctions has reminded us that geopolitics can often eclipse economics. Our overweight recommendation on Russian assets versus their EM peers was based on two pillars: (1) healthy and improving macro fundamentals and an unfolding cyclical economic recovery; and (2) easing tensions between Russia and the West. Clearly, the second part of our assessment is wrong, or at least premature. While BCA's Geopolitical Service team maintains that on a 12-month horizon tensions between Russia and the West will subside, the near-term risks are impossible to assess. For this reason we are closing our overweight allocation in Russian financial markets and recommend downgrading it to neutral. In particular, we are shifting Russia to a neutral allocation within the EM equity, sovereign and corporate credit and local currency bonds portfolios (Chart I-15). Consistently, we are closing the following trades: Long Russian / short Malaysian stocks (27.6% gain); Long Russian energy / short global energy stocks (2.8% gain); Long RUB / short MYR (3.1% loss); Short COP / long basket of USD & RUB (16.2% loss); Long RUBUSD / short crude oil (29.1% loss). Sell Russian 5-year CDS / buy South African 5-year CDS (317 basis points gain); Long Russian and Chilean / short Chinese Corporate Credit (12% gain); Long Russian 5-year bonds / short Brazilian 5-year bonds (flat). Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com 1 We have removed the Russian ruble from the version of this chart shown in March 29, 2018 EMS report to assure that the recent idiosyncratic developments - the selloff triggered by the U.S. sanctions - in Russia's financial markets do not impact the reading of this indicator. 2 Pease see Emerging Markets Strategy Weekly Report "EM Local Bonds And U.S. Twin Deficits", dated February 21, 2018, Page 14. 3 Pease see Emerging Markets Strategy Weekly Report "China Real Estate: A Never-Bursting Bubble?", dated April 6, 2018, Page 14. 4 Pease see Emerging Markets Strategy Weekly Report "EM: Perched On An Icy Cliff", dated March 29, 2018, available at ems.bcaresearch.com. 5 Pease see Geopolitcial Strategy Weekly Report "Expect Volatility... Of Volatility", dated April 11, 2018, available at gps.bcaresearch.com. Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights Capacity cuts in China's steel and aluminum industries over the winter produced little in the way of output reductions, confounding our expectations. The resulting unintended inventory accumulation in Asian markets, reflecting high production relative to demand, and slowing Chinese steel exports are a downside risk to our neutral view. U.S. sanctions against Russian oligarchs close to President Putin could tighten the aluminum market, countering the unintended inventory accumulations. For now, we remain neutral base metals. Energy: Overweight. We are closing our long put spread position in Dec/18 Brent options at tonight's close. The fast-approaching May 12 deadline for President Trump to renew sanctions waivers against Iran shifts the balance of price risks to the upside. Base Metals: Neutral. COMEX copper rallied above $3.10/lb on the back of Chinese President Xi's remarks at the Boao Forum earlier this week, which re-hashed plans to open China's economy to imports. Precious Metals: Neutral. Gold likely becomes better bid as the May 12 deadline to waive Iran sanctions nears. Our long gold portfolio hedge is up 8.9%. Ags/Softs: Underweight. European buyers are scooping up U.S. soybeans, as Chinese purchases of Brazilian beans makes U.S.-sourced crops relatively cheaper, according to Reuters.1 China also announced plans to start selling corn stocks from state reserves this week, offering an alternative protein for animals to partially offset the price impact of tariffs on their imports of U.S. soybeans. Feature Chart of the WeekAluminum Rebounds On U.S. Sanctions Despite much-ballyhooed capacity reductions in China's steel and aluminum capacity, these markets - both in China and globally - remained relatively well supplied over the winter. Higher global supplies, and falling Chinese steel exports, will result in unintended inventory accumulation, which already is showing up in Shanghai Futures Exchange (SHFE) inventories. While we remain neutral base metals, continued unintended inventory accumulation could cause us to downgrade the sector. The MySteel Composite Index we use to track steel prices is down more than 10% since the beginning of the year (Chart of the Week). Similarly, the first-nearby primary aluminum contract on the LME was down ~ 12% year-to-date (ytd) early last week, before regaining most of these losses on news of U.S. sanctions against Russian oligarchs, which hit shares of Rusal very hard. Given that these sanctions will restrict access to up to 6% of global aluminum supply, ex-China supply dynamics will dominate the aluminum market this year making the outlook relatively favorable, putting a floor beneath the London Metal Exchange Index (LMEX).2 Ex-Post Winter Production Production cuts over the winter - when Chinese mills in 28 smog-prone northern cities were ordered to reduce capacity by up to 50% - did not live up to our expectations.3 China's steel and aluminum sectors have undergone major supply-side reforms, particularly re the removal of outdated capacity, most of which has been completed. In addition to the winter capacity cuts, past reforms that have already been implemented, and have shaped current market conditions, are as follows: In an effort to eliminate outdated and unlicensed facilities, China removed an estimated 3-4 mm MT of annual capacity in 2017 - amounting to approximately 10% of total aluminum smelting capacity. In the case of steel, Beijing announced plans to shut down 150 mm MT of annual steel capacity between 2016 and 2020. To date, 115 mm MT of capacity have already been eliminated. Another estimated 80-120 mm MT of induction furnace capacity was shuttered in 1H17. Going forward, China's steel and aluminum markets will be driven by: An estimated 3-4 mm MT of updated aluminum capacity is expected to come on line this year, offsetting constraints from last year's supply cuts. 30 mm MT of steel capacity shutdowns are planned this year, putting Beijing on track to meet its five-year target two years ahead of schedule. The Chinese National Development and Reform Commission (NDRC) has communicated its resolve to keep shuttered capacity offline. Major steelmaking cities in Hebei province - accounting for 22% of 2017 Chinese crude steel output - have announced plans to extend the capacity cuts to November 2018. The mid-November to mid-March capacity cuts implemented this past season are expected to be a recurring event. Winter Shutdowns Minimally Impact China's Steel Output ... According to steel production data released by the World Steel Association (WSA), winter capacity closures in China did not significantly affect overall output levels. Crude steel output from China was up 3.9% year-on-year (y/y) in the November to February period (Chart 2). At the same time, production from the rest of the world increased by 3.6% y/y in the November to February. Thus global crude steel supply remained in excess over the winter season, as global steel output increased 3.8% y/y. A caveat to these data: China does not account for the historical output of induction furnaces, which produced an estimated ~30-50 mm MT of steel in 2016. As mentioned in our previous research, the output of these furnaces was illegal and thus not carried in statistics we use to track supply.4 These data problems mean it is possible that actual output in the November 2016 to February 2017 period was higher than suggested by the data, and as a result, actual output during this year's winter season may actually be lower than last year. As induction-furnace data lie in the statistical shadows, we cannot ascertain this with certainty. Nevertheless, a buildup in China inventories - which we discuss below - indicates an oversupplied market. It is also likely producers - incentivized by high steel prices earlier this year - kept capacity utilization at maximum levels throughout the winter. ... And Aluminum Output According to International Aluminum Institute data, primary aluminum output in China fell 2.3% y/y in the November to February period, suggesting the winter cuts likely had an impact on aluminum supply (Chart 3). Data from the World Bureau of Metal Statistics (WBMS) show an even sharper decline in winter aluminum output: primary production in China fell 8.7% y/y in the November to January period. Chart 2Steel Output Grew##BR##Amid Winter Cuts Chart 3China Aluminum Market In Surplus##BR##Despite Production Decline Both sources reveal an especially pronounced contraction in November, at the onset of the winter cuts. Despite reduced supply, WBMS data indicate a positive Chinese aluminum market balance throughout the winter. A large contraction in demand offset the supply shortfall, and kept primary aluminum in a physical surplus throughout the winter, ultimately leading to a buildup in domestic inventories. A Look At The Trade Data Despite our disappointment regarding the impact of the winter cuts on steel and aluminum markets, trade data increasingly suggests China's steel exports have peaked. Aluminum exports from China, on the other hand, are likely to continue rising. Chinese Steel Exports Continue To Fall ... Chinese steel product net exports have been falling since mid-2016, and have continued falling in y/y terms throughout the winter. According to Chinese customs data, steel product net exports fell 35.1% y/y in the November to February period, driven by both falling exports as well as rising imports (Chart 4). Steel product exports plunged 30% y/y in the November to February period, more or less in line with the 2017 average. The decline mirrors the 2017 contraction in domestic supply, bringing exports to their lowest level since 2012. This indicates fears of a China slowdown leading to a flood of metal onto global markets have not materialized, at least not yet. In fact, Customs data show a 1.7% y/y increase in Chinese steel imports during the November to February period - a reversal from falling imports prior to the winter season. The conclusion we draw from this is that, while in the past, China was a source of supply for the world, ongoing capacity cuts and production controls could mean China will lack the ability to ramp up output in case of a global physical supply deficit. If this becomes the new normal, price volatility will likely increase. This trend is important, especially given our expectation of strong world ex-China demand this year. As such, global steel prices may find support amid this new normal. ... But Aluminum Exports Move Higher In the case of aluminum, Chinese net exports were up 28.7% y/y during the winter, continuing their upward trend. Customs data show a 14.8% y/y increase in aluminum exports in November to February, bringing exports in this period to their highest level since 2014/15 (Chart 5). At the same time, imports of aluminum have come down during this period - by 37.2% y/y. According to China customs data, 2017 imports over these winter months registered their lowest level since 1994. Chart 4Steel Exports Continue Falling ... Chart 5...While Aluminum Exports Are On the Uptrend The combination of growing exports amid falling imports puts China's net exports in expansionary territory. This will be especially true given the planned increase in capacity this year amid weak Chinese demand. All in all, ceteris paribus global supply of aluminum looks set to increase. However, we do not live in a ceteris paribus world and, as we explore below, sanctions against the top aluminum producer outside of China will have massive implications on the global aluminum supply chain. Are Inventories Due For A Turnaround? Chart 6Larger Than Expected##BR##Seasonal Inventory Buildup China Iron and Steel Association data indicate that since the beginning of the year, steel product inventories have been re-stocked to levels last seen in 1Q14. Inventories of the five main steel products we track have more than doubled since the beginning of the year (Chart 6). Although the Q1 build is seasonal, the re-stocking since the beginning of the year has been especially pronounced. This buildup occurred in an environment of stable supply - with minimal impact from the winter capacity cuts - amid weak exports, indicating domestic demand for the metal was subdued. However, steel inventories have turned around, and we expect further destocking as demand accelerates post the Chinese New Year. The question remains whether this destocking will bring inventories back down to their 5-year average. Aluminum inventories on the SHFE show similar dynamics. However in this case, it is part of the larger trend of rising stocks since the beginning of last year. Aluminum inventories at SHFE warehouses are up more than nine-fold - or 0.87 mm MT - since the end of 2016. In fact, the pace of buildup seems to have accelerated: the average weekly build of 16.6k MT of aluminum coming into warehouse inventories since the beginning of the year stands above the 2017 average weekly build of 12.6k MT. This brought SHFE aluminum inventories to almost 1 mm MT, more than double their previous record in 2010. Although the Chinese physical aluminum surplus weighed down on prices in 1Q18, we expect global aluminum prices to remain supported from here due to the impact of U.S. sanctions on world ex-China aluminum supply. U.S. Russian Sanctions Could Be A Game-Changer Chart 7Sanctions Will Restrict##BR##Marketable Aluminum Supply Last Friday, the U.S. announced sanctions on Russian oligarchs close to President Vladimir Putin. Among those sanctioned is Oleg Deripaska who controls EN+ Group, which owns a controlling interest in top aluminum producer United Company Rusal. Given that UC Rusal accounts for ~6% of global aluminum production, we view this move as significant to global aluminum markets. As the top producer of the metal outside China, Rusal aluminum likely makes up the majority of Russian supply, which account for 14% of U.S. imports (Chart 7). In fact, almost 15% of Rusal's revenues comes from its business with the U.S. While it is clear that these sanctions will, in effect, terminate aluminum trade between Russia and the U.S., more significant are the implications on the global supply chain. A clause in the U.S. Treasury Department's order extending the restrictions to non-U.S. citizens dealing with U.S. entities means the impact could be far-reaching, requiring a major re-shuffle in global aluminum trade. Earlier this week, the LME announced that it will no longer accept Rusal aluminum produced after April 6, effectively preventing the company's products from being delivered on the LME. These sanctions will likely turn global aluminum buyers off from Rusal products, as they can no longer deliver it to the LME. The net effect will be a contraction in global usable aluminum supply. Furthermore, these sanctions will likely disrupt supply chains as aluminum users scramble to avoid purchasing metal from the Russian producer. While the details of these restrictions are still unclear, the sanctions are a game changer in the global aluminum market - effectively restricting access to a major source of the metal. As such, primary aluminum on the LME is up more than 10% since the announcement last Friday. Bottom Line: While China's crude steel output increased y/y during government-mandated output cuts over the winter, seasonally weak demand meant that the metal piled up in inventories. Falling exports indicates that at least for now, the domestic surplus is not flooding global markets. The main risk to our neutral view here is that demand in China remains weak, and that this will lead to the offloading of Chinese metal to global markets, i.e. a pickup in exports. This has not yet materialized, so we are holding on to our neutral view for now. China's primary aluminum production declined y/y during the winter cuts. However the decline in domestic demand was greater - likely due to the decline in auto production and sales following the loss of tax credit incentives. Consequently, China's aluminum market remained in surplus throughout the winter. Some of the excess supply was exported, but SHFE inventories continued building. Our outlook on the aluminum market had been bearish, due to additional capacity coming online this year amid an uncertain China demand environment. However, the sanctions on Rusal could be a game changer, putting a floor beneath aluminum prices. This improves our near term outlook for the aluminum market. This makes our outlook on aluminum prices much more favorable. Roukaya Ibrahim, Associate Editor Commodity & Energy Strategy RoukayaI@bcaresearch.com 1 Please see "As U.S. and China trade tariff barbs, others scoop up U.S. soybeans," published by reuters.com on April 8, 2018. 2 The six non-ferrous metals represented in the LMEX and their respective weights are as follows: aluminum: 42.8%, copper: 31.2%, zinc: 14.8%, lead: 8.2%, nickel: 2.0%, and tin: 1.0%. 3 China's winter smog "battle plan" targeted polluting industries in the northern China region by mandating cuts on steel, cement and aluminum production during the smog-prone mid-November to mid-March months. Steel and aluminum production cuts targeted a range between 30-50% during this period. This event is expected to be an annually recurring event until 2020. 4 Please see BCA Research's Commodity & Energy Strategy Weekly Report titled "China's Environmental Reforms Drive Steel & Iron Ore," dated January 11, 2018, available at ces.bcaresearch.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed in 2018 Summary of Trades Closed in 2017

Highlights There is more downside risk ahead as the geopolitical calendar is packed in May; Protectionism remains in play, but markets could also fall on Iran-U.S. tensions, military intervention in Syria, and Russia-West confrontation; Investors should expect volatility to go up as we approach a turbulent summer; We were wrong on Russia-West tensions peaking and are closing all of our Russian trades for now, but may look for new entry points soon; Go long a basket of NAFTA currencies versus the Euro and expect reflation to remain the "only game in town" in Japan. Feature "I'm not saying there won't be a little pain, but the market has gone up 40 percent, 42 percent so we might lose a little bit of it. But we're going to have a much stronger country when we're finished. So we may take a hit and you know what, ultimately we're going to be much stronger for it." President Donald Trump, April 6, 2018 Chart 1Teflon Trump There are times when conventional wisdom is spectacularly wrong. Last week was such a moment. Since Donald Trump became president, the "smart money" has believed that he was obsessed with the stock market. Therefore, the view went, none of his policies would threaten the bull market. We have pushed back against this assumption because our view is that geopolitical risks - specifically the lack of constraints on the executive branch in foreign and trade policy - would become investment relevant.1 This view has been correct thus far: we called the volatility spike and trade protectionism in 2018. Not only have President Trump's tariff pronouncements produced stock market drawdowns, but his popularity appears to be unaffected. Astonishingly, President Trump's approval rating collapsed as the stock market went up in 2017 and recovered as the stock market went in reverse this year (Chart 1)! It is therefore empirically incorrect that President Trump is constrained by the stock market. His actions over the past month, as well as his approval ratings, suggest that he is quite comfortable with volatility. There are two broad reasons why we never bought into the media hype. First, there is no real correlation, or only a weak one, between equity declines of 10% and presidential approval ratings (Chart 2). Generally, presidential approval rating does decline amidst market drawdowns of 10% or greater, but the effect on the presidency is only permanent if the momentum of the approval rating was already heading lower, otherwise the effect is minimal and temporary. Second, the median American does not really own stocks (Table 1). President Trump considers blue collar white voters his base and they care more about unemployment and wages, not their equity portfolios. At some point, equity market drawdowns will affect hard data and the real economy. This is the point at which President Trump will care about the stock market. Given that the market is already down 10% from the peak, we are not far away from this pain threshold. But in this way, President Trump is no different from any other president. Chart 2AThe Stock Market Mattered For Eisenhower, JFK, Bush Sr., And Obama... Chart 2B...But Not For Johnson, Nixon, Ford, Carter, Reagan, And Bush Jr. The pessimistic view on trade protectionism risk, that there is more downside to equities ahead, is therefore still in play. Investors should be careful not to overreact to positive developments, such as President Xi's speech at the Boao Forum where he largely reiterated previous Beijing promises to open up individual sectors to foreign investment. In fact, it is the investment community itself that is the target of President Trump's rhetoric. In order to convince Beijing that his threat of protectionism is credible, President Trump has to show that he is willing to incur pain at home, which explains the quote with which we began this report. Table 1Stock Ownership Is Concentrated Amongst The Wealthiest Households This is not dissimilar to President Trump's doctrine of "maximum pressure" which, when applied to North Korea, produced a significant bond rally last summer. The 10-year Treasury yield topped 2.39% on July 7 and then collapsed to a low of 2.05% in September.2 The vast majority of the yield decline, at the time, came from falling real yields as investors flocked into safe-haven assets amidst North Korean tensions and not lower inflation expectations. It is therefore dangerous to rely on conventional wisdom when assessing the limits of volatility or equity drawdowns. Any buoyant market reaction may in fact elicit a more aggressive policy from Washington. As if on cue, President Trump shocked the markets on April 7 by suggesting that he would impose another round of tariffs on a further $100bn worth of Chinese imports, bringing the total under threat to $160 billion. The announcement came after the market closed 0.89% up on April 6. Perhaps President Trump was irked that the market was so dismissive of his trade threats and decided to jolt it back to reality. In addition to trade, there are several other reasons to be bearish on risk assets as we approach May: Chart 3Inflation Will Pick Up In 2018 Chart 4Service Sector Wage Growth Is At A Cyclical Peak Inflation: Unemployment is low, with wage pressures starting to build (Chart 3). Meanwhile, teacher strikes in Red States like Oklahoma, Kentucky, West Virginia, and Arizona are signalling that public service sector wage pressures are building in the most fiscally prudent states. Service sector wages cannot be suppressed through automation or outsourcing and are therefore likely to add to inflationary pressures (Chart 4). The Fed remains in tightening mode, despite the mounting geopolitical risks. "Stroke of pen risk:" Another sign that President Trump is comfortable with market drawdowns is his increasingly aggressive rhetoric on Amazon. There is a rising probability that the current administration decides to up the regulatory pressure on the technology and retail giant, as well as a possibility that other technology companies like Facebook and Google face "stroke of pen" risks. Iran: This year's premier geopolitical risk is the potential for renewed U.S.-Iran tensions.3 Ahead of the all-important May 12 deadline - when the White House will decide whether to end the current waiver of economic sanctions against Iran - President Trump has staffed his cabinet with two hawks, new Secretary of State Mike Pompeo and National Security Advisor John Bolton. Meanwhile, tensions in Syria are building with potential for U.S. and Iranian forces to be directly implicated in a skirmish. The U.S. is almost certain to militarily respond to the alleged chemical attack by the Syrian government forces against the rebel-held Damascus suburb of Douma. Throughout it all, investors appear to remain unfazed by the rising probability that Iran's 2 million barrels of oil exports come under renewed sanction risk, mainly because the media is ignoring the risk (Chart 5). Chart 5The Media Is Ignoring Iran As A Risk Russia: As we discuss below, tensions between the West and Russia appear to be building up anew. Particularly concerning is the aforementioned chemical attack in Syria, which Moscow considers a "false flag operation." The Russian government hinted in mid-March that precisely such an attack may occur and that the U.S. would use it as a pretext to attack Syrian government forces and structures.4 Our view that tensions have peaked, elucidated in a recent report, therefore appears to have been spectacularly wrong. Chinese reforms: Now that Xi Jinping has finished setting up his new government, his initiatives are starting to be implemented. While some slight tax cuts are on the docket, and interbank rates have eased significantly, there is no sign of broad policy easing or economic recovery (Chart 6). Rather, both Xi and his economic czar Liu He have continued to stress the "Three Battles" of systemic financial risk, pollution, and poverty - the first two requiring tighter policy. Xi has stated that deleveraging will focus on state-owned enterprises (SOEs) and local governments. SOEs will have debt caps and will not be allowed to lend to local governments. Instead, local governments will have to borrow through formal bond markets, giving the central government greater control. Meanwhile, the Ministry of Housing says property restrictions will remain in place. All in all, the risk of negative surprises in China this year remains significant, with a likely negative impact on global growth.5 There is also a fundamental reason for equity market weakness: the market is likely coming to grips with a calendar 2019 EPS growth of a more reasonable 10% annual rate compared with this year's near 20% peak growth rate. This transition, which our colleague Anastasios Avgeriou of BCA's U.S. Equity Strategy has highlighted in recent research, will be turbulent.6 In addition, Anastasios has pointed out that stocks are reacting to a more bearish mix of soft and hard data (Chart 7), suggesting that not all of the market volatility is due to headline risk. Chart 6China Will Slow Down Further In 2018 Chart 7Trade Is Not The Only Risk To The Market How should investors make sense of these budding risks? Going forward, we would fade any enthusiasm or narratives of "peak pessimism" on trade protectionism. It is in the interest of the Trump administration that investors take his threats seriously. President Trump literally needs stocks to go down in order to show Beijing that he is serious. The summer months could be volatile as market confusion grows amidst the upcoming event risk (Table 2). This may be a good time to be risk-averse, with the old adage "sell in May and go away" appropriate this year. Table 2Protectionism: Upcoming Dates To Watch There are several reasons why protectionism is a much bigger deal than it was in the 1980s when investors last had to price a trade war between two major economies (Japan and the U.S. at the time): Chart 8This Time Is Different... Because Of Supply Chains... Chart 9...Globalization... Supply chains are a much bigger deal today than thirty years ago (Chart 8); The share of global exports as a percent of GDP is much higher today (Chart 9); Interest rates are much lower, leaving little room for policymakers to ease (Chart 10); Stock market valuations are higher, leaving stocks exposed to drawbacks (Chart 11); Unlike 1981-88, when Japan and the U.S. waged a nearly decade-long trade war while remaining allies in the Cold War, China and the U.S. are outright rivals. This increases the probability that Beijing's reprisal, given its constraints in retaliating against U.S. exports (Chart 12), could take a geopolitical turn. Chart 10...Policymaker Ammunition... Chart 11...And Valuations Chart 12China May Run Out Of U.S. Exports To Sanction Investors should therefore prepare for volatility of volatility. Amidst the confusion, there could be some not-so-positive news that the market overreacts to with optimism, and some not-so-negative news that the market reacts to with pessimism. In our six years of publishing geopolitically driven investment strategy, we have not seen a similar period where a confluence of risks and tensions are building up at the same time. May should therefore be a busy month. Mexico: A Silver Lining Amidst Mercantilism Risk? Mexico began the year with clouds over its head due to the Trump team's tough negotiating line on NAFTA. The third round of negotiations, in September 2017, ended on a bad note. The peso tumbled and headline and core inflation soared, portending both tighter monetary policy and weaker domestic demand.7 Today, however, the odds of renewing NAFTA have improved significantly. We have reduced our probability of Trump abrogating the trade deal from 50% to 20%. The administration appears to be focused on China and therefore looking to wrap up the NAFTA negotiations quickly over the summer. This would give time to send the new deal to the Mexican and U.S. congresses prior to the September changeover in Mexico's legislature and January changeover in the U.S. legislature. The U.S. has reportedly compromised on an earlier demand that NAFTA-traded automobiles have a U.S. domestic content of 50%.8 Meanwhile, inflation has peaked and the peso has firmed up (Chart 13), which will help buoy real incomes and boost purchasing power. Economic policy has been prudent, with central bank rate hikes restraining inflation and government spending cuts producing a primary budget surplus (and a much-reduced headline budget deficit of -1% of GDP) (Chart 14).9 Chart 13Mexico: Peso & Inflation Chart 14Mexico: Improved Macro Fundamentals In this more bullish context, the Mexican elections on July 1 are market-neutral. True, it is hard to present a strong pro-market outcome. The public is shifting to the left on the economic spectrum while the outgoing "pro-market" administration of Enrique Pena Nieto has lost credibility. The latest polling suggests that Andres Manuel Lopez Obrador (AMLO) is polling in the lower 30-percentile (around 33%), above his next competitors, Ricardo Anaya (PAN) at 26% and Jose Antonio Meade (PRI) at 14% (Chart 15). However, the latest data point of the admittedly volatile polling gives AMLO a much less commanding lead of 6-7% over Anaya than he had before. AMLO is polling around his performance in the 2006 and 2012 elections (35% and 32%, respectively), has increased his lead over the other candidates, and his National Regeneration Movement (MORENA) and "Together We'll Make History" coalition are also polling with double-digit leads (Chart 16). The general shift to the left is also apparent in the fact that Ricardo Anaya's PAN has been forced to combine with the left-wing PRD in order to garner votes. Chart 15AMLO's Lead Is Not Insurmountable Chart 16Likely No Majority In Congress Nevertheless, political risk is overstated for the following reasons: AMLO is not Hugo Chavez:10 True, he is a leftist, a populist, and has a reputation for egotism. He is Mexico's fitting anti-Trump. Nevertheless, he is also a known quantity, having run for president and engaged with the major parties for over a decade. While he elevates headline political risk, we would fade the risk based on the fact that Mexico is a relatively right-wing country (Chart 17), and his movement will probably not garner a majority in Congress (see next bullet). Notably, AMLO's rhetoric on Trump and NAFTA has been restrained, and his personnel decisions have been competent and orthodox. He has not suggested he will revoke new private Mexican oil concessions, under the outgoing government's privatization scheme, but only halt the auctions. AMLO will be constrained by Congress: The trend in Mexico is towards "pluralization" or fragmentation in Congress (see Chart 18), meaning that ruling parties will have to share power. This is not a negative development. As we recently pointed out, political plurality engenders stability by drawing protest parties into centrist coalitions and by allowing establishment parties to coopt protest narratives without having to actually protest or revolt.11 At this point in time, it is difficult to see how AMLO's MORENA garners enough support to get a majority in Congress. AMLO's closest challenger is right-wing and pro-market: If AMLO loses the election, Ricardo Anaya of PAN will not be scorned by financial markets. In 2006, AMLO looked like he would win the election but then lost to Felipe Calderon (PAN). Of course, a victory by Anaya is not very market positive either, as PAN is in an unstable coalition with the left-wing PRD and would also be constrained in Congress. Still, there would be a lower probability of reversing the outgoing PRI administration's policies than under AMLO. AMLO is unlikely to repeal NAFTA: Mexico's exports to NAFTA partners comprise 30% of GDP, and it would be exceedingly dangerous for a Mexican leader to provoke Trump on the issue. A plurality of the Mexican public (44%) supports the ongoing NAFTA negotiations as they have been handled by the current government (Chart 19), as of late February polling by the Wilson Center. The same polling shows that Mexicans are generally aware of how important NAFTA is for their economy. This is despite the polls showing that a majority of Mexicans have a negative view of the U.S., due largely to Trump's rhetoric (though that majority has fallen considerably since last year to 56%). In other words, anti-American sentiment is not turning the Mexican public against compromising on a new NAFTA deal. Chart 17Mexicans Lean Right Chart 18Mexico's Rising Political Plurality Finally, Mexico is more exposed to U.S. growth (which is charged with fiscal stimulus), and to BCA's robust outlook on oil prices (as opposed to our weaker metals outlook), while it is less exposed to weakening Chinese demand than other EMs (such as South Africa or Brazil).12 The peso looks particularly attractive relative to the latter two currencies (Chart 20). Chart 19Mexicans Want NAFTA To Survive Chart 20A Major Bottom In MXN's Cross? None of the above should suggest that the Mexican election will be a smooth affair. The rise of AMLO will create jitters in the marketplace, particularly as he faces off against Trump, who will continue to try to pressure Mexico over immigration and border security even once NAFTA negotiations are squared away. Nevertheless, the cyclical backdrop has improved while the major headwind of NAFTA abrogation seems to be abating. Bottom Line: Mexico's presidential campaign, election, and aftermath will give rise to plenty of occasion for volatility, particularly as President Trump and a likely President Obrador will not shy from a war of words. Nevertheless, Mexico's economic policy is stable and the NAFTA headwind is abating. We recommend going long Mexican local currency bonds relative to the EM benchmark. We also recommend that clients go long a NAFTA basket of currencies - the peso and the loonie - versus the euro. Our currency strategist - Mathieu Savary - has recently pointed out that the euro has moved ahead of long-term fundamentals and is ripe for a near-term correction.13 Japan: Abe Will Survive Japanese Prime Minister Shinzo Abe has come under rising public criticism in recent that is dragging down his approval ratings (Chart 21). Three separate scandals are weighing on his administration: one relating to the government's sale of land at knockdown prices to a nationalist school, Moritomo Gakuen, tied to Abe's wife; another relating to the discovery of "lost" journals of Japan Self-Defense Force activity during the Iraq war; another tied to the mishandling of statistics in promoting the government's new revisions to the labor law. Abe's popularity has tested lower lows in the past, but he is approaching the floor. And while Abe is still polling in line with the popular Prime Minister Junichiro Koizumi at this stage in his term (Chart 22), nevertheless he is approaching his 65th month in office when Koizumi stepped down. Chart 21Abe's Approval Testing The Floor Chart 22Abe Holding At Koizumi's Levels Of Support More importantly, the all-important September leadership election is approaching. The challenges arising today are at least partly motivated by factions within the LDP that want to challenge Abe's leadership. Koizumi stepped aside in September 2006 because he could not contend for the LDP's leadership due to party rules that limited the leader to two consecutive three-year terms. Abe is not constrained on this front. He has already revised those rules to three terms, giving him until September 2021 to remain eligible as party leader. He wants to run again and incumbents are heavily favored in party elections. Abe also secured his second two-thirds supermajority in the House of Representatives, in October 2017. This was a remarkable feat and one that will make it difficult for contenders to convince the rank and file in Japan's prefectures that they can lead the party more effectively. While Abe's 38% approval is now slightly below the psychologically important 40% level, and below the LDP's overall approval rating (Chart 23), there is no alternative to the LDP heading into July 2019 elections for the House of Councillors. This is manifest from the October election result. Chart 23Still No Alternative To LDP What happens if Abe's popularity sinks into the 20-percentile range? Financial markets will selloff in anticipation that he will be ousted. He could conceivably survive a scrape with the upper 20% approval range, but markets will assume the worst once he dips beneath 30% in the average polling on a sustainable basis. Markets will also assume that the remarkably reflationary period in Japanese economic policy is coming to an end. Even when Abe's successor forms a government, investors may believe that the best of the reflationary push is over. We think that the market would be wrong to doubt Japan's inflationary push. First, if Abe is ousted, the LDP will remain in power: it has until October 2021 before it faces another general election that could deprive it of government control. (A loss in the upper house election in 2019 can prevent it from passing constitutional changes but not from running the country.) This ensures that policy will be continuous in the transition and that any changes in trajectory will be a matter of degree, not kind. Second, the phenomenon of "Abenomics" is not only Abe's doing but the LDP's answer to its first shocking experience in the political wilderness, from 2009-12. This experience taught the LDP that it needed to adopt bolder policies. The result was dovish monetary policy under Haruhiko Kuroda, who just began his second five-year term on April 9 and whose faction has the majority on the monetary policy board. Looser fiscal policy was another consequence - and ultimately it came to pass.14 It will be hard for a new LDP leader to tighten policy. Factions that are criticizing Abe or Kuroda today will find it harder to phase out stimulus once they are in office. Abe's successor will, like him, have to try policies that boost corporate investment, wages, the fertility rate, immigration, social spending and military spending.15 Without such initiatives, Japan will sink back into a deflationary spiral. As for BoJ policy, over the next 18 months the biggest challenges are meeting the 2% inflation target while the yen is rising due to both China's slowdown and trade war risks.16 Tokyo is also ostensibly required to hike the consumption tax in October 2019. This is more than enough to convince Kuroda to stand pat for the time being.17 In the meantime, Abe's push to revise the constitution is a significant factor in encouraging persistently loose monetary and fiscal policy. The national referendum on the matter could be held along with the early 2019 local elections or the July 2019 upper house election. It will be hard to win 50%+ of the popular vote and nigh impossible if the economy is failing. What should investors look for to determine if Abe's downfall is imminent? In addition to Abe's approval rating we will watch to see if the ongoing scandal probes produce any direct link to Abe, or if top cabinet ministers are forced to resign (like Finance Minister Taro Aso or Defense Minister Itsunori Onodera). It will also be a telling sign if Abe's "work-style" reforms to liberalize the labor market, which have received cabinet approval, wither in the Diet due to lack of party discipline (not our baseline view).18 But even granting Abe's survival, we would expect that China's slowdown and the U.S.-China trade war will keep the yen well bid. We are sticking with our tactical long JPY/EUR trade, which is up 2.6% thus far. Bottom Line: Shinzo Abe is likely to be re-elected as LDP leader in September and to lead his party in the charge toward the 2019 upper house election and constitutional referendum. Should he fall into the 20% of popular approval, the markets should sell off. His leadership and alliances have been remarkably reflationary and the policy tailwind could dwindle. We would fade this risk, but we still think the yen will remain buoyant due to China's internal dynamics and the U.S.-China trade war. We remain long yen/euro until we see signs that Washington and Beijing are able to defuse the immediate trade war. Russia: Tensions With The West Have Not Peaked Our view that tensions between Russia and the West would peak following President Putin's reelection has been spectacularly wrong.19 We still encourage clients to review the report, penned in early March, as it sets out the limits to Russia's aggressive foreign policy. The country is geopolitically a lot more constrained then investors think, and thus there are material limits to how far the Kremlin can take the rivalry with the West. What we did not account for is that such weakness is precisely the reason for the tensions. Specifically, the Trump administration - riding high following the success of its "maximum pressure" doctrine in the Korea imbroglio - smells blood. President Trump is betting that the view of Russian constraints is correct and therefore the time to pressure Putin - and prove his own anti-Kremlin credentials - is now. But has the market gotten ahead of itself? The expanded sanctions target specific individuals and companies - EN+ Group, GAZ Group, and Rusal - and yet the broad equity market in Russia has tumbled.20 Sberbank, which is nowhere mentioned in the sanctions, fell by an extraordinary 16% since the announcement. On one hand, there does appear to be a material step-up in sanctions. Despite being focused on specific companies, the new restrictions are designed to make the entire Russian secondary bond market "not clearable." The targeting of specific companies, therefore, was merely a shot-across-the-bow. The implication for the future - and the reason that Sberbank fell as much as it did - is that U.S. investors could be forbidden - or the compliance costs could rise by so much that they might as well be forbidden - from participating in Russian debt and equity markets in the future. On the other hand, our Russia geopolitical risk index has not priced in the renewed tensions (Chart 24). This means that either our currency-derived measure is wrong or the sell off in equity and debt markets is not translating into bearishness about the overall economy. Given our bullish oil outlook and our view of the limits of Russian aggression investors should expect, the index may actually be signaling that these tensions are an opportunity to buy Russian assets. Chart 24The Russia GPI Says No Risk That said, we have learned our lesson. There is no point in trying to catch a falling knife as the Kremlin and the White House square off over Syria and other geopolitical issues. As such, we are closing all of our Russia trades until we find a better entry point to capitalize on our structural view that there are material limits to geopolitical tensions between the West and Russia. The long Russia equities / short EM equities has been stopped out at 5% loss. Our buy South African / sell Russian 5-year CDS protection is down 20 bps and our long Russian / short Brazilian local currency government bonds is up 1.07 bps. Investment Implications In April 2017, we penned a report titled "Buy In May And Enjoy Your Day!," turning the old "sell in May and go away" adage on its head.21 At the time, investors were similarly facing a number of geopolitical risks, from the second round of French elections to concerns about President Trump's domestic agenda. However, we had a very high conviction view that these risks were overstated. This time around, we fear that the markets are mispricing constraints on President Trump. Geopolitical risks ahead of us are largely in the realm of foreign policy, where the U.S. Constitution gives the president large leeway. This includes trade policy. As such, it is much more difficult to have a high conviction view on how the Trump administration will act towards China, Iran, and Russia. Furthermore, the success of the "maximum pressure" doctrine has emboldened President Trump to talk tough, worry about consequences later. Investors have to understand that we are the target of President Trump's rhetoric. There is no better way for the White House to show China, Iran, and Russia that it is serious - that its threats are credible - than if it strongly counters the view that it will do nothing to harm domestic equities. We therefore expect further volatility in the markets. We propose that clients hedge the risks this summer with our "geopolitical protector portfolio" - equally-weighted basket of Swiss bonds and gold - which is currently up 1.46%, although adding 10-Year U.S. Treasurys to the mix may make sense as well. We would also recommend that clients expect both a spike in the VIX and a rise in the volatility of the VIX (volatility of volatility). Marko Papic, Senior Vice President Chief Geopolitical Strategist marko@bcaresearch.com Matt Gertken, Associate Vice President Geopolitical Strategy mattg@bcaresearch.com 1 Please see BCA Geopolitical Strategy Weekly Report, "Political Risks Are Understated In 2018," dated April 12, 2017, available at gps.bcaresearch.com. 2 Please see BCA Geopolitical Strategy Weekly Report, "Can Equities And Bonds Continue To Rally?" dated September 20, 2017, available at gps.bcaresearch.com; and Global Fixed Income Strategy Weekly Report, "Have Bond Yields Peaked For The Cycle? No," dated September 12, 2017, available at gfis.bcaresearch.com. 3 Please see BCA Geopolitical Strategy Weekly Report, "We Are All Geopolitical Strategists Now," dated March 28, 2018, available at gps.bcaresearch.com. 4 Please see "Russia says U.S. plans to strike Damascus, pledges military response," Reuters, dated March 13, 2018, available at reuters.com. 5 Please see BCA Geopolitical Strategy Weekly Report, "Upside Risks In U.S., Downside Risks In China," dated January 17, 2018, available at gps.bcaresearch.com. 6 Please see BCA U.S. Equity Strategy Weekly Report, "Bumpier Ride," dated March 26, 2018, available at uses.bcaresearch.com. 7 Please see BCA Geopolitical Strategy Special Report, "Five Black Swans In 2018," dated December 6, 2017, available at gps.bcaresearch.com. 8 Please see "US drops contentious demand for auto content, clearing path in NAFTA talks," Globe and Mail, March 21, 2018, available at www.theglobeandmail.com. 9 Please see BCA Emerging Markets Strategy Weekly Report, "EM: Perched On An Icy Cliff," dated March 29, 2018, available at ems.bcaresearch.com. 10 Please see BCA Geopolitical Strategy Weekly Report, "Update On Emerging Markets: Malaysia, Mexico, And The United States Of America," dated August 9, 2017, available at gps.bcaresearch.com. 11 Please see BCA Geopolitical Strategy Weekly Report, "Should Investors Fear Political Plurality?" dated November 29, 2017, available at gps.bcaresearch.com. 12 Please see BCA Geopolitical Strategy Outlook, "Three Questions For 2018," dated December 13, 2017, available at gps.bcaresearch.com. 13 Please see BCA's Foreign Exchange Strategy Weekly Report, "The Euro's Tricky Spot," dated February 2, 2018, available at fes.bcaresearch.com. 14 Please see BCA Geopolitical Strategy Special Report, "Japan: Kuroda Or No Kuroda, Reflation Ahead," dated February 7, 2018, available at gps.bcaresearch.com. 15 Please see "Japan: Abe Is Not Yet Dead, Long Live Abenomics," in BCA Geopolitical Strategy Weekly Report; "The Wrath Of Cohn," dated July 26, 2017; and "Japan: Abenomics Will Survive Abe," in Geopolitical Strategy Weekly Report, "Is King Dollar Back?" dated October 4, 2017, available at gps.bcaresearch.com. 16 Please see BCA Geopolitical Strategy Weekly Report, "We Are All Geopolitical Strategists Now," dated March 28, 2018; and "Politics Are Stimulative, Everywhere But China," dated February 28, 2018, available at gps.bcaresearch.com. 17 Please see Cory Baird, "BOJ Chief Haruhiko Kuroda Begins New Term By Vowing To Continue Stimulus In Pursuit Of 2% Inflation," Japan Times, April 9, 2018, available at www.japantimes.co.jp. 18 Please see "Work style reform legislation gets Abe Cabinet approval," Jiji Press, April 6, 2018, available at www.the-japan-news.com. 19 Please see BCA Geopolitical Strategy and Emerging Markets Strategy Special Report, "Vladimir Putin, Act IV," dated March 7, 2018, available at gps.bcaresearch.com. 20 Please see Department of the Treasury, "Ukraine Related Sanctions Regulations - 31 C.F.R. Part 589," dated April 7, 2018, available at treasury.gov. 21 Please see BCA Geopolitical Strategy Weekly Report, "Buy In May And Enjoy Your Day!" dated April 26, 2017, available at gps.bcaresearch.com.

Special Report

Highlights In China, the central bank and commercial banks conducted outright monetization of real estate inventories, which caused the property markets' recovery post 2015. Despite destocking, aggregate property inventories remain excessive. Elevated inventories, poor affordability, and policy tightening will depress property demand and lead to a contraction in construction activity. Slumping construction, along with a slowdown in infrastructure investment, pose downside risks to China's demand for commodities, materials and industrial goods. This is the main risk to EM stocks and currencies and the primary reason we maintain our negative stance on EM risk assets. Continue shorting Chinese property developers stocks versus U.S. homebuilders. Feature With a flurry of policy tightening directed at the real estate market in the past year, property demand in China has weakened. The latter typically leads property starts and real estate investment, and is coincident with real estate prices (Chart I-1). Is China entering another property downturn, and if so will it be shallow, or severe? Answers to these questions are important not only for Chinese stocks, but also for China-plays throughout the rest of the world. To shed light on this issue, this week we re-examine how large the imbalances in the Chinese real estate market actually are - with respect to both affordability and supply (the stock of housing and inventories). We also discuss policy objectives and investment implications. Proper Measures Of Inventories And Housing Stock Both purchases and prices of Chinese residential properties surged between 2015 and 2017, when the authorities implemented a property de-stocking policy. As a result, housing inventories declined significantly. Does this mean that one of the major imbalances, namely swelling inventories, has been eliminated? If imbalances, namely inventories and prices, in a property market are very minor, one can expect an ensuing adjustment to be benign. Conversely, if imbalances are large, it is reasonable to bet on a meaningful property market downturn. With respect to China's real estate inventory levels, data from the National Bureau of Statistics (NBS) which many analysts follow, indicates inventories of residential buildings have indeed declined, with a significant 33% drop in residential vacant floor space for sale (Chart I-2). The term "vacant" is used by the data provider to denote the floor space completed but not sold. Clearly, China's de-stocking strategy since 2015 has worked well. Chart I-1China: Real Estate Is Slowing Down Chart I-2Property Developers' Inventories: ##br##Completed But Not Sold However, data from the NBS on vacant space for sale is not all-encompassing. First, it includes only commodity buildings - i.e., those developed by real estate developers - and does not include buildings built by non-real estate developers. For example, companies, universities, organizations and even a group of individuals can construct both residential and non-residential buildings for their own use. Commodity buildings are just a small subset of total constructed buildings in China. According to NBS data, residential buildings by property developers account for only 26% of total constructed residential buildings in terms of floor space area completed. In brief, the inventory data that the majority of analysts use covers only a part of property construction (Figure I-1). Figure I-1The Breakdown Of Residential ##br##Real Estate Inventory Second, the vacant floor space data - shown in Chart I-2 and used by many analysts - only measures commodity buildings that have been completed but not sold. It does not account for those units that are under construction and have not been sold. The latter should also be counted as inventory because in China both residential and non-residential properties can be sold even when they are in the construction phase. Unlike advanced economies, in China the housing market is by far dominated by new construction. In particular, about 80% of residential commodity floor space sold are properties that are still under construction. This is drastically different from real estate markets in the U.S. and other developed countries, where the secondary housing market is a major source of supply. Given the above,1 we propose several alternative measures that aim to more accurately reflect the real picture of Chinese property inventory. Real Estate Inventory To capture the flow of the entire residential property supply in China, we calculate the difference between cumulative floor space started and cumulative floor space sold over the period of 1995-2017. This produces a new measure of total space not yet sold (i.e., available for sale), which includes areas both under construction and completed. This is a much more comprehensive measure of the total inventory than other commonly used measures. It is important to note that this measure takes into account both types of floor space available for sale: under construction and completed. The top panel of Chart I-3 illustrates that our derived measure of residential inventory - cumulative floor space started minus cumulative floor space sold - currently stands at 2.5 billion square meters or 27 billion square feet. This is about eight times greater than the NBS measure of vacant floor space - completed by property developers but not sold, which presently amounts to only 0.3 billion square meters or 3.23 billion square feet. On the bottom panel of Chart I-3, we estimate how many months of sales it will take to clear this housing inventory. Our findings reveal that even though our new inventory measure for the residential sector has fallen sharply due to the de-stocking policy, it still takes 22 months of last year sales to clear it. This is much higher than the completed by property developers but unsold vacant space, which presently stands at 2.5 months of last year sales. Provided that (1) most housing for sale in China is new construction, and (2) it can be sold at any stage of the construction cycle, we believe our new estimate of residential inventory that is equal to 22 months of last year sales is a more accurate reflection of reality. We computed a similar measure of inventory for non-residential properties that includes malls, offices, and warehouses. The top panel of Chart I-4 shows that the proper inventory levels for the non-residential sector have kept rising to new record highs in absolute terms. Relative to floor space sold last year, inventories still stand at 170 months of sales (Chart I-4, bottom panel). Chart I-3Our Measure Of Residential Inventories: ##br##Floor Space Available For Sale Chart I-4Our Measure Of Non-Residential Inventories: ##br##Floor Space Available For Sale Clearly, China's non-residential markets still carry excessive inventories. It would be misleading to use completed but unsold data for the non-residential sector, which accounts for roughly 14 months of sales. Similar to the residential commodity buildings market, about 65% of non-residential commodity buildings sold are those that are still under construction. In short, despite the decline from 2015's exceptionally high levels, inventories for both residential and commercial properties are still extremely elevated. Furthermore, the inventory-to-sales ratio is not a good indicator for the property market outlook because it is heavily influenced by sales. When sales - the denominator of this ratio - are weak, this inventory ratio is high, and vice versa. In particular, this ratio has been a poor indicator for the property market in China, where sales of properties have been deeply influenced by government policies. Whenever sales dropped and this ratio surged, the authorities would begin easing policies, spurring sales to rise and allowing the market - prices, floor space starts and construction - to recover. As a final note, these inventory data show floor space built by property developers only. Stock Of Housing The measure of per-capita living space gauges the existing stock of housing. Hence, it is a structural measure. Still being a low-income country, China is often perceived to offer enormous construction potential. However, some statistics on per-capita living space are revealing. The NBS data show that the 2016 per-capita living space for both urban and rural area has risen to 36.6 square meters and 45.8 square meters, respectively (Chart I-5). By comparison, in Korea and Japan, living space per capita (the entire population average) is only 33 and 22 square meters, respectively. Chart I-5China: Per Capita Living ##br##Has Grown Dramatically Our calculation of per-capita urban living space based on the NBS building construction data also show similar results - 38 square meters for 2017. Consequently, these statistics on per-capita living space are supported by historical construction data, and hence are reliable. Both NBS per-capita living space data and our calculated per-capita living space data confirm that there is already massive stock of residential property in China - the nation's current existing residential floor space area already amounts to 30.8 billion square meters (332 billion square feet). Furthermore, the stock of housing is relatively new with 88% of this living space built in the past 20 years. Assuming the floor space area of each house is on average 90 square meters (970 square feet), we infer that on average every urban household already owns 1.3 houses. This is actually in line with the results of several domestic household surveys, which conclude that 20-25% of houses owned by urban residents are neither being used for living nor for renting out. Provided not every household in China owns a house, and that a meaningful share of the population still lives in smaller and older housing, these data suggest there have been considerable speculative/investor purchases of housing over the past 10 years. Many high-income individuals own multiple properties (that are often kept vacant) while a still-considerable number of families live in poor conditions. Bottom Line: China has constructed enormous amounts of real estate since 2002. Furthermore, inventories are vast for residential and non-residential sectors alike. Such an oversupply of properties poses a considerable risk to construction activity going forward. Property Demand Weakness: Cyclical Or Structural? Very poor affordability, slowing rural-to-urban migration, demographic changes, tightening mortgage lending, a successful government-led clampdown on speculative activity and the promotion of the rental housing all point to both a cyclical and structural slippage in housing purchases in China. House Price-Income Ratios and Affordability House prices in China remain extremely high relative to disposable income. By using NBS 70-city residential average price, our calculation shows for an average household (assuming double income earners) it will take 10.5 years of its disposable income to buy a 90-square-meter (equivalent to 970 square feet) house at current prices (Chart I-6). The same ratio for the U.S. is presently 3.4 and at the peak of U.S. housing bubble in 2006 it was 4. In regard to the ability to service mortgage payments, annual interest costs account for 45% of average household disposable income (assuming a double income household) when buying a 90 square meter house and assuming 20% down payment (Table I-1). Chart I-6House Price-Income Ratio: ##br##China & The U.S. Table I-1House Price-To-Income Ratios ##br##And Affordability If we use another data provider - Choice, covering 100 cities, house price per a square meter is 60% higher than the NBS 70-city residential average price. Using Choice house price data, the house price-to-income ratio is 17, and affordability - the share of interest payments as a percentage of disposable household income - is 72%. Clearly, there is a huge gap between these two aggregate measures of residential property prices. In this report, we use conservative (low) prices from the NBS, which still reveals that house prices and interest payments are exceptionally high relative to disposable income for a double-income family. Table I-1 contains house price-to-income ratios and affordability ratios for 31 provinces using the house prices from NBS. Given the average urban household already owns more than one property, it is reasonable to expect that a considerable proportion of potential future demand for housing will come from rural residents as urbanization continues, or as rural residents seek to buy homes in the city for access to better quality education in the urban areas for their children. However, rural residents' current and potential (when they move to cities) disposable income is much lower than the urban's. Therefore, housing affordability is a bigger challenge for them. Rural-to-Urban Migration Even though urbanization is an ongoing process in China and will continue for many years, the pace is slowing (Chart I-7). The number of individuals moving from rural areas to cities as a percentage of the urban population is decreasing. This will translate into decelerating growth rate in demand for urban residential properties. Chart I-7China: The Pace Of Urbanization Is Slowing The second panel of Chart I-7 illustrates that rural-to-urban net migration accelerated in the early 1990s and has been between 15-18 million people per year over the past 20 years. However, as a share of the urban population, net migration has fallen from 4.5% in the late 1990s to 2% today (Chart I-7, third panel). Overall, urban population growth has slowed below 3% (Chart I-7, bottom panel). In brief, the slowdown in net migration and, consequently, decelerating urban population growth will cap structural housing demand that has been booming over the past 20 years. Poor Demographics The Chinese population is aging rapidly. The proportion of citizens who are over the age of 65 has risen from 8% of the population in 2007 to 11.4% as of last year and will continue rising rapidly. Given Chinese life expectancy is currently at about 76 years, senior citizens cohort will leave a large number of houses to their children or grandchildren over the next 10-15 years. The reason behind this is because the former demographic cohort (11.4% of the total population) is larger than the 10-19-year-age group which accounts for only 10.5% of the total population. The latter would have been a major source of property demand over the next 10 years, as Chinese tradition requires them to own a house before marriage. However, this is no longer the case. For this generation - born in the late 1990s and 2000s and by the time they get married (in general at the age of around 25 or a bit later), each newly-formed family could potentially inherit four houses from their parents and grandparents. Tightening mortgage lending As part of the current property related restrictive policies, mortgage interest rates have been on the rise for both first- and second-home buyers. Mortgage rates have risen by 74 basis points in the past 12 months - from 4.52% to 5.26%. Additionally, banks have been tightening credit standards. Given house prices are very high relative to income, a small increase in mortgage rates meaningfully increases the share of disposable income that must be allocated to interest payments on mortgages. For example, with the house price-to-income ratio at 10.5 and down payment of 20% of house price for the average home buyer in China, a 75-basis-point increase in mortgage rates would lift the share of interest payments on a mortgage from 45% to 51% of disposable income. Hence, higher borrowing costs over the past year as well as the ongoing tightening in credit standards will continue to discourage property buyers. Mortgage loan growth has rolled over after booming between 2015 and 2017, yet at a 22% annual growth rate, it remains very high (Chart I-8). Policy-led clamp-down of speculation President Xi Jinping's mantra that "housing is for living in, not for speculation" - proclaimed in December 2016 - is the focal point of the government's current policies. Many regulations implemented by both the central government and local governments over the past 15 months have been aimed at reducing speculative purchases. The promotion of the housing rental market In large cities residential rental yields fluctuate between 1-2.5% (Chart I-9). This compares with mortgage rate of 5.3%. Currently, renting is significantly cheaper than buying. This may encourage renting in the long term. Rising demand for rental housing might be met by the available stock of empty apartments that investors have been accumulating over the years. If this occurs, it will reduce demand for new home purchases. Chart I-8China: Mortgage Lending Has Been Booming Chart I-9China: Residential Rental Yields Are Very Low Meanwhile, the central government is determined to develop a rental market by constructing rental housing. If building of rental housing offsets the potential decline in property construction, it will make our negative view on construction volumes widely off the mark. The crucial factor to watch is financing. If credit supply slows meaningfully, there will be less available financing for overall construction, including rental. Any gains by rental construction will be overwhelmed by a decline in the building of residential and commercial real estate. In turn, financing is contingent on the government deleveraging campaign. If the authorities adhere to their pledge of deleveraging, a slowdown in credit growth will dampen overall construction activity. There can be no construction without credit. Furthermore, it takes only a deceleration in credit growth, i.e., a negative credit impulse, to depress construction volumes. That is why we cover China's credit cycle dynamics in such details in our regular reports. Bottom Line: Chinese property demand is facing numerous cyclical and structural headwinds. Policy Driven Market China's central and local government policies have over time and in different combinations substantially influenced the country's housing market on both the supply and demand sides. Over the past two decades, each time the government implemented restrictive policies (for example, raising down-payment ratios, increasing policy or mortgage rates, setting restrictions on mortgage lending, and so on), the real estate market slowed and housing prices softened. The opposite has also held true - each time the government introduced stimulus, housing prices surged as buyers quickly dove into the market. Chart I-10 illustrates the interaction between government property related regulations and the domestic housing market. Chart I-10China: Policy-Driven Property Market The biggest problem with such policies in the long run is that the authorities want to control both prices and volume - they want flat prices and moderately rising volumes. However, no government can control both prices and volumes simultaneously in any industry. China's real estate market is not an exception. Even in a completely closed socialist system, controlling prices and volume simultaneously is almost impossible. As the authorities adhere to their policy objectives of controlling financial risks and unwinding financial excesses, thereby focusing on property price control over the next 12 months, we believe property starts and construction activity will shrink. Monetization of Housing Inventories In 2015-'17 Understanding what was behind the housing market's strong recovery since late 2015 is critical to assessing the outlook. Since the summer of 2015, authorities were not only easing purchasing restrictions and lowering mortgage rates, but they were also implementing outright monetization of housing inventories. After inventories of both residential and non-residential properties swelled, the central government commenced a de-stocking strategy in 2015, mainly through a monetized slum reconstruction program and by encouraging migrant workers to buy housing in smaller cities near their hometowns. The de-stocking strategy focused on smaller cities where inventories had mushroomed. Given tier-1 cities account for only 6% of floor space started by property developers, and most construction in recent years has been taking place in tier-2 and smaller cities, these policies had a substantial positive impact on national sales, as well as drawing down inventories - ultimately spurring a construction recovery. 1. The government's slum area reconstruction policy has been the major driver behind de-stocking within the residential property market. The People's Bank of China (PBoC) has provided a significant amount of financing in the form of pledged supplementary lending (PSL) directly to homebuyers that was intermediated by three policy banks (China Development Bank, Agricultural Development Bank of China and Export-Import Bank of China). To shed more detail on the PSL mechanism, the central bank lends credit to the three policy banks at very low interest rates. These policy banks in turn lend directly to local government and regional property developers (mainly in tier-2 and smaller cities). These entities then turn and buy slums from their owners which puts cash in the hands of these sellers. Consequently, a large number of households suddenly receive large cash infusions - essentially disbursed by the central bank - that can be used to purchase new and better properties. The outstanding amount - total financing - via the PSL has risen from RMB 383 billion in 2014 to RMB 971 billion in 2016. The total amount of the PSL disbursed for the slum reconstruction program over 2014-2017 amounted to 3 trillion, or 3.6% of 2017 GDP, as of March 31, 2018. The interest rate on the PSL currently stands at a mere 2.75%. It appears that huge amounts of cheap money have been directly injected into the real estate market by the central bank alone. This slum reconstruction program has had a material impact on construction activity. Chart I-11 portends that slum area reconstruction accounted for about 20% of floor space sold in 2017. Chart I-11China: Slum Reconstruction ##br##Has Had Meaningful Impact 2. In addition to the PSL financing, Chinese housing mortgages have increased by 85%, or by 11 trillion RMB in the past two and a half years - since the beginning of China's de-stocking policy. The sum of PSL financing and mortgage lending has been RMB 14 trillion (or $2.2 trillion) during the same period. Hence, not only has the PBoC financed the real estate market directly, but it has also allowed banks to flood the system with money to liquidate housing inventories. As we have argued in our series of reports, bank credit does not come from anyone's savings. Commercial banks originate loans out of thin air.2 In short, altogether these actions constitute outright monetization of real estate inventories and that caused the property markets' recovery post 2015. A Downturn Ahead? Since early 2017 and especially in the wake of last October's Party Congress, the authorities have shifted their policy focus from "de-stocking" to "eliminating speculative demand". Recent weakness in both demand and prices are a reflection of the current policy focus. This time, the government seems to have more determination to break popular perception that property prices will rise forever, and that investing in property markets cannot go wrong. Therefore, we sense the government's objective is to achieve flat or mildly declining property prices to prevent the return of speculators. In order to avoid a further ballooning of the real estate bubble, the government will raise the bar for another round of property stimulus. Therefore, if the authorities are successful in persuading speculators that prices will not rise much further in the years to come, speculative demand will wane. At the same time, not many first-time homebuyers can afford to buy at current prices. This will create an air pocket in sales and prices will deflate, at least modestly. Facing shrinking revenues and being overleveraged, real estate developers will reduce new starts, and property construction volumes will likely contract by 10% or so. Notably, floor space started by property developers in aggregate declined by 27% between 2012 and 2016 (Chart I-12). The construction slump in China, in tandem with rising supplies of commodities, led to a collapse in commodities prices in 2012-'15 (Chart 12). Hence, a decline in property construction is not unprecedented, even amid robust national income growth. We believe the acute structural imbalances will likely result in a property market downturn commensurable if not worse than those that occurred in 2011-'12 and 2014-'15. While the government will try to avoid a sudden bust, a 10% decline in both property prices and construction volumes in the next 12-18 months is our baseline scenario. The budding contraction in cement and plate glass production suggests that overall construction activity is already decelerating (Chart I-13). Chart I-12China: Property Cycles ##br##And Commodities Prices Chart I-13China: Nascent Contraction In Cement ##br##And Plate Glass Production Bottom Line: The Chinese authorities will for now maintain their current restrictions on the property market to contain financial excesses and risks in the system. This, amid lingering elevated inventories and price excesses, poses considerable downside risks to the mainland real estate market. Investment Implications Our view remains that construction activity in China is set to slump from a cyclical perspective, at least. At 13.2 billion square-meter (142 billion square-feet) the total 2017 residential and non-residential floor area under construction was immense (Chart I-14). This, along with a slowdown in infrastructure investment due to tighter control on local government finances, pose downside risks to China's demand for commodities, materials and industrial goods. This is the reason why we have been and remain bearish on commodities, Asian trade and EM risk assets. It appears that several commodities prices are finally beginning to roll over which is consistent with a slowdown in the mainland's construction activity (Chart I-15). Chart I-14China's Total Building Construction: ##br##Level And Annual Growth Chart I-15A Budding Downtrend In ##br##Commodities Prices China's construction activity is much larger than exports to the U.S. and EU combined. Hence, overall industrial activity in China is set to decelerate dragging down Asian trade flows and commodities prices despite robust domestic demand in the U.S. and EU. This heralds underweighting/shorting EM stocks, currencies and credit versus their DM counterparts. We also reiterate our long-standing recommendation of shorting Chinese property developers versus U.S. homebuilders. Chart I-16 depicts that the Chinese property developers listed in A-share market have a debt-to-equity ratio of 6 and the cash flow from operations for the median of 76 property developers has begun contracting again. Further relapse in property sales will cause their financial position to deteriorate and limit their ability to launch new or complete existing construction. In regard to U.S. homebuilders, the fundamentals in the U.S. housing market are much better than those in China. While rising U.S. interest rates could be a headwind for U.S. homebuilder share prices, they stand to resume their outperformance versus Chinese property developers (Chart I-17). Chart I-16China: Median Property Developer's ##br##Financial Ratios Are Worsening Chart I-17Short Chinese Property Developers / ##br##Long U.S. Homebuilders Ellen JingYuan He Senior Editor/Associate Vice President EllenJ@bcaresearch.com Arthur Budaghyan, Senior Vice President Emerging Markets Strategy arthurb@bcaresearch.com 1 Other oft-used measures of inventories are not correct either. Some analysts use floor space under construction data as a proxy for inventory - this is technically not correct as the data includes both the area that has already been sold in advance and the area that has been completed and sold. Others use cumulative floor space started minus cumulative floor space completed - this is also not correct as cumulative floor space completed includes areas that have not yet been sold. 2 Please see Emerging Markets Strategy Weekly Report "Is Investment Constrained By Savings? Tales Of China And Brazil," dated March 22, 2018, the link is available on page 20. Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights R-star is higher in the U.S. than in most other large economies. This includes China, where an elevated savings rate has depressed the neutral rate of interest. Countries with relatively high neutral rates like the U.S. will tend to run structural current account deficits, whereas countries with relatively low neutral rates will tend to run surpluses. The failure of the Trump administration to understand this basic economic lesson could inflame the ongoing trade spat between the two countries, at a time when populism is on the rise and China is challenging the U.S. for global influence. Fortunately, trade protectionism is less attractive when jobs are plentiful, as is the case in the U.S. today. Thus, we continue to see a market-friendly resolution to the ongoing conflict. Our base case remains that another global recession is still about two years away, which should keep the bull market in global equities intact. However, with global growth decelerating, financial conditions tightening at the margin, and the near-term signal from our proprietary MacroQuant model stuck in bearish territory for the second month in a row, the tactical picture for stocks remains rather murky. Feature Blame It On The Neutral Rate If the world of macroeconomics were set in a superhero universe, the real neutral rate of interest, otherwise known as R-star, would undoubtedly be cast as an arch-villain. R-star is the interest rate consistent with full employment and stable inflation. A depressed R-star has made the zero lower-bound constraint on nominal rates a vexing problem for central bankers. Not long after the Global Financial Crisis began, policy rates fell to ultra-low levels. But even this was not enough to engender a strong recovery. Most economies needed negative real rates. However, with inflation stuck at low levels, there was a limit to how far below zero real rates could go. Japan, of course, has been no stranger to this problem. Policy rates have been close to zero for over 20 years, yet inflation remains stubbornly low (Chart 1). Some commentators have dismissed this issue, noting that real per capita GDP has still managed to grow at a reasonably healthy clip. Unfortunately, this misguided optimism ignores the fact that Japan was only able to keep the economy from sinking into a depression by relying on massive budget deficits. With Japanese monetary policy rendered impotent, fiscal policy had to pick up the slack. High levels of excess private-sector savings were absorbed with continued government dissavings (Chart 2). The result is a gross government debt-to-GDP ratio of 240%. A low R-star has also been a major problem in the euro area. Before the European sovereign debt crisis erupted, Germany was able to export its excess savings to the peripheral countries, who were more than happy to load up on cheap debt so that they could live beyond their means (Chart 3). Chart 1Japan: Even Zero Interest Rates ##br##Were Not Enough To Spur Inflation Chart 2Japan Relied On Large Fiscal Deficits And Current Account Surpluses To Offset The Rise In Private-Sector Savings Chart 3The European Periphery Is No Longer ##br##Absorbing Germany's Excess Savings Those days are over. Today, Germany's current account surplus stands at a gargantuan 8% of GDP, but much of Germany's savings are exported to the rest of the world. Consequently, the euro area current account balance has gone from roughly breakeven in the pre-crisis period to a surplus of 3% of GDP. This likely means that the neutral rate in the euro area has fallen further. R-Star In China Chart 4China Saves A Lot What about China? One might think that China's fast trend GDP growth rate would translate into a high neutral rate. However, the neutral rate is not just a function of trend growth. Most economic models state that the savings rate also affects the neutral rate.1 The more income people wish to save at any given interest rate, the lower the neutral rate will be. For a variety of institutional and cultural reasons, the Chinese save a lot (Chart 4). The national savings rate has averaged 50% of GDP for the past decade. In fact, despite an investment-to-GDP ratio of 44%, China still manages to run a current account surplus (remember the current account balance is just the difference between savings and investment). A Simple Thought Experiment The earth does not trade with Mars. As a result, the global current account balance must be zero; current account surpluses in one set of countries must be offset by current account deficits in another set of countries. Interest rates and exchange rates play a vital role in ensuring that this identity is satisfied. Imagine a bunch of island economies - all with different neutral rates - that do not trade with one another. Now suppose a technological breakthrough occurs that permits free trade and capital mobility. What would you expect to happen? Standard economic theory says that capital will flow towards the islands with relatively high interest rates. As shown in Chart 5, the flood of capital will push down the interest rate in those economies. A lower interest rate, in turn, will discourage saving and encourage investment, leading to a current account deficit. Capital inflows will also drive up the currency, while higher spending will push up consumer prices. Such a "real appreciation" of the exchange rate is necessary to ensure that increased spending falls primarily on foreign-made goods.2 Chart 5Interest Rates And Current Account Balances In An Open Economy On the flipside, capital will flow out of economies with low neutral rates, putting upward pressure on interest rates. A higher interest rate will lead to more savings and less investment, translating into a current account surplus. Countries with relatively low neutral rates will also see a real depreciation of their exchange rates. If there is complete free trade and capital mobility, the final equilibrium will be one where interest rates are equalized across all islands and the current account deficits of the islands with relatively high neutral rates are exactly offset by the current account surpluses of the islands with low neutral rates. In addition, countries with relatively high neutral rates will end up with exchange rates that appear somewhat overvalued relative to their fair value, while those with low neutral rates will have exchange rates that appear somewhat undervalued. U.S.-China Trade Tensions: An Inevitable Conflict There are many structural reasons why the U.S. and China are at loggerheads over trade these days. We predicted that Trump would win the presidency largely because we thought the political/media establishment was underestimating the importance of the populist wave sweeping across the U.S. and much of the world. Our geopolitical analysts share this view. They have also argued that China's growing economic, military, and technological prowess will inevitably put it into conflict with the U.S., which has been the world's sole hegemon ever since the Soviet Union collapsed.3 This week's report adds another structural reason to the list. While R-star in the U.S. is fairly low by historic standards, it is higher than in most other countries, reflecting America's favorable demographics, large fiscal deficit, and relatively spendthrift culture. This means that the U.S. must run a structural current account deficit. This, of course, is at odds with the Trump administration's stated objectives. Efforts by China or any other country to "talk up" their currencies in the hopes of placating Trump will fail. The U.S. economy is already operating at close to full employment. A weaker dollar would only shift the composition of spending towards domestically-produced goods. The U.S., however, does not have enough spare labor to produce these additional goods. All that would happen is that inflation would rise, rendering U.S. exporters less competitive. More stimulative fiscal policy will further increase the neutral rate of interest in the United States. Chart 6 shows that the budget deficit is set to widen to nearly 6% of GDP by 2019 even if the unemployment rate continues to decline. A larger budget deficit will drain national savings, shifting the savings schedule in the savings-investment diagram discussed earlier to the left. This will result in a bigger current account deficit (Chart 7). Chart 6The U.S. Budget Deficit Is Set To Widen Even If The Unemployment Rate Continues To Decline Chart 7A Bigger U.S. Budget Deficit Will Cause The U.S. Neutral Rate To Rise, ##br## Leading To A Larger Current-Account Deficit Investment Considerations The specter of trade protectionism is here to stay, as is the prospect of escalating U.S.-China geopolitical tensions. Fortunately, beggar-thy-neighbor policies are less attractive when jobs are plentiful, as is the case in the U.S. today. Trump also remains constrained by the stock market's view of his actions. After all, this is a president who likes to measure the success of his economic agenda by the value of the S&P 500. As such, we expect both the U.S. and China to follow a two-pronged approach to trade issues over the coming months. Publicly, they will snipe at one another, threatening each other with tariffs and other trade barriers. Privately, they will seek out a compromise that avoids a full-out trade war. China's announcement this week that it will retaliate in kind to the U.S. decision to impose tariffs on $50 billion in Chinese imports should not have taken anyone by surprise. The Chinese government had repeatedly said that they would do precisely this. Importantly, U.S. tariffs do not kick in until June. Between now and then, negotiators from both sides will try to hammer out a deal. Just as with the steel and aluminum tariffs, the final set of tariffs will be a watered-down version of the original proposal. Political theatre will be the name of the game. As discussed in last week's Q2 Strategy Outlook, our base case remains that another global recession is still about two years away, which should keep the bull market in global equities intact.4 We warned investors to "Take Out Some Insurance" on February 2nd, one day before the VIX spike began.5 Now that the S&P 500 is 7% off its highs, our bet is that the path of least resistance for global equities over the next 12 months is up. Nevertheless, with global growth decelerating, financial conditions tightening at the margin, and the one-month ahead signal from the beta version of our forthcoming proprietary MacroQuant model stuck in bearish territory for the second month in a row, the tactical picture for stocks still looks rather murky (Chart 8). For the time being, short-term investors should sell the rallies and buy the dips. Chart 8MacroQuant Model: Tactical Picture For Stocks Still Looks Rather Challenging Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com 2 The real exchange rate can be thought of as the volume of foreign goods and services that can be acquired by selling a basket of U.S. goods and services. Mathematically, the real exchange rate between two currencies is the product of the nominal exchange rate and the ratio of prices between the countries. A real appreciation tends to make a country less competitive, either through a nominal increase in its currency or through an increase in domestic prices relative to foreign prices. 3 Please see Geopolitical Strategy Special Report, “Sino-American Conflict: More Likely Than You Think, Part II,” dated November 6, 2015; and Global Investment Strategy Special Report, “The Looming Conflict In The South China Sea,” dated May 29, 2012. 4 Please see Global Investment Strategy Q2 Strategy Outlook, “It’s More Like 1998 Than 2000,” dated March 30, 2018. 5 Please see Global Investment Strategy Weekly Report, “Take Out Some Insurance,” dated February 2, 2018, available at gis.bcaresearch.com. Strategy & Market Trends Tactical Trades Strategic Recommendations Closed Trades

Highlights Solid fundamentals will keep the backwardation in the forward curves of the benchmark crude-oil streams - WTI and Brent - intact. If our long-held thesis is correct and OPEC 2.0 becomes a durable producer coalition, we believe it will maintain some level of production cuts in 2019.1 This will, in part, keep OECD commercial oil inventories close to their 2010 - 2014 levels, thus keeping oil forward curves backwardated beyond this year. Backwardation serves OPEC 2.0's interests by limiting the rate at which shale-oil production grows.2 It also drives returns from long-only commodity-index exposure, particularly the energy-heavy index exposure we favor, by maintaining an attractive roll yield for investors.3 We expect the S&P GSCI to return 10 - 20% this year. Energy: Overweight. Our recently concluded research shows commodity index exposure hedges portfolios against inflation risk. We remain long index exposure. Base Metals: Neutral. COMEX copper traded back through $3.00/lb on the back of strong official Chinese PMI data, indicating manufacturing activity continues to expand. It has since fallen back to ~ $3.00/lb, as U.S. - Sino trade-war fears grew. Precious Metals: Neutral. Gold remains range-bound, between $1,310 and $1,360/oz. Ags/Softs: Underweight. In a tit-for-tat fashion, Beijing announced on Wednesday that it would retaliate to the U.S. tariffs on $50 billion worth of Chinese imports. U.S. soybeans and beef are among the list of 106 items China plans to impose a 25% tariff on. Feature An unlikely commonality of interests unites the fates of OPEC 2.0 and long-only commodity index investors: The desire to see the crude-oil forward curves backwardated. Turns out, both interests benefit from the same configuration of the forward curves, in which prompt prices trade premium to deferred prices. Backwardation achieves a critical goal of OPEC 2.0 by making the prices most member states in the coalition receive on their crude oil sales - i.e., the spot price indexed in their term contracts - the highest point along the forward curve. A backwardated curve means the average price U.S. shale-oil producers realize over their hedging horizon - typically two years forward - is, perforce, lower than the spot price. We have shown rig counts are highly sensitive to the level and the shape of the WTI forward curve. A backwardated curve reduces the revenue that can be locked in by hedging. This reduces the number of rigs shale producers send to the field, which restrains - but does not quash - the rate at which they can grow their production (Chart of the Week). For commodity index investors - particularly those with exposure to the energy-heavy S&P GSCI index, where ~ 60% of the index is crude oil, refined products or natural gas - backwardation drives roll-yields, which are a critical component of the index's total returns. The steeper the backwardation, the higher the roll yield.4 Our balances modeling indicates oil markets will remain tight this year, given strong global growth in demand in excess of production growth, which will keep the market in a physical deficit (Chart 2). This will cause inventories to continue to draw this year (Chart 3), which will keep the crude-oil backwardation in place. This backwardation is one of the principal drivers of returns in the S&P GSCI. Chart of the WeekBackwardation Constrains##BR##Shale's Rate Of Growth Chart 2Balances Model Indicates##BR##Physical Deficit Persists This Year Chart 3Tighter Inventories Keep##BR##Backwardation In Place As for the other components of the S&P GSCI, we are neutral base and precious metals, expecting them to remain relatively well-balanced this year, and underweight ag markets, even though they appear to have bottomed, as the USDA indicated recently. As a result, we expect an energy-heavy commodity index exposure like the S&P GSCI will continue to perform for investors, driven largely by the stronger oil prices we expect this year, and the roll yields from backwardated energy futures. Any price upside from the other commodities will be a marginal contribution to returns, as energy price appreciation plus roll yields will be the primary driver of the long-index exposure. Can Crude Oil Backwardation Persist? Beyond 2018, reasonable doubts exist as to whether OPEC 2.0 can remain a durable coalition. These doubts arise from apparent differences in the long-term goals of OPEC 2.0's putative leaders, KSA and Russia. We believe that, over the short term (two years or so) KSA favors higher prices, and that the Kingdom's preferred range for Brent is $60 to $70/bbl, at least until the Saudi Aramco IPO is fully absorbed and trading in the market. Russia's apparent preference is for lower prices ($50 to $60/bbl), which will disincentivize U.S. shale producers from adding even more volume to the market and threaten its market share. How these goals are resolved within OPEC 2.0 as it negotiates its post-2018 structure will determine whether oil forward curves remain backwardated - the likely outcome if production cuts are extended into 2019 - or if OECD inventories start to rebuild and the backwardation returns to contango (i.e., deferred prices exceed prompt prices). This would happen if Russia and its allies decide they are uncomfortable with prices staying close to or above $70/bbl for too long, and therefore lift production and exports to bring them down. OPEC 2.0 Has Reconciled KSA's And Russia's Goals We believe OPEC 2.0 has reconciled KSA's desire for higher prices over the short term to allow a smooth IPO of Aramco. Both KSA and Russia share a longer-term goal of not overly incentivizing U.S. shale production, and production by others - e.g., Norway's Statoil - which also have significantly reduced their costs in order to remain competitive.5 If OPEC 2.0 is successful in achieving higher prices over the short term, it will have to offset them with lower prices further out the forward curve to reconcile KSA's and Russia's goals. This is the principal reason we believe backwardating the forward curve, and keeping it backwardated, achieves OPEC 2.0's short- and longer-term goals. After Aramco is IPO'd - something that, from time to time, seems doubtful - and the market's trading the stock, we believe KSA and Russia will want average prices to drift lower. KSA will, by that time, have lowered its fiscal break-even cost/barrel to $60 (they're at or below $70 now) and will be executing on its diversification strategy. But even with spot prices lower - we're assuming the target level would be ~ $60/bbl - the forward curve will have to remain backwardated to keep U.S. shale's growth somewhat contained. This can be done by keeping deferred contracts (2+ years out) close to $50/bbl using OPEC 2.0 production flexibility, global inventory holdings and forward guidance re production, export and inventory policies. By keeping the average price realization over the shale producers' hedging horizon in the low- to mid-$50s, OPEC 2.0 restrains rig deployment in the U.S. shales. Keeping the front of the forward curve closer to (or above) $60/bbl, means OPEC 2.0 member states get the high price on the forward curve, since their term contracts are indexed to spot prices. Once a persistent backwardation becomes a reliable feature of the forward curve, the short-term inelasticities of the global supply and demand curves - but mostly the supply curve - mean small changes by a production manager like OPEC 2.0 can readily change the price landscape and alter expectations along the forward curve covering the shale-oil producers' hedge horizon. OPEC 2.0 states already have lived through the alternative of not managing production to the best of their abilities during the 2014 - 2016 price collapse: A production free-for-all similar to what the market experienced then would again lead to massive unintended inventory accumulations globally. This would put the Brent and WTI forward curves into super-contangos, which occurred at the end of 2015 into early 2016. At that point, the market would, once again, begin pricing sub-$20/bbl oil as a global full-storage event becomes more probable. At that point, it's "game over" for OPEC 2.0 member states. The stakes remain sufficiently high for OPEC 2.0 member states to keep the coalition intact and to maintain production cuts to keep OECD inventories tight, and thus keep markets backwardated beyond 2018. Backwardation Works For Commodity Index Investors, Too We expect the S&P GSCI to continue to perform well this year - posting gains of 10 to 20% - given our expectation OPEC 2.0 will remain committed to maintaining production discipline. We've recently shown there is a close relationship between oil forward curves and oil inventories, expressed as the deviation of Days-Forward-Cover (DFC) from its 2- or 3-year average, and y/y percentage change (Chart 4).6 This analysis supports our view that - based on our expectation of a continuation of OECD commercial inventory decline - backwardation will continue throughout 2018 and early-2019. This tight relationship, allows us to include OECD commercial inventories as a proxy among our explanatory variables for the shape of the oil forward curves, when modeling and forecasting the GSCI total return. For 2018, we are modeling a continuation of the production cuts put in place at the beginning of 2017 to year end. At some point later this year, we expect the market to get forward guidance on what to expect in the way of OPEC 2.0 production levels for next year. In lieu of actual guidance, we've modelled three different scenarios for OPEC 2.0's production levels next year, leaving everything else affecting prices unchanged. This is a sensitivity analysis on OPEC 2.0's production only (Chart 5).7 Chart 4Oil Inventories, Spreads,##BR##DFC, Closely Related Chart 5BCA's 2019 Scenario Analysis##BR##For OPEC 2.0 Production Scenario 1: Our actual balances, most recently updated in our March 22, 2018, publication, with no production cuts in 2019; Scenario 2: An extension of the OPEC 2.0 production cuts to end-2019 at 100% of 2018 levels; Scenario 3: An extension of the OPEC 2.0 production cuts to end-2019 at 50% of 2018 levels. Under scenario 1, the GSCI's y/y returns slow in 2H18 and become negative in 3Q19. Returns peak in Feb/19 at 28%, and average 21% in 2018, and 9% in 2019. In scenario 2, y/y growth remains positive this year and next, peaking in Feb/19 at 30%, then falling to 13% in 2019. Average returns in 2018 are 21%, and in 2019 19%. In scenario 3, y/y growth remains positive in both years, and bottoms close to 0% but never turns negative. GSCI returns peak in Feb/19 at 29%, then fall to 3% in 2019. Average returns in 2018 are 21%, and in 2019 14%. Given the guidance already conveyed by KSA's oil minister Al-Falih, we would put a low weight on scenario 1, and attach a 50% probability to each of the 2019 simulations in scenarios 2 and 3. GSCI As An Inflation Hedge Our analysis shows the GSCI Total Return (TR) also is highly sensitive to the USD broad trade-weighted dollar (TWIB) and U.S. headline CPI inflation (Chart 6).8 This has powerful implications for the evolution of commodity-indices going forward. A decrease (increase) in the USD TWIB increases (decreases) USD-denominated commodity demand from buyers ex-U.S., thus raising prices, all else equal. An increase (decrease) in the U.S. CPI can lead to higher commodity costs, which are reflected in the GSCI, or to a positive (negative) net-inflow of cash into commodity-indices as a hedge against inflation risks. Importantly, we found the GSCI TR and U.S. CPI relationship to be bi-directional, enhancing the magnitude of the impact of a change in any of those variables. In other words, a rise in the GSCI TR causes inflation to rise which leads to a rise in the GSCI TR, and vice-versa until a new equilibrium is reached.9 Our colleagues at BCA's Global Fixed Income Strategy desk expect inflation pressures will continue to build this year. In particular, they note, "the global cyclical backdrop is boosting inflation."10 With 75% of OECD countries operating beyond full employment, capacity-utilization rates in the developed economies are approaching 80% - the highest level since mid-2008 (Chart 7, top panel). This closing of the global output gap likely will stoke inflation. Chart 6GSCI Highly Sensitive To USD, U.S. CPI Chart 7Inflation Risks Picking Up Consistent with our overweight view, we expect oil prices to move higher from current levels, as refiners come off 1Q18 maintenance turn-arounds and summer-driving-season demand picks up in the Northern Hemisphere (Chart 7, middle panel).11 Lastly, global export price inflation is showing no signs of slowing, suggesting that global headline inflation will continue moving higher (Chart 7, bottom panel). From the model shown in Chart 6, which captures ~ 82% of the variance in the y/y GSCI TR, we have high conviction that three of the four explanatory variables for the GSCI - crude spreads, DFC and U.S. CPI - will support the GSCI this year, leaving only a significant appreciation in USD TWIB as a potential risk to our view. Away from our modelling, other risks to our bullish oil case as a driver of GSCI returns remains a greater-than-expected economic deceleration in China arising from a policy error in Beijing as policymakers execute a managed slowdown, or a trade war with the U.S.12 These would affect our inflation and commodity-demand - hence commodity price - outlooks. Bottom Line: We expect persistent backwardation in the benchmark crude-oil forward curves- WTI and Brent - as OPEC 2.0 extends production cuts beyond 2018. This will achieve the goals of OPEC 2.0's leadership and underpin returns in the S&P GSCI, which we expect will post gains of 10 - 20% this year. Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com Hugo Bélanger, Research Analyst Commodity & Energy Strategy HugoB@bcaresearch.com 1 Last month, the Kingdom of Saudi Arabia's (KSA) oil minister, Khalid Al-Falih, indicated OPEC 2.0 production cuts could be extended into 2019. Al-Falih suggested the level of the cuts could be at a reduced level. Please see "Saudi expects oil producers to extend output curbs into 2019," published by uk.reuters.com March 22, 2018. 2 OPEC 2.0 is the producer coalition led by KSA and Russia, which, at the end of 2016, agreed to remove 1.8mm b/d of production from the market. 3 Commodity-index total returns are the sum of price appreciation registered by being long the index; "roll yield," which comes buying deferred futures in backwardated markets, letting them roll up the forward curve as they approach delivery, selling them, then replacing them with cheaper deferred contracts in the same commodity; and collateral yield, which accrues to margin deposits on the futures comprising the index. For a primer on commodity index investing, please see "Convenience Yields, Term Structures & Volatility Across Commodity Markets," by Michael Lewis in An Investor Guide To Commodities (pp. 18 - 23), published by Deutsche Bank April 2005. 4 By way of a simplistic example, assume the oil exposure in an index is established in a backwardated market - say, spot is trading at $62/bbl and the 3rd nearby WTI future trades at $60/bbl. Assuming nothing changes, an investor can hold the 3rd nearby contract until it becomes spot, then roll it (i.e., sell it in the spot month and replace it with another 3rd nearby contract at $60/bbl) for a $2/bbl gain. This process can be repeated as long as the forward curve remains backwardated. 5 Please see "How we cut the break-even prices from USD 100 to USD 27 per barrel" on Statoil's website at https://www.statoil.com/en/magazine/achieving-lower-breakeven.html and "OPEC 2.0 Getting Comfortable With Higher Prices," published by BCA Research's Commodity & Energy Strategy February 22, 2018, where we discuss how KSA's and Russia's goals have been reconciled. It is available at ces.bcaresearch.com. 6 Please see BCA Research's Commodity & Energy Strategy Weekly Report titled "Oil Price Forecast Steady, But Risks Expand," dated March 22, 2018, available at ces.bcaresearch.com. 7 This sensitivity analysis allows only for the path of OECD commercial inventories to vary while everything else is held constant. To obtain the forecasted values, we've combined the estimates of a set of different modelling techniques (i.e., a Markov switching model, threshold and break-OLS estimators). This increased the information and granularity obtained from the model and allowed us to capture time-varying characteristics in the global inventory/GSCI TR relationship. 8 We found there is two-way Granger-causality between the S&P GSCI and U.S. CPI y/y changes. This feedback loop indicates the GSCI will move with, and cause movement in, the CPI, as discussed herein. 9 This is supported statistically using Granger Causality tests in a VAR model of the GSCI TR and U.S. CPI inflation. 10 Please see BCA Research's Global Fixed Income Strategy Weekly Report titled "Nervous Complacency," published March 27, 2018. Available at gfis.bcaresearch.com. 11 Please see BCA Research's Commodity & Energy Strategy Weekly Report titled "Oil Price Forecast Steady, But Risks Expand," for our latest oil price forecast. It was published March 22, 2018, and is available at ces.bcaresearch.com. 12 Please see BCA Research's Commodity & Energy Strategy Weekly Report titled "China's Managed Slowdown Will Dampen Base Metals Demand," for a discussion of this risk. It was published March 29, 2018, and is available at ces.bcaresearch.com. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades Commodity Prices and Plays Reference Table Trades Closed in 2018 Summary of Trades Closed in 2017

Highlights The U.S. and China have a roughly 60-day period to prevent the current trade "skirmish" from metastasizing into a full-blown trade war; The revised U.S.-Korea trade deal suggests that Trump's trade negotiators are credible and are targeting China, not U.S. allies; The U.S. will demand that China's recent RMB appreciation is backed by a long-term reduction in foreign exchange intervention; Tariff reciprocity is not significant, but market access and investment reciprocity are; China will offer concessions first, and will only go to a trade war if Trump imposes sweeping tariffs anyway; Short Chinese technology stocks; remain short China-exposed S&P500 stocks in expectation of further volatility. Feature The market is coming to terms with the fact that President Trump is willing to put his policies where his campaign rhetoric was, at least on trade policy. U.S. equities are down 5.7% since the White House announced Section 232 tariffs on steel and aluminum and 2.34% since it announced forthcoming Section 301 tariffs against China. Although we have cautioned clients since November 2016 that protectionism is a real risk to global growth and risk assets,1 we believe that the current set of U.S. demands on China justify the moniker of a "trade skirmish," rather than a full-out war.2 That said, the 5.7% drawdown is appropriate, if a bit sanguine. Our "trade skirmish" view is low-conviction. President Trump remains unconstrained on trade policy, giving him leeway to be tougher than the market expects. As such, it is appropriate for the market to price a 20%-30% probability of a full-blown trade war. Given that the market drawdown in such a scenario could be 20% or more, the current market action is appropriately pricing the worst-case scenario. Why would a trade war between the U.S. and China elicit a bear market in U.S. equities if a similar confrontation between Japan and the U.S. did not in the late 1980s? For three reasons. First, the overvaluation of stocks is much greater today. Second, interest rates are much lower, restricting how much policymakers can react to adverse risks. Third, supply chains are much more integrated today, globally and between China and the U.S. Nearly every major S&P 500 multinational corporation is in some way exposed to these supply chains. As such, we think the current drawdown is appropriate. That said, the administration's policy is not haphazard. President Trump and U.S. Trade Representative (USTR) Robert Lighthizer are on the same page, making China - and not NAFTA trade partners or South Korea - the main target of U.S. protectionism (Chart 1). The rapid pace at which the administration pivoted from global tariffs to targeting China gives a clear indication of what is afoot. The U.S. is using the threat of tariffs to cajole its allies into tougher trade enforcement against China (Table 1).3 We think this strategy can work, as outlined last week, but there is plenty of room for mistakes that could derail it. Chart 1China, Not NAFTA, In The Crosshairs Table 1U.S. Gradually Exempting Allies From Tariffs Trump also wants to change U.S. policy on immigration and could use the NAFTA negotiation to gain leverage over Mexico. There is therefore still some probability that Trump triggers Article 2205 to leave NAFTA, but we believe it has declined substantively since we put it at 50% in November, particularly given the U.S.-South Korea negotiations we discuss below.4 This week we take a look at the revised U.S.-Korea trade deal and what it suggests about the Trump administration's trade agenda more broadly. Then we update the status of the U.S.-China trade frictions, which are only temporarily subsiding, if at all. Lessons From The KORUS Talks The just-completed renegotiation of the U.S.-Korea free trade agreement (the "KORUS FTA") offers some clues to the Trump administration's trade tactics that may be relevant for future negotiations with NAFTA partners, China, and others. President Trump has repeatedly criticized the KORUS FTA, as the U.S. trade deficit with South Korea has ballooned since its implementation in March 2012 (Chart 2). Trump used the threat of withdrawing from the deal to pressure South Korean President Moon Jae-in not to ease sanctions on North Korea too rapidly. Chart 2Why Trump Likes Tariffs Now USTR Lighthizer and his South Korean counterpart, Hyun Chong-Kim, have agreed to the outlines of a revised deal.5 The key points are as follows: Steel tariff waiver for Korea: South Korea will receive a country-level exemption from the U.S.'s recently imposed steel tariffs.6 Going forward, Korean steel exports will be subject to quotas equivalent to 70% of the average annual import volume during 2015-17. Greater market access for U.S. autos: Korea will double the number of autos it imports on the basis of U.S. safety standards, from 25,000 to 50,000 per year from each U.S. carmaker. It can import more subject to its own safety standards. It will refrain from any new emissions-standards tests, will accept U.S. safety standards on auto parts, and will ease ecological policies and the customs process of verifying the origin of exports. Delayed market access for Korean trucks: The U.S. will retain the existing 25% tariff on Korean trucks through 2041, instead of 2021 (Chart 2, second panel). Fair treatment of U.S. pharmaceutical imports: Korea promises not to discriminate against U.S. drugs but to grant them fair treatment under KORUS provisions. Ancillary currency agreement: The two sides appended a "gentleman's agreement" on currency policies, which is not a formal part of the deal and not subject to legislative confirmation. South Korea agreed not to devalue the won competitively, or to manipulate it more broadly, and to provide greater transparency regarding its interventions in foreign exchange markets. There are three main takeaways from the above. First, the U.S. is obviously focusing on non-tariff barriers to trade, the main hindrance to trade in a world with already low tariff rates. The grievances with Korea were primarily due to safety standards, environmental policies, and burdensome administration that hindered U.S. exports despite the reduction of tariffs under the KORUS agreement. Second, USTR Robert Lighthizer - the seasoned negotiator of the historic 1980s trade disputes with Japan, and the man in charge of the current NAFTA and China negotiations - deserves his reputation as a competent policymaker. He apparently makes concrete demands and is capable of compromising to conclude deals. This reduces the risk, overstated by the media, that the inexperienced U.S. president is driving the trade negotiations. Third, the U.S. is not deliberately trying to punish its allies in pursuit of some mercantilist fantasy of closing every single trade imbalance. Strategic logic dictated that Washington and Seoul needed to conclude a deal quickly so as to better coordinate on North Korea, and they did so. It is highly unlikely that the concluded deal will end the U.S. trade imbalance with South Korea, but it will likely improve it substantively. Moon Jae-in continues to be a pragmatist in his dealings with Trump and Trump is joining Moon's "Moonshine" policy of engagement with North Korea. Talk of the U.S. abandoning its allies did not materialize. (Japan and Taiwan are likely to get deals soon.) Most importantly, this deal is a strong indication that the U.S. will continue to pressure China on its foreign exchange practices. It would make no sense for the U.S. to require its allies to disavow competitive devaluation and reduce currency interventions while not demanding similar assurances from China. On this front, China's recent appreciation of the yuan will not ultimately satisfy the U.S., as it is arbitrary. The U.S. will need to extract deeper guarantees, with the implicit threat of tariffs to prevent China from backsliding. Otherwise the U.S. would yield Chinese exporters a foreign exchange advantage relative to American trade partners who agree to stop intervening to preserve a favorable exchange rate with the USD. A simple comparison of these countries currency moves over the past eight years reveals how they have allowed less appreciation relative to the U.S. than in trade-weighted terms, and how China would benefit if the others were forced to stop this practice while it was left off the hook (Chart 3). Chart 3The U.S. Will Demand Currency Appreciation This last conclusion fits with our study of previous cases of U.S. trade protectionism, in which the end-game was dollar depreciation relative to key trade partners.7 The KORUS case can be considered alongside Lighthizer's and the Trump administration's handling of the Section 301 investigation into China's forced tech transfer and intellectual property theft. The Trump administration came out swinging with unilateral 25% tariffs on about $60 billion worth of goods, to be listed on April 6 and enacted sometime in June. But it also signaled that it would allow a consultation period, and initiated a case through the World Trade Organization, thus reinforcing (rather than undermining) the global trading system. These developments give some grounds for optimism in the NAFTA negotiations and (less so) in the China negotiations. While China is preempting U.S. demands on its currency policy, it will be averse to providing any permanent guarantees, or to painful structural demands. This is due to its concerns about overall stability and its suspicion that the U.S. is pursuing a broader strategic containment policy against it. We discuss these issues below. Bottom Line: The preliminary conclusions of the KORUS FTA negotiation suggest that the Trump administration's trade leadership is credible, while Trump himself is looking for quick and concrete trade "wins" that can be presented to his domestic voter base. This is a marginally market-positive sign. But its ramifications are limited with regard to China, where strategic tensions and geopolitical competition will make it much harder to strike a similar deal quickly. U.S.-China: Fade The "Mirror Tax," Focus On Market Access And Tech China announced tariffs on roughly $3-$3.5 billion worth of U.S. goods on April 2 - ranging from fruits and nuts to wine and pork - in retaliation for the steel and aluminum tariffs that the U.S. imposed in March under Section 232 of the Trade Expansion Act of 1962. China used the exact same tariff rates as the U.S. - 25% and 10% - while selecting the product list so as to produce roughly the same net trade impact in USD terms (Chart 4). The implication is that China will retaliate in kind to deter the U.S., but does not wish to "up the ante." This is largely what we expected, but the implication is significant: the U.S. is about to release a preliminary list on April 6 of $50-$60 billion worth of goods on which it will slap tariffs. This second round of tariffs - which is China-specific - follows from the probe under Section 301 of the Trade Act of 1974. China's recent decision suggests that if negotiations fail, it will respond with tariffs worth roughly the same amount, which is a much bigger exchange of fire for these two economies. The actual retaliatory action would most likely occur in June, when the U.S.'s list is finalized and implemented, though China may hint at its product list much sooner, adding to trade fears and market volatility.8 The Trump administration claims that its product list will be chosen by an algorithm to maximize the impact on Chinese exporters while minimizing the impact on the American consumer. Consistent with this aim, some reports indicate that the goods will be advanced technological products set to benefit from China's "Made in China 2025" plan, in which China has laid down aggressive domestic content requirements (Chart 5). Chart 4Tit For Tat Chart 5China's High-Tech Protectionism What is the Trump administration's goal? Treasury Secretary Steve Mnuchin declared at the G20 finance ministers' meeting that he did not want to penalize Chinese imports so much as promote U.S. exports. Is this a credible basis for assessing the administration's policy? Yes and no. We think Mnuchin is telling the truth, but not the whole truth. When it comes to blocking imports or boosting exports, Mnuchin is right: the U.S. goal is not simply to punish Beijing for past unfair trade practices by blocking imports of Chinese goods. True, the Trump administration has focused on a lack of reciprocity in tariff rates. But a "mirror tax" or "mirror tariff" with China, which Trump has referred to, would not make much of a difference to the trade balance: Chart 6AThe U.S. Exports Soybeans And Cars To China Chart 6BChina Exports Phones And Computers To The U.S. Taking a look at the top ten exports of the U.S. and China to each other (Chart 6 A&B), it is quite clear that China imposes higher tariffs on U.S. goods than the U.S. imposes on Chinese goods (Chart 7 A&B). This follows from World Trade Organization rules and the relative level of economic development of the two countries. Chart 7AAmerican Exports To China Face Higher Tariffs... Chart 7B... Than Chinese Exports To America If we equalize these tariffs by raising U.S. tariffs to the same level as their Chinese counterparts for the same good, we wind up with a very small $6.2 billion gain to the U.S. trade balance (Chart 8). If we focus only on the top ten goods that both countries export to each other, and impose a hypothetical mirror tax, we wind up with an even smaller gain for the U.S. of $3.9 billion (Chart 9). This is small fry and cannot be the administration's goal (at least not its main goal). The real goal is to gain greater market access for U.S. exports in China. Here the U.S. may have a case, as China lags both its developed and emerging market peers in sourcing its imports from the U.S. (Chart 10). While China comprises 24% of total EM imports, it comprises only 15% of U.S. exports to EM. Even in commodity exports, where the U.S. has made major inroads in China, Beijing has recently limited the American share (Chart 10, middle panel). Chart 8Equalizing Tariffs Has Little Impact Chart 9Equalizing Tariffs Has Little Impact (2) Chart 10U.S. Grievance Is About Market Access A simple, back-of-the-envelope comparison of the U.S.'s top exports to China and EM ex-China suggests that the U.S. can make a case that its exports are suffering unduly in China: China's share of top U.S. exports is lower than one might expect it to be relative to EM or EM-ex-China (Chart 11 A&B). The U.S.'s market share of China's imports in key goods is lower than it is in EM or EM-ex-China (Chart 12 A&B). The U.S. share of China's top imports is smaller than the DM-ex-U.S. share (Chart 13 A&B). Chart 11AChina Is Not A Large Enough Share Of U.S. Exports (Broad) Chart 11BChina Is Not A Large Enough Share Of U.S. Exports (Detailed) Chart 12AU.S. Is Not A Large Enough Share Of Chinese Imports (Broad) Chart 12BU.S. Is Not A Large Enough Share Of Chinese Imports (Detailed) Chart 13AU.S. Has Less Market Access In China Than Other Exporters Chart 13BU.S. Has Less Market Access In China Than Other Exporters China has granted the legitimacy of U.S. complaints by pledging several times in the last few months to open market access. The latest news from the negotiations suggests that some progress is being made.9 Clearly the above is a very rough measure. Chinese consumers may not want to buy as much stuff from the U.S. as from Europe and Japan. The U.S. doubtless needs to improve its global competitiveness, and even then it may not gain as much market share in China as its DM peers. Nevertheless, Washington sees itself as the power that brought China into the global economy and allowed it to join the WTO. If China wants the U.S. to allow it to play a greater role in running the world, the U.S. is demanding a beneficial economic relationship in return. One way China is offering to deal with the problem is by buying American goods at the expense of U.S. allies' goods. For instance, Beijing has offered to buy more semiconductors from the U.S. and fewer from Taiwan and South Korea. This would alleviate the U.S. trade deficit a little, but at a greater expense to U.S. allies (Table 2). It would open up an opportunity for China to make more strategic acquisitions in those weakened, neighboring industries. It is not clear that the Trump administration will accept such a "concession," unless it is coupled with much greater concessions as compensation for selling out the allies. Table 2China's Trade Concessions To The U.S. Could Impose Costs On U.S. Allies Similarly, China's concessions that have been offered so far - like lowering the 25% tariff on car imports - are tokens in the right direction but not sufficient to satisfy the U.S. at the current juncture. This means that the U.S. will demand structural changes that increase market access, from a stronger RMB to a more consumer-oriented economy, as part of what will be a drawn-out effort to encourage China to rebalance its macroeconomy. Of course, Treasury Secretary Mnuchin was only telling half the truth: the U.S. also wants to prevent China from stealing too much of America's market share too fast. When we look at China's comparative advantage - the goods categories in which China's export growth has been fastest in recent years, weighted by contribution to the total - the U.S. is the country that has the largest global market share in these very goods (Chart 14). For instance, telecoms equipment, car parts, TVs, electrical circuits, etc. The U.S.'s export mix is not as dependent on these goods as that of China's neighbors (Taiwan, Vietnam, Malaysia, Singapore, South Korea), but it is the chief exporter of these goods nevertheless. Because many of China's most competitive goods are still low value-added (toys, plastics, textiles, furniture), China is pursuing tech upgrades, innovation, and intellectual property: it would eat away at the U.S. share of more advanced goods. Chart 14China's Comparative Advantage Threatens U.S. Global Market Share The Trump administration is trying to slow China's advance and put a stop to China's aggressive poaching of foreign tech and IP.10 This will include restrictions on Chinese direct investment and acquisitions to be announced by Mnuchin on May 21. We expect him to intensify an inherently stringent vetting process. The administration has already taken a proactive stance by blocking Canyon Bridge Capital Partners from acquiring Lattice Semiconductor and Singaporean company Broadcom's attempted acquisition of Qualcomm.11 Rumor has it that the administration is now considering invoking the International Emergency Economic Powers Act of 1977, which authorizes the president to take actions "to deal with any unusual and extraordinary threat, which has its source in whole or substantial part outside the United States, to the national security, foreign policy, or economy of the United States, if the President declares a national emergency with respect to such threat." Trump would be able to cite China's use of state-backed companies, corporate espionage, and cyber-attacks in pursuit of technology and IP (Table 3). Table 3Trump Lacks Legal Constraints On Trade Issues... Especially When National Security Is Involved This is entirely aside from legislation pending in Congress, which the White House appears to support, that would provide the Committee on Foreign Investment in the United States (CFIUS) with the ability to block investments across entire industries, rather than on a case-by-case basis, and with a broader definition of national security and sensitive property and technologies.12 While American presidents have historically vetoed similar legislation against China, the Trump administration may not, depending on the outcome of talks. The key point is that the U.S. political establishment - across the spectrum - is alarmed about China's economic mercantilism. As Senator Elizabeth Warren recently declared to a group of top policymakers in Beijing: "Now U.S. policymakers are starting to look more aggressively at pushing China to open up the markets without demanding a hostage price of access to U.S. technology."13 Warren, a staunchly liberal senator from the Democratic stronghold of Massachusetts, is entirely on the same page as Trump. The takeaway for investors? China's tit-for-tat response to Trump's steel and aluminum tariffs should not be dismissed out of hand. The market is sensitive to trade fears and there is a clear avenue for them to get worse if the 60-day consultation period lapses without any major Chinese concessions. True, negotiations are ongoing and Trump's trade team has been shown to be both credible and willing to pursue trade disputes through the WTO. Nevertheless there are substantial measures aimed at China coming down the pike and the usual restraints on U.S. policy, centered on the U.S. business establishment lobbying policymakers, are not as effective as in the past. Bottom Line: The U.S.'s primary economic goal in the China negotiations is not to equalize tariffs but to open market access. The strategic goal is much larger. The U.S. wants to see China's rate of technological development slow down. As such, Washington will expect robust guarantees to protect intellectual property and proprietary technology. Investment Conclusions Several clients have asked about the constraints on the different players if trade conflict should escalate over the coming months. On the surface the U.S. is in a stronger position because its outsized deficit with China means that measures constricting bilateral trade are inherently more damaging to China's output (Chart 15). Even some of China's best retaliatory options are difficult to put into practice, including selling U.S. treasuries or imposing sanctions on U.S. commodities (Table 4).14 Chart 15China More Exposed To Trade Than U.S. Table 4China's Retaliation Options Are Limited... Even In Agriculture The U.S. also faces a constraint in imposing measures on China because manufacturing value chains today sprawl across various countries and multinational corporations. Tariffs therefore punish countries, including U.S. allies, that provide inputs to China or American companies that profit from them - think Apple. Moreover, tariffs will not in themselves change the U.S.'s fundamental savings-investment balance, suggesting that demand for foreign goods will simply shift to other producers and the trade deficit will be unaffected. However, supply chain risk is ultimately not prohibitive for the U.S. China has long ranked among the most exposed to supply-chain disruptions, while the U.S. ranks among the least (Chart 16). Moreover, U.S. allies in Europe and ASEAN stand to benefit if supply chains are rerouted from China (Chart 17). While the U.S. and allies would suffer higher initial costs as a result, they would gain the strategic advantage of reducing China's centrality to global supply chains. The latter has given Beijing an advantage in acquiring technology and moving up the value chain. Chart 16China Most Exposed To Supply-Chain Risk Chart 17U.S. Allies Benefit If Supply Chains Move While the Xi Jinping administration is weaning China off export reliance and U.S. reliance, the country still employs 28% of its workers in the manufacturing sector, which leaves it more exposed to disruptions than the U.S. if trade frictions should spiral out of control and weaken overall demand (Chart 18). While American workers are intimately familiar with the boom-and-bust cycle of free labor markets, China has not struggled with significant unemployment since 2003 (Chart 19). Its middle class was much smaller then. Chart 18Employment Is A Constraint On China Chart 19China Unfamiliar With Large-Scale Job Loss In short, China will first attempt to appease the Trump administration through market access (and keeping the RMB strong) to maintain its supply-chain centrality and overall stability. If Trump accepts China's concessions, trade frictions will not spiral out of control - at least not this year. China will only accept a full-fledged trade war if Trump rejects its concessions and imposes punitive measures that threaten its stability. At that juncture, Xi would probably find it useful to demonize Trump and execute long-term changes to make China more self-sufficient, blaming the U.S.-initiated trade war for the painful consequences. This is why it matters if Trump's demands go beyond foreign exchange rates, improved market access, and IP enforcement - for instance, if they extend to capital account liberalization, the holy grail of American trade negotiations with China. Thus far, Trump's team has not raised this demand, but it is a subject we will revisit soon as it is likely to be China's red line, at least within the economic sphere. In light of our expectation for further trade-war related volatility, we would recommend shorting Chinese tech stocks15 and remaining short China-exposed U.S. stocks. The latter trade has been in the black by over 5% in just a week, but is currently up only 0.7%. It is a way to hedge the risk of further tensions between U.S. and China. Risks to this view are: if the U.S. reduces the Section 301 tariffs that it is threatening on or after April 6; if Treasury Secretary Mnuchin's investment restrictions due on May 21 are watered down; or if the U.S. makes no structural demands on China's economy but merely accepts temporary RMB appreciation and some big-ticket import orders. Otherwise the risk that trade tensions spiral out of control will remain elevated at least through the U.S. midterm elections on November 6. By then, Trump will need either to have cut a small-scale deal with China that he can tout for voters or to have taken more aggressive trade action pursuant to the Section 301 findings. Matt Gertken, Associate Vice President Geopolitical Strategy mattg@bcaresearch.com Marko Papic, Senior Vice President Chief Geopolitical Strategist marko@bcaresearch.com Jesse Anak Kuri, Research Analyst jesse.kuri@bcaresearch.com 1 Please see BCA Geopolitical Strategy Special Report, "Constraints And Preferences Of The Trump Presidency," dated November 30, 2016, available at gps.bcaresearch.com. 2 Please see BCA Geopolitical Strategy Special Report, "Market Reprices Odds Of A Global Trade War," dated March 6, 2018, available at gps.bcaresearch.com. 3 Please see BCA Geopolitical Strategy Weekly Report, "We Are All Geopolitical Strategists Now," dated March 28, 2018, available at gps.bcaresearch.com. 4 Please see BCA Geopolitical Strategy Special Report, "NAFTA - Populism Vs. Pluto-Populism," dated November 10, 2017, available at gps.bcaresearch.com. 5 A 60-day consultation period with both legislatures will follow but the deal will probably remain in more or less the same form. 6 Aluminum was not included, but South Korea is not a major source of aluminum products for the U.S. 7 Please see footnote 2 above. 8 Please see David Lawder, "Trump to unveil China tariff list this week, targeting tech goods," Reuters, April 2, 2018, available at www.reuters.com. 9 Treasury Secretary Steve Mnuchin spoke with Politburo member Liu He, who is Xi Jinping's top economic policymaker, and they reportedly pledged that they are "committed" to a solution on reducing the U.S. trade deficit. The U.S. is asking for a $100 billion reduction to the trade deficit within the year, as well as some progress on intellectual property enforcement. Supposedly the specific demands involve reducing the Chinese tariff on car imports and raising the foreign ownership cap on Chinese financial companies, the latter of which China has previously promised to do. Please see Andrew Mayeda, "U.S. Pushes China On Cars And Finance In Tariff Talks," Bloomberg, March 26, 2018, available at www.bloomberg.com. 10 Please see the U.S. Trade Representative, "Findings of the Investigation into China's Acts, Policies, and Practices Related to Technology Transfer, Intellectual Property, and Innovation under Section 301 of the Trade Act of 1974," March 2018, available at ustr.gov. 11 In September 2017, the White House and Department of Treasury intervened in the attempt by a group of investors, including the state-owned China Venture Capital Fund, from acquiring Lattice, on the advice of CFIUS. Lattice makes computer chips that are highly versatile and can be used in military functions; the Chinese SOE was suspected of pursuing China's state-backed efforts to improve its semiconductor industry. Separately, in March 2018, President Trump blocked Singapore-based Broadcom's attempt to acquire Qualcomm, which would have been a hugely consequential tech merger due to the two companies' dominance in making processors. The Treasury Department feared that Chinese state entities might get access to Qualcomm's IP or that the merger might otherwise hinder Qualcomm's "technological leadership." Please see "CFIUS Case 18-036: Broadcom Limited (Singapore)/Qualcomm Incorporated," dated March 5, 2018, available at www.sec.gov. 12 Please see Andrew Mayeda, Saleha Mohsin, and David McLaughlin, "U.S. Weighs Use of Emergency Law to Curb Chinese Takeovers," March 27, 2018, available at www.bloomberg.com. 13 She was speaking with Liu He, seasoned diplomat Yang Jiechi, and Defense Minister Wei Fenghe. Please see Michael Martina, "Senator Warren, in Beijing, says U.S. is waking up to Chinese abuses," April 1, 2018, available at www.reuters.com. 14 Please see BCA Commodity & Energy Strategy Weekly Report, "Ags Could Get Caught In U.S. Tariff Imbroglio," dated March 15, 2018, and "Oil Price Forecast Steady, But Risks Expand," dated March 22, 2018, available at ces.bcaresearch.com. 15 Please see BCA China Investment Strategy Weekly Report, "After The Selloff: A View From China," dated February 15, 2018, available at cis.bcaresearch.com. Geopolitical Calendar

Special Report

Highlights Hong Kong's leverage burden is a corporate sector rather than a household sector problem. But this corporate sector debt is highly concentrated in the finance and real estate industries, meaning that investors should be legitimately concerned over Hong Kong's extremely elevated debt service ratio. Our BCA Hong Kong Debt Risk Monitor serves as an important tool to help investors gauge the risk of a serious credit-driven downturn in the region. While the risk from excessive leverage is real, the current message from our DRM is that the odds of a deleveraging event over the coming year are low. Due to the importation of U.S. monetary policy, Hong Kong may "enjoy" easy monetary policy on a permanent basis. This suggests that Hong Kong's private sector may continue to leverage itself even in the face of rising interest rates, setting up the potential for a cataclysmic future recession. Stay neutral Hong Kong stocks versus the global benchmark over the coming 6-12 months. While equities may rise in relative terms if earnings momentum converges with that of the global benchmark, it is not a sufficiently compelling prospect to outweigh the significant structural risk facing the region. Feature Hong Kong has appeared in the headlines of the financial press for two reasons over the past few months. The first is due to the recent weakness in the Hong Kong dollar (HKD), a topic that we addressed last week.1 The second was prompted by the BIS' March 2018 Quarterly Review, which noted that mainland China, Hong Kong, and Canada stood out among 26 jurisdictions as being the most vulnerable to a banking crisis according to their research. The BIS's warning is rooted in the fact that Hong Kong is a highly leveraged economy, but there are two additional reasons for investors to be cautious about the region: China's industrial sector is slowing, and monetary policy is tightening due to the region's direct link to U.S. interest rates. While Hong Kong has avoided the full brunt of rising U.S. rates over the past year thanks to plentiful interbank liquidity (which has limited the rise in 3-month HIBOR), we noted in last week's report that the weakness in the HKD likely means that gap between interbank rates and the base rate cannot get much wider. This means that further Fed rate hikes over the coming year are likely to feed more fully into tighter Hong Kong monetary conditions. In this report we review the extent and disposition of Hong Kong's indebtedness, and develop an indicator for investors to monitor in order to gauge the risk of a serious private sector deleveraging event. We conclude that while it is too early to position aggressively against Hong Kong stocks, the risk from excessive leverage is real and is very likely to eventually cause a serious credit-driven downturn. For now, however, that appears to be a story for another day, and as we explain below, potentially a distant one. Breaking Down Hong Kong's Debt Chart 1 presents the basis for concern about Hong Kong's debt. The chart shows the BIS' nonfinancial private sector debt service ratio ("DSR", which includes both households and nonfinancial corporations) for the G10 countries alongside that of China, Hong Kong, and Canada. The chart shows that Hong Kong's DSR has risen nearly to 26%, a full 10 percentage points higher than the G10 average, and is now the highest among the 32 economies that the BIS has debt service data for. One important point to note is that among the three countries that the BIS recently singled out for concern, the disposition of Hong Kong's private sector debt is more similar to that of China than Canada. Chart 2 highlights that the private sector debt in China and Hong Kong is predominantly owed by the nonfinancial corporate sector, whereas in Canada the debt is more equally split among the two sectors, with households owing more in total. Chart 1Hong Kong's Debt Burden Hits##br## A New High Chart 2Unlike In Canada, Hong Kong's Leverage##br## Is A Corporate Sector Problem Normally we would be inclined to suggest that the skew in Hong Kong's debt towards the corporate sector makes it less risky than in other jurisdictions where elevated leverage is a household sector problem. The rationale is that while corporations can (and often do) misallocate their capital, firm borrowing is usually employed to acquire income-producing assets, with problems arising only when the value of those assets (or their potential to generate income) declines sharply. Household leverage problems, on the other hand, are almost always the result of a sharp rise in residential mortgage credit, and our view is that the purchase of residential property is fundamentally an act of consumption rather than a true investment. In addition, the past experiences of several countries have shown that housing-related leverage busts are particularly pernicious, in that the resulting recessions tend to be followed by long periods of subpar economic growth. But unlike in China where the majority of nonfinancial corporate sector debt is held on the balance sheets of state-owned enterprises, Hong Kong's corporate debt does not have de-facto state backing and appears to be enormously concentrated in the real estate and financial sector. Over 80% of Hong Kong's total nonfinancial sector debt (which includes households) is provided by domestic banks, and Chart 3 shows that among bank loans to firms, 35% have been granted to property building & construction companies and another 22% to "financial concerns" and stockbrokers. This high concentration of corporate sector debt in the real estate sector means that investors should be legitimately concerned over Hong Kong's extremely high DSR. On the household side, we have made the case in a previous report that a replay of another spectacular housing bust (similar to what occurred in 1997) is highly unlikely despite the fact that Hong Kong house prices have vastly outstripped income over the past decade2 (Chart 4). Chart 3Loans To Businesses Are Highly Concentrated ##br##And Exposed To Property Chart 4Lofty House Prices Are A Red Herring: ##br##The Risk Is On The Business Side This suggests that, despite extremely elevated residential property prices, investors should be more concerned about a shock that will destabilize the commercial real estate market. Hong Kong households would not likely escape the impact of such a shock, since commercial and residential real estate prices move strongly in tandem (Chart 5). But in terms of watching for a "tipping point" that could push Hong Kong's private sector into a balance sheet recession, the trigger seems more likely to occur in the market for the former, rather than the latter. Bottom Line: Hong Kong's leverage burden is a corporate sector rather than a household sector problem. But this corporate sector debt is extremely concentrated in the finance and real estate industries, meaning that investors should be legitimately concerned over Hong Kong's extremely high debt service ratio. Chart 5Still, Households Will Be Hurt##br## If CRE Prices Fall Chart 6The BIS' Warning Thresholds ##br##Don't Seem To Apply To Hong Kong Timing The Onset Of A Balance Sheet Recession Our analysis above supports the recent warnings from the BIS that the risk of a banking crisis / private sector deleveraging event in Hong Kong is nontrivial. This raises the obvious question of how to gauge the timing of such an event in order for investors to properly position their exposure towards Hong Kong's financial markets. The BIS has itself investigated this question, and has published several reports on its "Early Warning Indicator" (EWI) approach.3 Table 1 presents a list of these indicators for several countries, and highlights that the two of the most informative measures (the credit-to-GDP gap4 and the overall debt service ratio) are flashing red for Hong Kong. In fact, Table 1 served as the basis for the BIS' warning in their most recent Quarterly Review that we noted above. The BIS' EWI research has focused on identifying thresholds for these measures that can predict a banking crisis within a three-year window based on the historical record. But in the case of Hong Kong, it is not clear that these thresholds apply. Chart 6 shows the credit-to-GDP gap and overall private sector DSR along with the more stringent BIS threshold noted in Table 1, and highlights that these measures have been flashing red for 4-8 years. Based on this approach, Hong Kong should have experienced a banking crisis long ago. Table 1BIS Early Warning Indicators For Stress In Domestic Banking Systems Rather than relying on the BIS' framework, we have instead constructed our own private-sector debt risk monitor for Hong Kong. In contrast to the BIS' measures, which have been specifically constructed to predict a banking crisis, the goal of our indicator is to help predict a serious credit-driven downturn regardless of its character (i.e. we abstract from whether the result of the downturn is a full-blown financial crisis or simply a prolonged period of economic stagnation). Chart 7Low Risk Of A Serious Credit-Driven ##br##Downturn, For Now Chart 7 presents our BCA Hong Kong Debt Risk Monitor (DRM) and its five equally-weighted components, a summary of which is provided below. All series have been scaled such that an increase in the DRM represents higher risk. Alpha: We have highlighted the importance of examining the alpha as well as the beta of regional equity returns in a previous report,5 and we include a composite indicator of Hong Kong's rolling alpha versus the global benchmark as a measure of Hong Kong-specific stock performance that adjusts for Hong Kong's riskiness. While this component of our DRM was quite elevated in early-2016 (signaling weak Hong Kong stock performance), it is presently in line with its historical average, and thus is not flashing a warning sign. Property Prices: Given the high concentration of Hong Kong's corporate sector debt in the real estate sector, our DRM includes the deviation of office & retail property prices from their 9-month moving average. Similar to the first component of our indicator, Hong Kong property prices are roughly in line with their trend and are not signaling serious economic weakness. Credit Impulse: The third component of our DRM is a simple bank credit impulse, calculated as the flow of credit over the past year as a percent of GDP. While this component has fallen well into "low risk" territory, over the past year, there are some tentative signs of a reversal that investors should monitor. Monetary Policy Stance: The fourth component of our DRM is a structural variable that attempts to measure whether U.S. (and thus Hong Kong) interest rates are either consistent or out of alignment with economic conditions in Hong Kong. This component is an average of two measures of the stance of monetary policy: 1) the difference between U.S. 10-year government bond yields and Hong Kong nominal GDP growth, and 2) the difference between the base rate and a Taylor Rule estimate for the region (with the latter acting purely as an estimate of the cyclical equilibrium interest rate).6 The chart shows that despite the onset of tighter monetary policy in the U.S. over the past few years, our gauge of Hong Kong's policy stance suggests that conditions are still easy, and that material further increases would likely be required in order to see this component rise to +1 sigma territory. Debt Service Ratio: The final component of our DRM is the BIS' total private sector DSR shown in Chart 6, acting as a second structural variable that captures the underlying debt servicing risk that the BIS has warned about. We extent the BIS' series back to the early-1990s on a best efforts basis, by adjusting the product of Hong Kong's prime rate and the total private sector debt-to-GDP ratio to best align with the official DSR series over the course of its history. Our extended series suggests that Hong Kong's debt servicing burden is indeed the highest that it has been over the past three decades, underscoring that our DRM is likely to rise materially if the cyclical factors included in the indicator deteriorate. The overall message of our DRM is that a threat to Hong Kong's economy from excessive debt does not appear to be imminent, despite the underlying risks highlighted by the BIS. While the risk from excessive leverage is real and is very likely to eventually cause a serious credit-driven downturn, the odds of this occurring over the coming 6-12 months appear to be low. Bottom Line: Our BCA Hong Kong Debt Risk Monitor serves as an important tool to help investors gauge the risk of a serious credit-driven downturn in the region. While the risk from excessive leverage is real, the message from our DRM is that the odds of a deleveraging event over the coming year are low. The Spooky Implications Of The Natural Interest Rate Gap Interestingly, at least part of the benign reading of our DRM is due to the fourth component of the indicator, our gauge of Hong Kong's monetary policy stance, which suggests that there is ample room for further rate increases. In fact, in our view this observation carries much deeper significance than many may initially perceive, as it may explain why the BIS' early warning indicator thresholds have not worked in the case of Hong Kong, and why the region may avoid a debt crisis for a further significant period (but ultimately experience a much more painful collapse when it finally arrives). At root, the reason that U.S. 10-year Treasury yields remain exceedingly low relative to U.S. nominal GDP growth is because investors believe that real U.S. policy rates are likely to be much lower on average over the next 10-years than they have been historically (Chart 8). Abstracting from calendar-based cyclical considerations (such at the timing of the next U.S. recession), this fundamentally reflects the prevalent view among fixed-income investors that the U.S. natural rate of interest (or "r-star") has likely permanently declined. If true, this is of enormous importance for Hong Kong, as it suggests that the region will permanently "enjoy" easy monetary policy. This is because the substantial leveraging that has occurred in Hong Kong in response to low interest rates implies that there has been no impairment (yet) to Hong Kong's natural rate of interest (Chart 9). Chart 8A Low Estimate Of R-Star Has Depressed##br## U.S. Bond Yields Chart 9No Evidence Of A Low R-Star##br## In Hong Kong In some ways the dynamic we are describing is not new: the importation of easy monetary policy from the U.S. via competitive currency devaluation over the past decade has been a well-known phenomenon that was quite prominent during the early phase of the global economic recovery. But the fixed exchange rate regime in Hong Kong means that this process cannot be avoided without abandoning the peg, an event that itself could trigger a deleveraging event via a sharp decline in asset prices. The key point for investors is that if the U.S. natural rate of interest has indeed fallen materially and permanently below potential GDP growth, then Hong Kong will not experience tight monetary conditions even once the Fed has normalized short-term interest rates, unless it raises them well above equilibrium levels. This suggests that Hong Kong's private sector may perpetually leverage itself until debt service burdens reach some, as yet, unknown maximum level, precipitating what would likely become a cataclysmic recession. The fact that no crisis erupted in late-2015/early-2016 when the cyclical components of our DRM deteriorated significantly suggests that this level may be materially higher than is presently the case. Bottom Line: Due to the importation of U.S. monetary policy, Hong Kong may "enjoy" easy monetary policy on a permanent basis. This suggests that Hong Kong's private sector may continue to leverage itself even in the face of rising interest rates, setting up the potential for a cataclysmic future recession. Investment Implications: Stay Neutral, For Now Chart 10Room For A Rise In Relative Earnings Momentum The picture painted by our above analysis suggests that a benign cyclical outlook for Hong Kong is arrayed against a negative (and potentially horrific) structural outlook. How should investors position towards Hong Kong equities in response? First, as noted above, our Debt Risk Monitor does not signal that there is an imminent threat facing the Hong Kong economy that would herald the potential for a major deleveraging event over the near-term. Second, while Hong Kong's earnings momentum is stretched in absolute terms, Chart 10 highlights there is room for a catchup versus global stocks, which could boost relative performance over the coming year. Third, relative valuation and technical conditions are at neutral levels (Chart 11), and thus do not provide any compelling basis to avoid Hong Kong stocks. But to us, the weight of this modestly positive assessment over the coming year is overshadowed by the structural outlook, meaning that we continue to recommend a neutral allocation towards Hong Kong stocks over the coming 6-12 months. The most investment-relevant conclusion from our analysis is that investors will one day be able to earn significant risk-adjusted returns from underweighting / shorting Hong Kong stocks once a serious credit-driven downturn begins. As an example, Chart 12 shows the impact of the Asian financial crisis on Hong Kong's relative performance, a period where our DRM rose sharply and persistently into "high risk territory". It took 12½ years for Hong Kong to rise to a new high in relative total return terms, and it has yet to do so in price terms. Chart 11Neutral Relative Valuation And ##br##Technical Conditions Chart 12One Day, Shorting Hong Kong Stocks##br## Will Be Enormously Profitable So while the economic and financial market conditions are not yet in place to act on a bearish structural view, we will be closely watching our Debt Risk Monitor over the coming months and years for signs of a significant deterioration, as it will likely provide a major opportunity for investors to earn outsized returns. Stay tuned! Bottom Line: Stay neutral Hong Kong stocks versus the global benchmark over the coming 6-12 months. While equities may rise in relative terms if earnings momentum converges with that of the global benchmark, it is not a sufficiently compelling prospect to outweigh the significant structural risk facing the region. Jonathan LaBerge, CFA, Vice President Special Reports jonathanl@bcaresearch.com 1 Please see China Investment Strategy Weekly Report, "Chinese Stocks: Trade Frictions Make For A Tenuous Overweight", dated March 28, 2018, available at cis.bcaresearch.com. 2 Pease see China Investment Strategy Weekly Report "Hong Kong Housing Bubble: A Replay Of 1997?", dated June 29, 2017, available at cis.bcaresearch.com. 3 For example, please see "Evaluating early warning indicators of banking crises: Satisfying policy requirements" by Mathias Drehmann and Mikael Juselius, BIS Working Paper No. 421, August 2013. 4 The BIS defines the credit-to-GDP gap as the difference between the credit-to-GDP ratio and its long-run trend, derived using a one-sided (i.e. backward-looking) Hodrick-Prescott (HP) filter. 5 Pease see China Investment Strategy Special Report "China: No Longer A Low-Beta Market", dated January 11, 2018, available at cis.bcaresearch.com. 6 Our Taylor Rule estimate for Hong Kong is constructed in a fashion similar to what we showed for China in our January 18 Weekly Report, using a neutral policy rate estimate of 5%. Cyclical Investment Stance Equity Sector Recommendations